International Journal of Intelligent Engineering and Systems

1

Authors: Prashant Digambar Hakim, Vinod Moreshwar Vaze

Electronic Health Record (EHR) is widely adopted by various providers for health care systems such as hospitals, medical practitioners, doctors, etc. and the medical technology involves collecting health information related to diseases, medication, medical images, etc. Personal information such as age, name, gender, billing, and weight details of an individual is highly sensitive data that need to be protected against unauthorized access, hence, the protection of medical data is a big challenge for healthcare systems. The importance of this research work to analyze the sharing of medical data securely without any leakage of patient’s data when the unauthorized user accesses the data. In this current research, Chest X-ray MedPix images from the EHR are used to share medical data through encryption and uploading the data into the cloud. The proposed framework uses SHA256 cryptographic hashing technique as it can enhance the security of a patient’s data. Meanwhile, the attacker is not able to reverse back the key by the process of hashing. The SHA256 has a verifiable key ability for matching and verifying the new passwords against the authentication to improve the security for sensitive data. The proposed SHA256 Verifiable Key (SHA256-VK) executed at a faster rate, as it consumed a lesser ‘average block time’ of 5.48612s when compared to Searchable Encryption-48.125s, Smart-Contract Ethereum Distributed Ledger-14s, Decentralized Security Architecture-Software Defined Networking (SDN)-10s and Elliptic Curve Cryptosystems (ECC) -5.88s.

20

Authors: Monika Rajput, Sanjay Kumar Sharma, Pallavi Khatri

The energy of a wireless sensor network is crucial. Energy optimization is one of the key areas where research is being conducted. The upgraded technology in wireless sensor network (WSN) protocols has been seen tending towards clustering of nodes for energy optimization. Sensor nodes are battery- operated, so the size of the WSN matters mainly in terms of energy and the lifespan of the network. This work proposes a new approach of clustering process for energy efficiency in WSN thereby increasing the network lifetime. The initial clustering uses K-medoid, and setup phase using timer-based execution of adaptive neuro-fuzzy inference system (ANFIS). The ANFIS neuron structure is optimized using the firefly algorithm. The timer-based call of ANFIS for re clustering process is the novel approach and shows optimum efficiency in energy consumption. The optimum choice of cluster head (CH) selection leads to uniform energy consumption amongst all the sensor nodes and avoids energy run out problem when CH charge is given to particular node the performance evaluation shows improvement in energy efficiency and lifetime of the network. This work shows that the lifetime of the network is enhanced by 4% as compared to existing methods.

40

Authors: Madhu Mallikarjuna Nayak, Sumithra Devi Kengeri Anjanappa

Magnetic Resonance Image (MRI) uses an automated system to detect the presence of tumors in brain tumor images. Many existing methods performed the classification based on machine learning approaches that include supervised and unsupervised machine learning approaches. There are various researches undergone with more efforts and many types of research related to brain tumor were examined using the small dataset. The existing models performed segmentation techniques that extracted irrelevant features showed over-fitting problems. Similarly, the segmentation process uses clustering approaches that identify the brain tumor at an early stage, which was a challenging task. Therefore, a Convolution Neural Network with Global Average Pooling (CNN-GAP) model is designed to automatically segment the tumor regions and from the segmented regions, the relevant features are extracted to overcome the problem of overfitting. The proposed CNN-GAP classifies the MRI data into Benign I and II grade or Malignant III or IV grade. The proposed CNN-GAP method showed a 2 % of improvement in accuracy when compared to the existing CNN technique.

66

Authors: Muhammad Ainul Yaqin, Riyanarto Sarno, Siti Rochimah

Business process models can represent course schedules. Every business process model has complexity. The greater the number of courses and the number of study groups, the more complex the business process model. The rate of increase in the complexity of the business process model is referred to as growth. The growth in the business process model complexity helps estimate business processes resource requirements at a certain complexity level. Therefore the equation for the growth of the complexity of the business process model needs to be formulated. The equation for the growth in the business process model complexity is obtained through several stages, namely developing a course schedule using the aSc TimeTables software. The order of the scheduling results is formatted as an event log. The event log is used as material in the mining process using the inductive mining algorithm to get a business process model represented using the Business Process Modeling Notation (BPMN). The complexity of the business process model is calculated using the Yaqin complexity formula. The equation for the growth function for the complexity of the business process model as a function of the number of study groups is obtained through logistic regression. Logistic regression is based on a logistic equation with 3 parameters, 4 parameters, and 5 parameters. Based on the three types of logistic equations regression results, we find that the logistic equation with 5 parameters has the highest average value of R2, namely 0.9698. Meanwhile, the equations for the growth of the business process model complexity are a function of the number of courses through power regression. The equations obtained through power regression have an average value of R2 = 0.9549. In this study, regression is carried out separately between the growth function as the number of study groups with the growth function as a function of the number of courses. This research novelty lies in using the number of study groups and the number of courses to measure the growth in the business process model complexity.

95

Authors: Wayan Firdaus Mahmudy, Adyan Nur Alfiyatin, Candra Fajri Ananda, Agus Wahyu Widodo

Inflation is an important tool to assess the current condition of a nation economy. Uncontrolled inflation rate may have a lot of negative impacts on economic development. Forecasting can be used to control the inflation by making appropriate economic policies. However, uncertainty pattern of the inflation rate may make it hard to forecast. This study proposes an extreme learning machine (ELM) for the inflation rate forecasting. As part of the machine learning algorithm, the ELM has an ability to address uncertainty in data input pattern. However, ELM had weakness in determining initial weights that may produce inaccurate results. So, we propose particle swarm optimization (PSO) to determine good initial weights for the ELM. PSO is a metaheuristic method that gives good results in local searches but requires longer computation time to locate its particles on the global optimum point in the vast search space area. To overcome this problem, auto-speed acceleration algorithm is employed to drive particles of the PSO in searching of the global optimum with lower computation time. The performance of the proposed approach is evaluated using root mean square error (RMSE). A series of computational experiments prove that the proposed approach achieves better results with the average RMSE of 0.01926. This result is better than RMSE of 0.02020 achieved by the original version of ELM.

115

Authors: Mallikarjuna Anantapur, Venkanagouda Chanabasavanagouda Patil

Mobile Ad hoc Network (MANET) is an infrastructure-less wireless network that is characterized by multi-hop communication, dynamic network topology, open medium, and so on. MANET is generally vulnerable to security threats because of its decentralized control architecture. Therefore, the malicious attack is required to be avoided in the routing path for improving the network efficiency. In this paper, the Ant Colony Optimization (ACO) based Modified Ad-hoc On-Demand Distance Vector (MAODV) is proposed to obtain secure data transmission over the MANET. The ACO considers the four fitness functions such as trust, residual energy, distance, and node degree to detect an optimal path under the constraints of blackhole attacks. The main objective of the MAODV-ACO method is to improve the Packet Delivery Ratio (PDR) while minimizing the delay. The performance of the MAODV-ACO method is analyzed in terms of PDR, throughput, average End to End Delay (EED) and overhead. This proposed MAODV-ACO method is compared with Context-Aware Routing Protocol (CARP) and, Ad-hoc On-Demand Distance Vector (AODV) using Bayesian Approach and Dempster Shafer Theory (BA-DST) to evaluate the network performances. The PDR of the MAODV-ACO method is 99.66 % for 100 nodes, which is high when compared to the CARP and AODV-BA-DST.

137

Authors: Divya Jain, Sanjaeev Narayan Sharma, Alok Jain

Digital differentiation is an important phenomenon in digital signal processing to finding out sudden changes in signal. In this paper design of digital differentiator based on Fourier series expansion with quadratic fall in response has been presented. Error minimization criterion is applied to obtain minimum error in band of interest and to optimize the designed digital differentiator. Differentiator produces noise after differentiation and differentiated signal become noisy therefore guided filtering has been detailed to further smoothen the differentiated signal. Criterion for sampling frequency has been evaluated for particular cut-off frequency so that efficient selection of sampling frequency for Electrocardiogram (ECG) signal may possible. The results are demonstrated by applying designed differentiator on ECG signal and it has been found that the response of smoothened digital differentiator is equivalent to ideal differentiator. In this work sensitivity and error rate are 99.97 percent, and 0.069 percent, respectively.

159

Authors: Madhavi Devi Lanka, Subbarao Chalasani

A low profile and highly efficient M-Shaped monopole antenna has been designed and its analysis is presented in this work. The designed model can work in the sub 6 GHz band with coverage in the communication applications like Bluetooth(2.4 GHz), Long Term Evolution LTE (2.45 GH), Industrial Scientific and Medical Band ISM(2.5 GHz), Wireless Fidelity Wi-Fi (3.1 GHz) and Wireless Local Area Network WLAN (5.5 GHz) respectively. It has been built on commercially available FR4 material with dimensions of 37X26X1.6 mm. Electromagnetic simulation-based modelling with optimization has been done and presented in this work. An average gain of 3.8 dB, average directivity of 3.2 dB and attained average efficiency reading of 82% from the current design. The results obtained in real time measurement with respect to reflection coefficient, gain, radiation characteristics and Voltage Standing Wave Ratio VSWR are matching with the results obtained in the simulation.

177

Authors: Hana H. kareem, Rana T. Saihood

Enhancing aerial images plays an important role in many remote sensing applications; aerial images are often affected by dust, smoke, and fog. In this study, a new method was proposed to enhance aerial images. The proposed method is enhanced the light value (Y) in the YCbCr color transform by using the Sigmoid function mapping and Adaptive Histogram Equalization techniques, While the chromatic compounds (CbCr) are enhanced by Dark Channel Prior (DCP). To know the efficiency of enhancement, the no-reference quality measures as Entropy and Average Gradient (AG) have been calculated, which are used to see the quality of the proposed method in comparison with Image Entropy and Information Fidelity (IEIF), Multiscale Retinex with color restoration (MSRCR), Color Enhancement Scene Estimation (CESE), traditional (DCP), and Histogram Equalization (HE). From the results, it is found that the proposed method has succeeded in enhancing the aerial images better than other methods. Where it had high values for the rate for Entropy (7.73) and AG (68.75).

198

Authors: Dimas Aryo Anggoro, Salsa Sasmita Mukti

The kidneys have an essential role in the body; if it does not work properly, it will cause disease, one of which is chronic kidney failure (CRF). Therefore, a machine learning algorithm is needed to help predict CRF, one of which is the Extreme Gradient Boosting (XGBoost) algorithm. However, XGBoost has one thing that must be considered, namely the presence of hyperparameters that need to be tuned. In this research, tuning was done using grid search and random search methods, which would be compared to get a good performance and help predict GGK. In the dataset used, several features were incomplete, so that data cleaning was carried out. Based on the unbalanced amount of data, oversampling with SMOTE and random oversampling methods were needed to balance the data. Next, the data were normalized using two: min-max and z-score normalization. After the data had been normalized, a splitting process was performed with a division of 70% for training data and 30% for testing data. Furthermore, the hyperparameter was tuned to obtain the optimal parameter value, which was then performed data processing and evaluated to obtain the accuracy and f-measure values of the tuned parameter values. In the grid search method with data performed with min-max and z-score normalization combined with oversampling random oversampling, the best results were obtained with an accuracy value of 99.28% and an f-measure value of 0.9942. Comparison with other methods was carried out and the proposed method obtained an accuracy of 99.33% and f-measure 1.0.

220

Authors: Seyed Alireza Bashiri Mosavi, Mohsen Javaherian, Mohammad Sadeghi, Halime Miraghaei

The increases in releasing high-dimensional radio galaxy images necessitate addressing automatic methods for compacting the curse of dimensionality. Hence, in this paper, the modularized filter-wrapper feature selection scheme (MFWFS) is exploited to select the most discriminative features (MDFs) of the galaxy images. First, we employed the Fanaroff–Riley (FR) radio galaxy images in the data-gathering phase. Next, we applied the MFWFS scheme to 528 moments of radio images for selecting MDFs in dual-phases. The preliminary optimal features (POFs) are selected in the filter phase concerning the triple information theory criteria. In the wrapper phase, the obtained POFs are fed to the twin support vector machine (TWSVM) classifier to extract MDFs. Finally, the quad-final MDFs are introduced to the experimental comparison strategy through the cross-validation technique. Our results show that the best-laid MDFs in a subset of 96 features have accuracy close to 80% with a dimension reduction rate of 5.5.

249

Authors: Pooja Manghnani, Asmita Moghe

Eye is a precious part of human body. But in the world, glaucoma has affected the human eye as second main cause of blindness. This disease grows very slowly in the eye and without noticing it destroyed the optic verves within the eye. Traditional glaucoma detection methods are costly and time consuming. Therefore, better methods are required to detect glaucoma more accurately. Hence, a new adaptive approach for glaucoma detection using bi-dimensional empirical mode decomposition (BDEMD) from retinal images is proposed. Adaptive bi-dimensional intrinsic mode functions (BDIMFs) are obtained using very simple steps, from pre-processed and coloured decomposed images, is the main idea of the proposed method. Moment and texture features contain more information hence these are extracted from decomposed BDIMFs. These features are then normalized and classified by support vector machine (SVM) with its different kernel functions. The achieved results like, accuracy, sensitivity, and specificity are 97.02 %, 98.23 %, 95.46 %, respectively for 10-fold cross validation. Experimental analysis shows that our method outperformed over the traditional methods for glaucoma detection.

274

Authors: Achraf Boumhidi, El Habib Nfaoui

Advanced reputation generation systems analyze user-generated content such as opinions and reviews expressed in natural language in order to produce a reliable and trusted reputation value. According to recent and relevant literature, while many important attributes are exploited such as semantic and sentiment orientation of the reviews, time, and opinions’ relevancy, other relevant attributes such as users’ credibility, as well as the sentiment intensity of the reviews, are not considered. In this paper, we propose a system that computes a single numerical reputation value between 0 and 10 from Twitter microblogging platform by incorporating the sentiment orientation of the tweets, the sentiment intensity of the positive tweets, as well as the users' and tweets' credibility score. To assess the effectiveness of our system we have compared its computed reputation value with the ground truth one ranging from 0 to 10. This later is the weighted average votes of thousands of users taken from IMDb, Amazon, TripAdvisor, and Yelp concerning respectively four products and services (movie: 7.9, phone product: 7.3, hotel: 9, restaurant: 7.1). The experimental results conducted on four real-world Twitter datasets related to the aforementioned products show that our system provides a reputation value (movie: 7.66, phone product: 7.18, hotel: 8.71, restaurant: 7.08) that is near to the ground truth one. Consequently, it can be applied in practice and used by consumers and businesses to generate a reliable and trusted reputation value from tweets in order to support them during their decision-making process in E-commerce platforms (buying, renting, booking, etc.).

301

Authors: Asha Kathigi, Krishnappa Honnamachanahalli Kariputtaiah

In recent times, identifying Kannada, Arabic and English handwritten characters is a challenging task in pattern recognition application. The low resolution, complex backgrounds, text orientation, text size, and the variations in the writing styles makes character recognition as a challenging task. To address the above stated issues, a new automated character recognition model is introduced in this paper. Firstly, skewed line segmentation technique is applied to the handwritten character for dissecting the document images into line elements and then split into phrases and single characters. Next, AlexNet model is used for extracting the deep features from the individual characters, where the extracted feature vectors are multi-dimensional in nature that increases the system complexity. So, an unsupervised feature selection algorithm is proposed to select the active feature vectors that are fed to multi support vector machine classifier for individual character classification such as 64 classes in English language, 10 classes in Arabic language, and 657 classes in Kannada language. Experimental analysis showed that the proposed model obtained 85.80%, and 95.55% of accuracy on the chars74K dataset for Kannada, and English characters, respectively. In addition, the proposed model obtained 71.79%, and 99.97% of recognition accuracy in Kannada and Arabic handwritten character recognition on a real time dataset and MADbase digits dataset. The obtained results are better compared to the existing deep learning models; DIGI-Net, feed forward neural network, and context aware model.

321

Authors: Shahbaa I. Khaleel

Since digital images have become an essential part in information systems and many other scientific applications, and due to the importance of some of these images in terms of the importance of the information they contain, there have been some problems in terms of the inability to determine the object in the image accurately or directly from the raw image. Because of the occurrence of errors affecting the process of transferring or taking these pictures, or there is a percentage of lack of clarity that leads to the loss of a part of the information, and this has led to the development and use of algorithms and methods that facilitate the process of detecting the edges of objects in the image and working to determine the spatial characteristics that are difficult to distinguish in sometimes. In this research, edge detection techniques and methods that depend on computational intelligence, and which are based in their work on biological content inspired by nature are presented. Two swarm intelligence methods represented by the beetle optimization method BSO and the butterfly optimization method BFM for conducting the edge detection process were presented in the images. In addition to developing a powerful edge detector based on the beetle and butterfly algorithms by improving the performance of the butterfly algorithm by choosing optimal values for its parameters that play an important role in finding the optimal solution. This was done by hybridizing it with the beetle algorithm that provides the butterfly algorithm with the best values for its parameters. The hybridization process was done in the first stage of the butterfly method, which is the initialization stage, this method called Beetle Butterfly Method BSO-BFM. All edge images that obtained using different swarm intelligence techniques of proposed system were compared by the reference edge image (ground truth) to ensure the performance, and accuracy of all methods used in the research, using mean square error MSE and root mean square error RMSE measurements as well as peak signal to noise ratio PSNR. From the results it turns out that the proposed hybrid method BSO-BFM is the best of all, this was evident through the values of the measures used here, as this method had the best values for the MSE measure, where its values ranged between 0.023375 - 0.466766, and the RMSE measure, where the best values were obtained in this method, which is between 0.152889 - 0.216048, and also for the PSNR measure, which ranges between 61.4398 - 64.443273.

344

Authors: Zahraa A. Ghazi, Kasim K. Abdalla

Multiple-input multiple-output (MIMO) systems are used in most modern wireless communication systems, including WiMAX, DVB-NGH, WiFi, HSPA+, LTE, and 4G. MIMO systems with space time coding are a potential technique for increasing data rates and improving wireless communications reliability.This work proposed a new technology for decoding Multiple-Input Multiple-Output Space–Time Block Code (MIMO-STBC) by using blind source separation (BSS), to enhance the BER performance for MIMO STBC channel estimator by using semi-blind independent component analysis (ICA) method, The kurtosis-based source extraction technique is achieved by using real imaginary decomposition (R-Im) of maximum ratio combiner (MRC), Finally, by applying the water cycle method in optimization, we were able to increase the work speed of channel estimation using the ICA technique. The MATLAB software was used to model the system's performance. The simulation results demonstrate that the proposed channel estimation technique is feasible and effective, with only a small difference in performance between the estimated and known channel parameters, such that the proposed algorithm using WCA-kurtosis is better than the performance of the LS algorithm for transmit and receive antennas 2×1, 2×2, 4×2, and 4×4 by 1.6, 1.5, 1.2, and 2.2 dB, respectively, at the frame length of the 1024.

369

Authors: Sakon Klongboonjit, Tossapol Kiatcharoenpol

Wind energy potential is the main data to making decision for green energy projects in resident areas. The decision makers generally prioritize wind energy of alternative areas with data of wind power that are hard collected. In this study, the AHP method is introduced to prioritize wind energy of alternative areas with data of average wind velocity generally receiving from the secondary data sources. With the example of 13 alternative areas in the south of Thailand, the data of average wind velocity can be used to construct the AHP pairwise comparison matrix. Finally, the priority values of these 13 alternative areas are calculated with C.R. = 0.1157 and the wind energy priority of alternative areas is prioritized. The priority results with the data of average wind velocity are really similar to the priority results with the data of wind power.

391

Authors: Ujjwala S. Rawandale, Mahesh T Kolte

The hearing aid system is used for hearing impair patient communication, but the hearing aid's main issues are improper communication, power dissipation, delay, matching error, and low complexity. Many existing filter bank techniques are developed to overcome these issues but still have the problem of delay, unwanted noise, etc. This paper has designed a Tuned Feed forward Variable Bandwidth Filter (TFVBF) to enhance the Hearing Aid System (HAS) by reducing unwanted noise, matching error, and complexity. Moreover, the developed technique is implemented in MATLAB and it achieve low rate in power consumption and delay. Thus, the gained performance metrics has enhanced the HAS's performance by reducing, unwanted noise, matching error, and complexity. Finally, the attained performance metrics of the developed technique using 50 bands are compared with existing replicas in terms of matching error, delay, power consumption, and complexity for checking the effectiveness of developed replica.

417

Authors: Pharida Jeebklum, Phumin Kirawanich, Chaiyut Sumpavakup

The dynamic wireless power transfer system consists of multiple power transmission sectors in parallel with a single inverter circuit for light duty electric vehicles. The use of multiple power transmission sectors reduces power loss. Usually, each power transmission sector needs to use an inverter for every sector but this system uses one inverter to save cost. This article proposed a study of the dynamic wireless power transfer efficiency between the power transmission sector and the power receiver sector. The performance of the proposed dynamic wireless power transfer system was experimented and compared by varying the gap between the power transmission sector and the power receiver sector, and the speed of the electric vehicle at a resonance frequency. The results show that the dynamic wireless power transfer system had a maximum power transfer efficiency of 82.01% at a distance between the power transmission sector and the power receiver sector of 0.10 m, with a speed of 0.28 m/s.

442

Authors: Yasameen Sajid Razooqi, Muntasir Al-Asfoor

Routing is one of the most critical issues in Wireless Sensor Networks (WSN) as it differs from routing in classical networks. The dynamic nature of WSN environment and the constrained limited resources availability have made routing process a real challenge to industry and academia. One of the best solutions in this routing case is a heuristic search to find the best and optimal path. Thus, many heuristic optimization algorithms are used to enhance routing performance in WSNs. Ant Colony Optimization (ACO) is a well-known heuristic algorithm which used in WSNs, as it showed an admirable performance improvement routing context. In this paper, two routing protocols are proposed: modified Ant Colony Optimization for flat routing protocol (MACOF), and the second is Naïve Baye’s with ranking for flat routing protocol (NBRF). We have used inverse of the distance between the node and it’s neighbours in the probability equations of ACO and Naïve Baye’s formulas, along with pheromone and residual energy. Those formulas have been used in choosing the candidate between neighbors for the next hop that will lead to consume less average energy of all nodes in each round. In addition, in the first protocol we have adjusted ACO’s parameters values after several experiments, and in the second protocol we have proposed ranking for network training before start working. The experimental analysis has been done using varied scenarios of networks topologies. Furthermore, the experimental results have been compared and contrasted against the well-known ACO approaches and IOP routing protocols. The performance of the proposed methods of MACOF and NBRF have been shown in terms of throughput, energy consumption, network lifetime, the size of data transferred over the network, and the length of paths travelled by the packets.

464

Authors: Purba Daru Kusuma, Abduh Sayid Albana

Intermediate storage is one important aspect of parallel permutation flow-shop scheduling (PPFSP). Unfortunately, research about optimizing intermediate storage is rare. Most existing studies were conducted to minimize or optimize the maximum completion time under various circumstances and constraints. Thus, this work aims to minimize intermediate storage with a tolerable maximum completion time level. In this work, the proposed model was developed by using an evolutionary algorithm (EA). This model consists of two steps. In the first step, the EA is used to determine the tolerable maximum completion time. In the second step, the EA is used to make a solution with minimum intermediate storage and a tolerable maximum completion time. This proposed model is then compared with the existing non-dominated sorting genetic algorithm (NSGA II), improved efficient genetic algorithm (IEGA), cloud-theory based simulated annealing (CTSA), and artificial bee colony algorithm (ABC). Based on the simulation results, the proposed model performs better than the IEGA, CTSA, and ABC models. Meanwhile, the proposed model is competitive enough compared with the NSGA II. And, the two-step evolutionary algorithm (TSEA) produces 26 percent higher maximum intermediate storage, and the improved two-step evolutionary algorithm (ITSEA) produces 7 percent higher maximum intermediate storage than the NSGA II. The ITSEA produces 16 lower maximum intermediate storage than the TSEA. Both TSEA and ITSEA produce a 21 percent higher maximum completion time than the NSGA II.

488

Authors: Nandeeswar Sampigehalli Basavaraju, Shanmugarathinam Ganesarathinam

In recent decades, the automatic recognition of diabetic retinopathy is an emerging research topic, where diabetic retinopathy is the major cause of vision loss in human beings. The early recognition of diabetic retinopathy lesions like the optic disc, cotton wools, hemorrhages, exudates, and Microaneurysms reduces the growth rate of diabetic retinopathy disease and assists clinicians in early treatment. However, the monitoring, and recognition of diabetic retinopathy lesions are tiring, cyclic and error-prone processes. To overcome these problems, a new automated diabetic retinopathy detection model is proposed in this paper. Firstly, the retinal fundus images are acquired from Kaggle retinal dataset and DiaRetDB1 dataset, and the acquired raw fundus image quality is improved using histogram equalization technique, Haar wavelet transform, discrete wavelet transform and Gaussian matched kernel filtering method. Further, the k-means clustering algorithm and Gaussian blur technique are applied for fovea and blood vessel removal, and to highlight the optic disc regions. Next, color histogram features are extracted from the segmented regions, and the obtained features are fed to an ensemble classifier for classifying the healthy, and DR subjects. The simulation results show that the proposed k-means based ensemble model achieved 93.29% and 99.37% of accuracy on the Kaggle retinal dataset and DiaRetDB1 dataset, which are better compared to the existing state of the art models.

507

Authors: Srikanth Parikibandla, Sreenivas Alluri

Nowadays, the Lightweight cryptography algorithm is mostly used for security level purposes, because it’s designed with the help of basic arithmetic logic blocks. The basic blocks are perfectly placed in respective modules to process the encryption with effective manner. The fundamental shifting based PRESENT algorithms are frequently used to help the hackers to retrieve the data. To overcome this issue, Separation With NOT (SWN) with Srikanth Parikibandla box (SP-box) architecture is used in this research work. Due to the usage of modified Register Transfer Level (RTL) structure, the process and the confidentiality of the PRESENT encryption architecture is improved in a significant manner. The Modified Single Coupled Linear Congruential Generator (MS-CLCG) architecture is used to generate the random number, which act as a key for encryption as well as decryption. For every clock cycle, MS-CLCG architecture generates the random value to perform the encryption. The SWN-SP architecture occupied 56 counts of LUTs, 33 count of Flip-flops and 30 counts of slices, which are less compared to the existing SRS-PRESENT architecture.

528

Authors: Thanh Long Duong, Ngoc Anh Nguyen, Thuan Thanh Nguyen

Optimal power flow (OPF) is an important problem in the power system operation. The purpose of the OPF problem is to optimize a defined objective function by modifying variables of control such as real power at generator buses except slack bus generator, voltages at all generator buses, reactive power of compensators and tap ratio of transformers while all constraints are satisfied. This paper is proposed an approach based on the artificial ecosystem optimization (AEO) to solve problem of optimal power flow. The suggested algorithm is tested on the IEEE-30 bus systems with five target functions consisting of fuel cost, emission, power loss, voltage deviations and L_index. The results obtained of the suggested AEO approach compared with equilibrium optimizer (EO), particle swarm optimization (PSO), sunflower optimization (SFO), genetic algorithm (GA) and other exiting methods. The results simulation shows that, standard deviation obtained value after 50 independent runs by the proposed AEO algorithm is better compared with EO, PSO, SFO and GA method. The fuel cost, emissions, active power loss and voltage deviations levels are reduced by 11.21%, 44.06%, 46.44%, and 92.13% respectively, compared to the initial case. Furthermore, for other exiting methods the improvement level percentage (IL) of the proposed AEO algorithm can be up to 0.2285 % for fuel cost objective, 0.137% for emission objective, 7.618% for total power loss objective, 89.85% for voltage deviation objective and 0.652% for L_index objective. Thus, the proposed AEO method is also one of effective and reliable algorithms for handling OPF problem.

554

Authors: Robert Marco, Sakinah Sharifah Syed Ahmad, Sabrina Ahmad

In the subject of software effort estimation, missing data is a significant issue that leads to information loss and bias in data analysis. The majority of data preprocessing procedures are simple reuse approaches built for numerical data, which presents a problem when missing data and irrelevant features are associated with categorical variables. The purpose of this paper is to evaluate and compare the performance of the proposed technique with the k-nearest neighbor imputation (kNNI) technique, random forest imputation, and multiple imputation by chained equations in terms of error and accuracy in the ISBSG dataset when missing data is present. This study relied on five machine learning approaches as its foundation with hyperparameter tuning using grid search. The results show that the three imputation methods have almost the same performance. However, the combination of kNNI, Genetic feature selection, and classification and regression tree (CART) yielded better results than other combinations of methods with MAE (0.015), RMSE (0.037) and R-squared (0.804) values.

579

Authors: Dwiki Kurnia, Istikmal, Ida Wahidah

Delay in the use of the 10 Gigabit Passive Optical Network (XG-PON) on the Centralized RAN (C-RAN) is known to be not very good since a small maximum capacity of Remote Radio Heads (RRH) used, only 8 units. Dynamic Bandwidth Allocation (DBA) is required on upstream PON to dynamically allocate bandwidth. Conventional DBA on XG-PON produces an upstream delay which is 0.6 ms at least. In fact, C-RAN has strict delay requirements, namely 0.3 ms, which is why a DBA with low delay is needed as a solution. In this study, a modified Round Robin (RR) DBA algorithm using NG-PON2 is proposed with consideration of delay performance in upstream by utilizing excess bandwidth and using forecasting methods. Modified RR DBA has the best upstream delay among other DBAs that can be applied to C-RAN using a maximum of 16 Optical Network Unit (ONU) with the result of 0.261 ms.

596

Authors: Sowmyashree Malligehalli Shivakumaraswamy, Chikkanayakanahalli sadashivaiah Mala

In the previous few years, innovations in the Wireless Sensor Network (WSNs) domain have acquired momentum because of WSN’s various applications in farming, mechanical assembling, military observation, ecological checking, clinical and medical care, debacle recuperation tasks and so forth. In WSN, the wireless sensor nodes are organized randomly to communicate with other nodes inside the network. Since the nodes are battery-powered, energy is one of the major constraints in WSN. Moreover, the nodes are affected by malicious attacks (ex. black hole attack) which leads to security issues in the network. In this paper, the Cost Centric Cuckoo Search Algorithm (CCCSA) is proposed to overcome the energy and security issues in WSN. Here, the clustering operation is performed using the K-means clustering algorithm followed by CCCSA is used to select the adaptive Cluster Head (CH) between the normal nodes in the cluster. Further, the CCCSA is integrated with Ad hoc On-Demand Distance Vector (AODV) routing protocol to generate the adaptive routing path. In CCCSA, the trust value is considered as primary cost value for accomplishing a secure communication in WSN. The Packet delivery rate of the CCCSA method is 95.91% and it is higher when compared to the existing Media access control Centralized Routing Protocol (MCRP), Secure Routing Protocol based on Multi-objective Ant-colony-optimization (SRPMA) method and Realisable Secure Aware Routing (RSAR) protocol.

616

Authors: Yulianto Triwahyuadi Polly, Sri Hartati, Suprapto, Bambang Sumiarto

Pigs have become an essential part of the cultural and economic life of the people in Nusa Tenggara Timur (NTT) Province. Diseases in pigs significantly affect the success of pig farming. Identification of disease in pigs is a classification problem. Metaheuristic algorithms are widely used in Neural Network (NN) optimization to solve classification problems. Flower Pollination Algorithm (FPA) is grouped into a metaheuristic algorithm that has been commonly used in optimization cases in the real world. To improve FPA performance, this study proposes replacing the FPA step vector parameter, namely Levy distribution, with Newton Polynomial Quadratic Interpolation (NPQI), known as Quadratic Interpolation Flower Pollination (QIFP). Quadratic Interpolation Flower Pollination Neural Network (QIFPNN), Flower Pollination Neural Network (FPNN), Bat Neural Network (BANN), and Particle Swarm Optimization Neural Network (PSONN) algorithms were used to train NN in real cases of disease identification in pigs, covering 11 diseases with 68 clinical symptoms. The results showed that the proposed algorithm, namely QIFPNN, outperformed FPNN, BANN, and PSONN in classification accuracy. QIFPNN is also able to improve classification accuracy and speed of convergence when compared to FPNN. QIFPNN and FPNN, respectively, provide 82.6159 % and 67.4766 % accuracy, and the training time is 6056.240 seconds and 6555.179 seconds. QIFPNN accuracy increased by 22.40%, and training time was 7.61 % faster. It concluded that QIFPNN could be used as a complementary model in disease identification in pigs.

642

Authors: Raghad M. Eid, Eman k. Elsayed, Fatma T. Ghanam

More than seventy million protein sequences exist now, with just around 1% of their functions known. Multi-domain protein prediction is one of the problems in the Bioinformatics field, which was conducted to find the function of proteins. As a result, the researchers attempt to develop algorithms that predict protein functions based on their sequences. We might think of the tremendous rise in sequencing and structural genomics as providing us with a lot of data to expose the complicated sequence, structure, and functional correlations that exist in proteins. However, because of the critical functions that these macromolecules perform in biological mechanisms, acquiring a thorough understanding of their activity has recently emerged as a major challenge. Furthermore, multi-domain protein function prediction approaches are more efficient than methods that are fully processed based on protein sequencing. In this research, we used a model for predicting multi-domain protein function using a Fuzzy Convolutional Neural Network. In this research, we used a novel hybrid CNN and Fuzzy for sequence labelling, specifically to predict the function of the protein. Our hybrid model managed to outperform recent previous studies in terms of accuracy (95.02%) and performance on the UniProtKB dataset, and they showed significant improvements, in the execution time spent in the prediction process. It's crucial to keep your attention on this strategy. Further, study will be needed.

666

Authors: Dinu Calin Secui, Gabriel Bendea, Monica Liana Secui, Cristina Hora, Codruta Bendea

Economic dispatch problem is found in the recent focus of specialists, solving this problem being based especially on metaheuristic algorithms. Social Group Optimization (SGO) is a new metaheuristic algorithm that models human behavior to solve a certain problem, and it has been successfully applied to optimize mathematical functions. This paper investigates the behavior of the SGO algorithm equipped with five chaotic maps (Logistic, Double, Iterative, Singer and Cat), applied to solve the economic dispatch problem with multiple fuel sources. The new algorithm, called Chaotic Social Group Optimization (CSGO), results from including the sequences generated by chaotic maps in the heuristics specific to the SGO algorithm. The performance of the CSGO chaotic algorithm is tested on four systems having various dimensions (10-units, 20-units, 30-units, and 40-units) and characteristics. The SGO and CSGO algorithms are compared with each other and with other algorithms. The obtained results indicate that CSGO algorithms equipped with any of the chaotic maps mentioned has a higher average performance than SGO algorithm. The application of CSGO, compared to SGO, leads to a reduction of the fuel cost between (0.6-6.7) $/h for the analyzed systems.

1

Authors: Nanang Triagung Edi Hermawan, Edi Winarko, Ahmad Ashari, Yus Rusdian Akhmad

Radiation monitoring data is very important to support the safety of nuclear installation. The data should be ensured generated and supplied by an authenticated sensor. The new sensor should apply mutual authentication to server before joint the system. This research proposes a high secure authentication protocol based on Eight Prime Number of Modified RSA Algorithm (EPNR) method as certificateless mechanism to protect authenticity and confidentiality of sensor’s identity. The main advantages of the protocol are designed for simpler direct point to point connection, using one-time random keys, and implementing five-barrier data protection mechanism. Its performances in context security, running time, and Quality of Services have been compared to other models. The proposed protocol is faster 29.9 times in random key generation, 7.27 times in encryption, and 11.1 times in decryption duration, compared to the others model based on RSA scheme. It has better performances in throughput and delay time. It also has enhanced resistance to statistical, replay, and MITM attack.

25

Authors: Divya Jegatheesan, Chandrasekar Arumugam

The intelligent transport system aims to control congestion and enhance the driving experience through a variety of technologies and communication systems. They provide us a lot of data which can be fed to machine learning technologies to provide further enhanced services to the common public as well as the transport officials. In this busy modern world, the usage of vehicles is highly increased and thus monitoring and detecting these vehicles onboard with better accuracy and less time cost is quite a challenging task. We propose an intelligent traffic management solution to detect vehicles (static and onboard) which finds its set-about in the field of machine vision technology called YOLO D-NET (You Only Look Once Dilated Net). YOLO D-NET puts forward a novel architecture which incorporates the implementation of traditional YOLO model alongside of dilated Convolution Neural Network (CNN). Our paper also focuses on making a constructive comparison between the four different models namely SSD (Single Shot Multibox Detector) Inception, Faster R-CNN (Region-based CNN), YOLO ResNet (Residual Neural Network) and the proposed model YOLO D-NET. We have used the predefined COCO (Common Objects in Context) dataset and the custom dataset to train the traditional models. Also YOLO D-NET makes use of the predefined YOLO weights followed by the dilated CNN layers. Our proposed model was found to be 97.5 percent accurate with enhanced precision and accuracy than the other models particularly with less training time than the traditional YOLO model.

46

Authors: Quyen Thi Nguyen, Minh-Phung Bui

This paper presents a new method based on Equilibrium Optimizer (EO) algorithm that is inspired from the mass balance of a control volume for traveling salesman problem (TSP). For enhancing the efficiency of EO, the 2-opt movement algorithm is used to update the solution generated by EO. The efficiency of the proposed EO for the TSP problem has been compared with Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) on different instances consisting of the 14-city, 30-city, 48-city and 52-city. The calculated results show that for the large scale instances such as 48-city and 52-city, EO has found the better tour than PSO. In comparison with GA, EO has ability finding the best tour with the smaller mean and standard deviation. The comparisons with previous methods in literature have also demonstrated that EO has ability to search the better tour than other methods. Thus, the proposed EO can be a potential method for the TSP problem.

67

Authors: Khin Hsu Wai, May The` Yu

Effective and automated lane marking detection system is one of the important part of many intelligent transportation systems. Many researches of lane detection techniques have been conducted on academic and automobile industry and applied in intelligent transportation systems. However, It is still a challenging problem because of the various environmental conditions such as shadows, lane occlusion and illumination invariance that can decrease the lane detection performance. To deal with these issues, a reliable lane marking detection system that use the block-wise Hough based features is proposed. We propose the Region of interest extraction that divides the image into sub-image blocks to detect the lane line and block-wise Hough Transform feature extraction approach to extract the lane line features. Artificial Neural Network (ANN) applys the extracted features to detect the lane. The efficiency and accuracy of our system is evaluated by using the Caltech lane dataset. The average correct detection rate is 94.67% and the computational time is 17.64 ms/frame by applying the proposed method.

88

Authors: Azeddine Loulijat, Najib Ababssi, Makhad Mohamed

The main problem of the Doubly-Fed Induction Generator (DFIG) is its sensitivity to network disturbances especially voltage dip due to its stator windings being directly linked to the network. A voltage dip in the network results from high current peaks in the stator windings, since the stator and rotor are electromagnetically coupled, so these peaks provoke inrush current hard at the delicate back-to-back converters and an overcharge of the DC-LINK capacitor. In this perspective, several researchers are realized a linear and nonlinear controller with a passive protection method (PPM). The objective of this paper is to develop modelling and control models based on a true DFIG model considering the effect of all the parameters of a stator winding. The Second- Order Sliding Mode controller (SOSMC) is the non-linear control strategy used to control the dynamics of the DFIG through the rotor-side and network-side power converters under normal network conditions. When a voltage dip is detected, the Kalman observer (KO) contributes to a non-linear controller and the latter is coupled to a passive protection method that contains crowbar and DC-chopper circuits to improve the dynamics of the DFIG system to cross the network fault. Also, a comparative study is carried out between the developed controller and other linear ,non-linear approaches exist in the literature. The results obtained, under a maximum voltage dip (100%) on the 60 kV network, confirm its robustness and superiority in terms of limiting the DFIG graders at the beginning and on termination of a fault. In terms of Ir and Vdc values during a voltage dip, our control strategy does not exceed 1.45p.u and 1.2p.u, on the contrary, the ABC with HGO which is developed recently reaches 2.2p.u and 0.4p.u.

112

Authors: Yosepu Cooli, Chitraivel Mahesh

The development of computing technology has led to enormous improvement in the medical data and machine learning processing techniques. The increasing data usability in various field leads to the term big data, which is required in various applications such as medical, finance and so on. The existing models such as Random Forest, Support Vector Machine (SVM) and Decision tree have low capacity in handling the large volume of the data. In this research, the Recursive – Parallel Random Forest is proposed to effectively classify the large volume of medical data. The recursive partition method has been used to segment the input data and process the data in segmented manner. The recursive partition method segment the data and subset features based on data centric to improve the learning and provide adaptive gain ratio value to build Random Forest. The recursive partition method provides the data with high correlation that supports the random forest to perform the Random Forest parallel. The partition the data in data centric and balance the classes that helps the proposed model to handle the imbalance data effectively. The Parallel Random Forest method has been used to improve the performance of the medical data classification and the performance of Recursive partition- Parallel Random Forest (R-PRF) method was estimated using three UCI medical dataset such as Chronic Kidney Disease (CKD), Heart Disease and Diabetes. The standard classifiers and state-of-art method has been used to compare with the proposed R-PRF method and the experimental result shows that the R-PRF has the accuracy of 92.01 % and the existing Improved SVM radial method has 89.9 % accuracy.

129

Authors: Chinthakunta Manjunath, Balamurugan Marimuthu, Bikramaditya Ghosh

Equity market forecasting is difficult due to the high explosive nature of stock data and its impact on investor's stock investment and finance. The stock market serves as an indicator for forecasting the growth of the economy. Because of the nonlinear nature, it becomes a difficult job to predict the equity market. But the use of different methods of deep learning has become a vital source of prediction. These approaches employ time-series stock data for deep learning algorithm training and help to predict their future behavior. In this research, deep learning methods are evaluated on the India NIFTY 50 index, a benchmark Indian equity market, by performing a technical data augmentation approach. This paper presents a Recurrent Neural Network (RNN), Long Short Term Memory (LSTM), and the three variants of Gated Recurrent Unit (GRU) to analyze the model results. The proposed three GRU variants technique is evaluated on two sets of technical indicator datasets of the NIFTY 50 index (namely TA1 and TA2) and compared to the RNN and LSTM models. The experimental outcomes show that the GRU variant1 (GRU1) with TA1 provided the lowest value of Mean Square Error (MSE=0.023) and Root Mean Square Error (RMSE= 0.152) compared with existing methods. In conclusion, the NIFTY 50 index experiments with technical indicator datasetTA1 were more efficient by GRU. Hence, TA1 can be used to construct a robust predictive model in forecasting the stock index movements.

154

Authors: Srinivasulu Sirisala, S. Rama Krishna

The successful routing in Mobile Ad hoc Networks (MANETs) completely depends on the cooperation established between the mobile nodes during data dissemination. The resource constraint feature of mobile nodes is the main reason behind its non-cooperativeness attributed towards other interacting nodes of the network. This degree of non-cooperation rendered by the mobile node disturbs the routing and degrades the network performance. Thus, a node cooperation enforcing approach that distinguishes genuine node from malicious node is essential for sustaining the network performance. In this paper, Gaussian Trust Factor-based Grey Decision-Making Technique (GTF-GDMT) is proposed for maintaining the network performance independent to the influence of non-cooperative malicious nodes. This proposed GTF-GDMT identified the malicious nodes based on the trust factor that integrates the parameters of link stability and node reputation determined through direct and indirect monitoring process. This computation of trust is achieved using the Gaussian probabilistic distribution model that utilizes multi-perspective parameters that forms the input to the decision-making process. It is proposed for attaining the trade-off between energy consumption, transmission performance and security. The simulation experiments of the proposed GTF-GDMT conducted using ns-2 portrayed its excellence over the baseline schemes in terms of throughput, packet delivery rate, packet delay, energy consumptions and packet drop independent to the number of mobile nodes and malicious nodes in the network.

177

Authors: Suhariyanto, Rachmad Abdullah, Riyanarto Sarno, Chastine Fatichah

Sentiment analysis can provide rough recommendations in the form of sentiment from a collection of reviews or can provide recommendations in more detail about sentiment in a particular aspect called aspect-based sentiment analysis (ABSA). Sentiment analysis based on many aspects has been carried out but its accuracy is still being developed. In previous research, most research was carried out on explicit and implicit aspects and opinions and was carried out in simple sentences. The purpose of this research is to analyze the sentiment of restaurant reviews using the rule grammar method to extract implicit aspects - explicit opinions in four sentence models, namely simple (Si-AIOE), compound (Co-AIOE), complex (Ce-AIOE), and compound-complex (CoCe-AIOE). The ABSA method is proposed using the development of a grammatical rule extraction method to extract explicit and implicit aspect words and opinion words as the basis for sentence extraction. Rules making is done to take explicit and implicit aspect words and opinion words in Si-AIOE, Co-AIOE, Ce-AIOE, and CoCe-AIOE so that the comparison of the evaluation values can be known. This research uses the Semeval 2015 dataset on Restaurant reviews from the Tripadvisor Website which has been annotated as sentence data for ABSA. The aspect categorization process is then used to categorize aspects into 4 aspect categories, namely Ambience, Food, Service, and Price using hybrid approach. The hybrid approach is combined using Elmo-Wikipedia, grammatical rule extraction, WordNet, TF-ICF, and semantic similarity methods. The results of the aspect extraction obtained value of precision, recall, and f1-measure of 0.80, 0.84, and 0.82, respectively. Meanwhile, the ABSA process uses SentiCircle to classify sentiments into two, namely positive and negative. The results of the ABSA showed that the performance of proposed method achieve for precision, recall, and f1-measure were 0.84, 0.89, and 0.87, respectively.

199

Authors: Reza Fauzi Iskandar, Edi Leksono, Endra Joelianto

This paper presents a control strategy of maximum power point tracking and focuses on hybrid q-learning and type-2 fuzzy logic control strategy. Photovoltaics have non-linear voltage and current characteristics which influenced by temperature and exposure of solar irradiation so that the maximum power point (MPP) can change at any time. Unfortunately, operating conditions outside the MPP could reduce the efficiency of electrical power transfer from the photovoltaic system to the load. Q-learning (QL) has the potential to provide action decisions of duty cycle signal based on the state of the power gradient percentage range. The Q-learning hybrid method with Type-2 fuzzy logic control is proposed to provide correction of QL control signal by considering the uncertainty direction of power point shifting. The proposed strategy is important particularly to avoid a sudden change in solar irradiation exposure. The simulation results show that the Q-learning hybrid type-2 fuzzy logic control based MPPT response in the photovoltaic control system has tracking efficiency about 97%, rapid rise time of 0.08s, settling time of 0.23s, low overshoot and stable response for handling the change of irradiation and temperature exposure simultaneously.

226

Authors: Manjunath Hanumanthappa Rangappa, Guruprakash Chirathahalli Dyamanna

The combination of both static sensors and mobile sensors is called as Hybrid Wireless Sensor Network (HWSN). Static sensors monitor the environment and report events occurring in the sensing field. Mobile sensors are then dispatched to visit these event locations to conduct more advanced analysis. These sensor nodes have major energy constraints as they are small and their battery can’t be replaced and collaborate together in order to gather, transmit/forward the sensed data to the base station. Consequently, data transmission is one of the biggest reasons for energy depletion in HWSN. Clustering is one of the most effective techniques for energy efficient data transmission in HWSN. This paper proposes an Adaptive Hybrid Cuckoo Search (AHCS) algorithm. AHCS improves the Lévy flight method and population evolution strategy of the cuckoo search algorithm, and introduced a mutation operation operator. Inspired by the idea of position update of Grey Wolf Optimization (GWO) algorithm, this paper introduces the inertia weight w in the Lévy flight method of CS algorithm, and gives the new dynamic adjustment methods of parameters α and β respectively. The proposed algorithm is evaluated in terms of response time, packet drop ratio, energy consumption, end to end delay and dead time. Furthermore, the performance of AHCS-GWO is evaluated with existing methods such as classical algorithms, Evolutionary Multipath Energy-Efficient Routing protocol (EMEER) and Yellow Saddle Goatfish Algorithm (YSGA). The residual energy of the AHCS-GWO method for 200 nodes is 0.57J, it is high when compared to the existing methods.

247

Authors: VietHung Nguyen, DinhToan Nguyen, NhatTan Nguyen

The trend of Industry 4.0 development in the maintenance field requires the big data analysis of system operation for predicting, or identifying any possible problem. This study proposed an automatic fault diagnosis method for rolling bearing. Ensemble empirical mode decomposition (EEMD) is used to preprocess the vibration signals of rolling bearing which solved the problem of mode mixing existing in empirical mode decomposition (EMD). The features are then extracted from the first IMFs in the time and frequency domain to form an initial fault feature of the complex vibration signal. The Laplacian score (LaS) method for selecting features is then utilized to construct a new sensitive feature according to their significances, named EEMD-LaS. An optimized pattern classification method is constructed to achieve an automatic diagnosis of bearing fault. This method uses a backtracking search optimal (BSO) algorithm to optimize the parameters of the least-squares support vector machine (LSSVM) classifier model, named BSO-LSSVM. Finally, the EEMD-LaS and BSO-LSSVM base proposed method for bearing fault diagnosis is applied to analyze the experimental data. The result indicates that the proposed EEMD-LaS – BSO-LSSVM method can effectively diagnose the fault categories with 100% accuracy of the outer-race fault and normal bearing condition; and 97% minimal accuracy of the severe conditions of the inner-race fault and roller fault of rolling bearing.

269

Authors: Awan Uji Krismanto, Indra Soegiarto, Abraham Lomi, Irrine Budi Sulistiawati, Herlambang Setiadi, Muhammad Abdillah

One of the main focus in integrating large scale wind power plant is how to maintain voltage stability under different power injection from wind farm. In this paper, effects of large-scale wind power plants on voltage profile of power system is investigated. Practical test system of South-West Sulawesi, Indonesia with integration of two large scale wind power plants are considered. It was monitored that the increasing wind enhanced the voltage profile of the system. Impacts of wind power integration on power system voltage profile were investigated in this paper. Probability analaysis based on MCS were conduted to observe the impacts of uncertain power injection from wind farm on voltage fluctuation. It was clearly monitored that the probability distribution of bus voltage varied accordingly depending to location and capacity of wind farm. It was also monitored that enhancement of voltage profiles increased in proportion with power injection from wind farm. Thus, having more power production from wind farm results in better load-ability and eventually improved voltage stability condition of power system as shown in bus 31 (the voltage magnitude increased from 0.942 to 0.952 pu).

294

Authors: Rachmad Abdullah, Suhariyanto, Riyanarto Sarno

TripAdvisor is one of the most popular e-commerce platforms in the tourism sector in Indonesia. TripAdvisor give Traveler Choice Award every year in Indonesia through user reviews. However, online text-based reviews are often associated only with evaluation scores that do not pay attention to the context and meaningful content of the review itself, either explicitly or implicitly. Moreover, the sentence structure of the review can have an impact on the goal of the target sentiment which is nothing but an aspect of the review itself. This research discusses aspect based sentiment analysis for explicit and implicit aspects. This research starting with taking the TripAdvisor website restaurant product review dataset to measure customer satisfaction based on four aspect categories of Ambience, Food, Service, and Price. Furthermore, the aspect word extraction and opinion word extraction processes in the case of explicit sentences for simple, compound, complex, and compound-complex sentence structures are carried out using grammatical rule extraction. This research also works on implicit sentence cases for simple sentence structures. Aspect categorization process uses hybrid approach. Aspect and opinion keyword extraction process uses the ELMo-Wikipedia. Then WordNet and TF-ICF are used to expand the meaning of aspect and opinion that has been taken. The last stage is the aspect based sentiment analysis process, both explicit and implicit sentences using SentiCircle. This research can produce two evaluations of sentiment classification, namely positive and negative. The results of the aspect extraction obtained an evaluation of the aspect categorization for each precision 0.82, recall 0.87, and f1-measure 0.86. Meanwhile, the results of the sentiment analysis showed that the respective evaluations for precision 0.87, recall 0.92, and f1-measure 0.89.

320

Authors: Antono Adhi, Budi Santosa, Nurhadi Siswanto

Studies and research related to the Vehicle Routing Problem (VRP) are mostly carried out on simple constraints. The study was conducted to find the shortest route that the carrier will take. Effective algorithms are sought to find better solutions. Another study will develop VRP cases to be closer to real conditions. In actual conditions, the VRP cases are very complex. Several solutions that are considered good can be implemented to solve this problem. VRP in shipping bulk products, such as oil, cements, and fertilizers is one of the daily logistical cases carried out in an archipelago country. Bulk products are shipped from the depot port city to other islands. The optimum solution is needed to make the delivery shorter distance and the products can be sent faster. The solution to the VRP case with a single product can no longer solve the problem of bulk product shipment. Bulk product shipment involves more than one product with more than one ship compartment. The number of ship compartments that are smaller than the number of products will further increase the complexity of the VRP of bulk product shipment. The combination of several metaheuristic methods and strategies of product selection to fill undedicated compartment of ship is used to find the shortest route. Modified Nawas Enscore Ham (M-NEH), Particle Swarm Optimization (PSO), and 3-Opt algorithms are combined to solve VRP. This hybrid algorithm called Hybrid Particle Swarm Optimization (HPSO) is improved and supports each other. HPSO provides a better solution than those algorithms separately and other metaheuristics namely Tabu Search (TS) and Genetic Algorithm (GA). The implementation of the algorithm to solve real data of bulk cement transportation supports the conclusion that HPSO is superior to the other algorithms. It has 0.99% more effective than the second best and about 3.42% more effective than the average total distance of five other compared metaheuristic methods.

348

Authors: Hussein A. A. Al-Khamees, Nabeel Al-A'araji, Eman S. Al-Shamery

Many different applications in the real world can generate huge amount of data, that has unconventional features including massive size, fast access, besides the evolving in its nature; this is data stream. Data stream clustering algorithms began to grow at breakneck speed. evolving Cauchy (eCauchy) is a significant algorithm of density-based data stream clustering. The major limitation of eCauchy is the high number of clusters generated in dynamic environments. This paper presents an evolving model for data stream by optimizing e-Cauchy algorithm to decrease the number of clusters and reach to an ideal number by implementing evolving mechanisms (adding, merging, splitting clusters) based on a specific membership function. Model is tested by two real datasets NSL-KDD99 and keystroke. Proposed model outperforms two other algorithms, e-Cauchy and FEAC-Stream. Model constructs five and four clusters with less time to implement 1.30 and 2.30 minutes respectively for each dataset.

369

Authors: Seyed Alireza Bashiri Mosavi, Seyed Yashar Banihashem

Monitoring transient excursions of power system dynamic response is the most significant task for a reliable power supply. Hence, pursuing the best-laid features alongside a fit learning model for precise mining on the inherent characteristic of fast transient phenomena is considered a great concern on transient stability analysis (TSA). To this end, first, we discussed the root cause involved in selecting the cross-relevance (CR) features that influenced TSA. Furthermore, we regarded the geometric form of CR (GCR) features to measure the partial-whole manner of power system stability limits. Next, we triangulated Python technology, PSS/E, and MATLAB for generating the GCR multivariate trajectories. In the learning phase, to shroud the linear alignment weakness, the elastic kernel plugged into the support vector machine is considered for transient stability prediction (TSP). The results show an average accuracy near 97% with an acceptable average processing time of 691.7 milliseconds on test cases for TSP.

395

Authors: S. Beghin Bose, S. S. Sujatha

Cloud computing is the on-demand availability of internet-based computing services, especially software, large amounts of data storage, operating systems, and other computing resources. Service Level Agreement (SLA) violation is the most critical problem in cloud computing. SLA violation creates many problems for cloud service providers and cloud customers. Due to this, cloud customer gets low-quality cloud service. Thus, designing an effective cloud service recommendation algorithm is a critical research problem in cloud computing. The primary objective of this research is to determine the optimal cloud service from functionally equivalent cloud services that better fit the user's requirements (latency, throughput, response time, and cost). The efficiency of cloud services varies according to the time and location of the virtual machine. First, this proposed method determines the correlation between active user requirements and cloud services. Second, strongly correlated services are separated into two clusters based on the virtual machine's location and the cloud service's data transmission rate. For this purpose, two lightweight clustering algorithms have been proposed. A modified multilayer perceptron algorithm has been developed to recommend the optimal cloud service to the active user from the two clusters. The open-source WS-Dream dataset is used to train and validate the proposed MLP. The training efficiency, prediction accuracy, and performance of the proposed MLP-based cloud service recommendation system are experimentally compared to the existing cloud service recommendation systems analysed in the literature study [20, 2, 22, 23]. Compared to existing cloud service recommendation approaches, the MAE and RMSE values of the proposed cloud service recommendation system are less than one. In terms of accuracy, the suggested method obtains a precession of 94 %, a recall of 97 %, and an F1-Measure of 96 %, all of these are significantly better than the existing cloud service recommendation methods. Finally, experimental results prove that the overall performance of the proposed method's throughput (increase 10 MBPS), latency (reduce 10 ms), the response time (reduce 17 ms) and service recommendation time (reduce 5ms) is more robust than existing methods.

422

Authors: Jai Bhagwan, Sanjeev Kumar

Cloud computing technology is becoming popular in both academia and industries these days because of its dynamic and flexible infrastructure which provides good computing facilities through web networks without a place specification. In order to process the workflows, the cloud technology uses datacenters which consist of hosts having various homogeneous and heterogeneous virtual machines. Virtual machines play the role of real machines with multiple operating systems environments. Due to a wide acceptance of the cloud, the load on the cloud is increasing regularly. So, scheduling plays a key role to manage a machine load by mapping several workflows with available virtual machines. The scheduling can be divided into two types: dynamic and static scheduling. Dynamic scheduling can be done effectively by meta-heuristic computing techniques. In the field of cloud computing, many scientists have developed various algorithms for workflow scheduling in order to schedule the workflows. The Cat Swarm Optimization works fine as compared to the Max-Min algorithm, Particle Swarm, and Ant Colony Optimization algorithms. In this research, a new algorithm is designed named H-CSO by using the concept of Heterogeneous Earliest Finish Time and Cat Swarm Optimization algorithms. After experiments, it is found that the proposed H-CSO algorithm gives efficient makespan at realistic costs as matched to the Cat Swarm Optimization. The newly designed H-CSO algorithm is 2.99%, 2.87%, 3.35%, and 5.77% efficient for CyberShake_1000, Montage_1000, Inspiral_1000, and Sipht_1000 datasets respectively as compared to the standard Cat Swarm Optimization in terms of average makespan reduction. In case of cost reduction, the proposed algorithm is 4.03%, 5.12%, 2.42%, and 3.93 effective as compared to the Cat Swarm Optimization for CyberShake_1000, Montage_1000, Inspiral_1000, and Sipht_1000 datasets respectively.

446

Authors: Lindung Parningotan Manik, Zaenal Akbar, Hani Febri Mustika, Ariani Indrawati, Dwi Setyo Rini, Agusdin Dharma Fefirenta, Tutie Djarwaningsih

Knowledge-based chatbot (KBC) has grown in popularity in recent years and has been widely used for various use cases. Building KBC from scratch using deep learning (DL) is challenging since no prior historical data exists. Meanwhile, DL systems need a vast Volume of data to be trained. This paper proposes a novel framework to create an intent classifier of the KBC used to detect in-scope (IS) and out-of-scope (OOS) intents. We introduce an automated queries generator to create IS intents employed as the training data from an ontology input. We utilize Bidirectional Encode Representations from Transformers (BERT) fine-tuning as the backbone of our DL system. Moreover, we present a Bayesian approach as an extension of the BERT to classify OOS queries with minimal OOS training data. The experiments result show that the proposed method manages to achieve an F1 score of 100% for IS intents and 86% for OOS queries.

469

Authors: Ali Saeed Alfoudi, Ali Hakem Alsaeedi, Mohammed Hamzah Abed, Abayomi M. Otebolaku, Yasmeen Sajid Razooqi

Palm Vein Identification (PVI) is a modern biometric security technique used for enhancing security and authentication systems. The key characteristics of palm vein patterns include its uniqueness to each individual, its unforgettability, non-intrusiveness and its ability for disallowing unauthorized persons. However, the extracted features from the palm vein patterns are huge with high redundancy. In this paper, we propose a combined model of two-Dimensional Discrete Wavelet Transform, Principal Component Analysis (PCA), and Particle Swarm Optimization (PSO) (2D-DWTPP) that feeds wrapper model with an optimal subset of features to enhance the prediction accuracy of -palm vein patterns. The 2D-DWT extract features from palm vein images, using the PCA to reduce the redundancy in palm vein features. The system has been trained to select high recognition features based on the wrapper model. The proposed system uses four classifiers as an objective function to determine PVI which include Support Vector Machine (SVM), K Nearest Neighbor (KNN), Decision Tree (DT) and Naïve Bayes (NB). The empirical results proved that the proposed model has the best results with SVM. Moreover, our proposed 2D-DWTPP model has been evaluated and the results show remarkable efficiency in comparison with AlexNet and other classifiers without feature selection. Experimentally, the proposed model has better accuracy as reflected by 98.65% whereas AlexNet has 63.5% accuracy and the classifier without feature selection process has 78.79% accuracy.

492

Authors: Mohammed Fadhil Ibrahim, Huda Khurshed Shawkat Aljader, Nawar Abbood Fadhil

Covid-19 pandemic affected our daily activities. Education is the fastest aspect that has been hindered due to the pandemic. The majority of the schools and universities were prone to global closure. The world turned toward using e-learning as an alternative to traditional learning due to the strict lockdown procedures. This study presents a model to measure the students’ intention to use e-learning during the pandemic in Iraq. Technology Acceptance Model (TAM) is utilized with some external variables. Task Technology Fit (TTF) is also adopted as a moderator for the proposed model. The model is implemented on a sample of students studying at Technical College of Management – Baghdad who have completed the academic year using e-learning. Eighteen hypotheses were formulated; out of them, only two were rejected. The model is tested and evaluated using PLS Tool. The results showed that the students have a positive intention to use e-learning during the covid-19 pandemic; also, most of the external variables were statistically significant. Furthermore, using TTF positively moderates the proposed model. The results open the way to perform more investigations in adopting e-learning during unusual situations.

517

Authors: Prabhakaran Mathialagan, Malathy Chidambaranathan

Upper Aero Digestive Tract (UADT) cancer is one of the most common cancer types in any gender. Early detection and diagnosing such type of cancer will reduce the risk of death in human. In this paper we addressed the state-of-the-art solutions for two major pathological constraints like, dysplasia type tumour grading and tumour grade classification with artifact’s present in biopsy images. A new Whole Slide and patch-based CNN model was proposed which involves different pre-processing techniques to detect and normalize the stain artifacts present in biopsy tissue samples. The proposed CNN model is developed to classify different oral cavity tumour sites and its gradings automatically. It was observed that the proposed CNN model achieved best accuracy of 98 % with zero classification error rate. These findings will throw a light for pathologists to handle digital biopsy images with artifacts and also will classify dysplasia type tumour grades automatically.

539

Authors: Dipak Patel, Kiran Amin

Nowadays, the rapid growth of internet has led to the way for easiest data generation. Sentiment Analysis (SA) is necessary to scrutinize the user-generated data for classifying the sentiment polarity into anyone of the two classifications (negative and positive). SA is implemented by machine learning techniques and lexicon oriented techniques. However, the traditional SA models could not deal with missing labelled data and internet slang data. This paper aims to propose a new multi-source sentiment analysis model that includes six stages to perform the task of domain adaptation. For checking similarity between statements, higher order statistics based on text features is exploited. The higher order statistics based features extraction includes use of modified cross entropy measure. Remaining features are then given to proposed classifier that predicts the polarity of target domain in a precise way. For classification purpose Neural Network (NN) is exploited. Particularly, the weights of NN are tuned in an optimal manner using Improved Grey Wolf Optimization (IGWO) algorithm, which is the enhanced version of GWO algorithm. Finally, the performance of the proposed model is compared over several conventional methods of domain adaptation and gained an improvement of 28% to 6% in accuracy.

559

Authors: Mohammed Amine Hassoine, Fouad Lahlou, Adnane Addaim, Abdessalam Ait Madi

This paper investigates the optimal layout of wind turbines (WTs) within a wind farm. Finding the best placement of WTs in a wind farm is a challenging process due to the existence of multiple wake effects. A biogeography based optimization (BBO) algorithm method is proposed to search for the optimal location of WTs in a wind farm (WF), to maximize the power produced by the WF and improve the annual economic performance of the WF. A wind turbine (WT) that operates in the wake of one or more other turbines is subject to lower flow and therefore produces less power. When designing a wind project, the arrangement of the turbines with each other in a wind farm is a very important factor. The best layout of a wind farm is to achieve the optimal placement of the turbines in relation to each other in a given area to maximize the efficiency of the whole wind farm and reduce its cost. A dense configuration would result in considerable power losses. Each turbine must have a sufficient distance from other turbines in the WF where the optimal number of turbines should be placed. The BBO approach is conducted on a 2 km x 2 km wind farm assuming a constant wind speed of 12 m/s with a fixed wind direction, for solving the wind farm layout optimization problem in two different configurations which include 26 and 30 WTs respectively. A comparison of the results obtained with the previous studies shows that the BBO is more efficient in terms of maximizing power output and economic profitability of the same wind farm model, which validates that BBO performs effectively in optimizing WTs placement within WF. BOO provides the greatest improvement in the optimal layout, for example, in the case of the layout for 30 WTs. The power output reaches 15,383 KW, the agreement between the ideal and the optimal layout is more favorable. The difference in output power is only 169 KW (1%). Knowing that the ideal layout is reached if all WTs receive a wind flow with a maximum wind speed of 12 m/s. Furthermore, a case study of wind turbine layout optimization using the BBO program on the Alta X wind farm has been performed under variable wind speed and variable wind direction. The results indicate that the optimized layout of the Alta X wind farm achieves a 12% increase in the power output for a similar cost when compared to the original layout of Alta X. It is also more appropriate for evaluating the wind farm layout in the Wind Power Project (WPP).

583

Authors: Abdelghani Aib, Djalal Eddine Khodja, Loutfi Benyettou, Salim Chakroune

This article provides a hardware implementation of torque and flux estimators for direct torque control (DTC) of an induction motor (IM) on the Field Programmable Gate Array (FPGA). Due to its high processing frequency, the FPGA circuit presents an alternative strategy for achieving a high performance DTC implementation. This cannot be achieved by Digital Signal Processing (DSP) or Microcontroller application. Optimized hardware architecture is also proposed for implementation of a complete estimator with explicit determination of all the quantities such as the flux phase, the flux amplitude and the torque amplitude. This work presents a precise hardware estimator of the torque and the flux, it developed with negligible estimation error (0.0002 [Wb] RMS estimate error for the stator flux and 0.0005 [Nm] RMS estimate error the torque), in addition to that this estimator is implemented on the Virtex-4 FPGA board with minimal resources(less 5% of slices registers, 5% of DSP48Es and around 17% of LUTS). This estimator is developed and synthesized independently of the command and can be reused for any task which requires the estimation of these quantities. After this the estimator was used to complete a DTC control in a hardware co-simulation procedure with the Xilinx system generator and MATLAB/SIMULINK tools on the Xilinx ML402 development kit.

1

Authors: Rashed Salem, Wafaa Abdel-Moneim, Mohamed Hassan

Large graphs are available in everywhere such as social networks, web link analysis, and computer networks. Traditional methods of clustering are not suitable to tackle the problem of clustering large graphs because the computation is very costly, which is solved by local graph clustering using a given vertex set as input to detect an accurate cluster. SimpleLocal (SL) algorithm detects a best conductance cuts close to seed vertices set. In this paper, a new Parallel SimpleLocal (PSL) system is proposed using multicore CPUs. OpenMP parallel library is utilized to parallelize the first and second stages of 3StageFlow algorithm whereas the SL algorithm is used for enhancing the runtime. The experiments are performed on various applications from different domains, which are image segmentation and community detection. From the experiments, the proposed method improves the runtimes with 75.43% using 4-cores and 81.01% when using 8-cores over the sequential single core.

22

Authors: Phyu Phyu Khaing, May The` Yu

Image captioning is systematically generating the caption of the image with a sentence description. In the past few years, the automatic process of creating image caption has fascinated the great interest in Artificial Intelligence (AI) field. Image captioning defines as the basic process of building the conjunction of image processing and natural language processing at input and output position. All image processing tasks, such as the segmentation of image, object tracking, object detection, image recognition, and many others, are mostly performed using Convolutional Neural Networks (CNNs). To perform the natural language processing tasks, just as semantic role labelling, neural machine translation, speech recognition, question and answering, and many others, Recurrent Neural Networks (RNNs) and long-term memory networks (LSTMs) are essential for some of the biggest breakthroughs. This paper proposes the efficient encoder-decoder framework-based image captioning model, namely Two-Tier LSTM (TT-LSTM) Model. The TT-LSTM model is designedly implemented upon the encoder-decoder framework with two LSTM layers. This research is implemented on the MSCOCO, Flickr30k, and Flickr8k datasets; and evaluated with standard evaluation matrixes such as ROUGE-L, CIDEr, and four BLEU scores. The outcomes of the experiments on the typical datasets reveal that the proposed model generates meaningful natural language sentences. The proposed model also improves the sentence generation efficiency and can achieve better performance for image caption generation.

43

Authors: Nitesh Sureja, Avani Vasant, Nandini Chaudhari

Data clustering is a collection of data objects similar to one another within the same cluster and dissimilar to the objects in other clusters. The shuffled frog-leaping algorithm is a nature-inspired algorithm that mimics the natural biological evolution process of frogs. This algorithm also consists of elements like local search and exchanging information globally. This algorithm faces the problem of converging in local optima due to the limitations of the local search method used to explore search space. In this paper, a hybrid shuffled frog-leaping algorithm is introduced for clustering. The proposed algorithm uses a simulated annealing search method instead of a simple local search to improve the search behavior for selecting fitter solutions required in each iteration. Six benchmark datasets are used to validate the performance of the proposed algorithm. Quality measures used are purity, entropy, completeness score (CS), homogeneity score (HS), and FMeasure (FM). Fitness functions used to optimize are total within-cluster variance (TW) and the Silhouette coefficient (SC). Results obtained are compared with the results of twelve other state of the art algorithms. Results stored in the tables clearly shows that our proposed algorithm outperforms other algorithms in terms of quality. Results also prove that the proposed algorithm converges in the significantly less amount of time and eliminates local optima problem also.

56

Authors: Santhosh Kumar Kenkere Basavaraj, Bangalore Ramchandra Sujatha

Multiple-Input Multiple-Output Orthogonal Frequency Division Multiplexing (MIMO-OFDM) is widely used to provide high speed data transmission and spectrum efficiency in modern wireless communication systems. Specifically, the transmission of Electrocardiogram (ECG) signal plays a main role in health monitoring systems. The privacy and security of the patient identification and medical records are considered as a main concern in health monitoring systems. In this paper, the Lightweight cryptography (LWC) is proposed to secure the ECG signal transmission from unauthorized users through the MIMO-OFDM system. The LWC is mainly used to minimize the amount of logical elements using the gate level architecture and simple key schedule in the MIMO-OFDM. The turbo code is used in MIMO-OFDM is due to its error correcting capacity that minimizes the amount of error caused during communication under the constraints of burst error and Inter Symbol Interference (ISI). Here, the ECG signals from the MIT arrhythmia database is used to analyse the secure ECG signal transmission of LWC-MIMO-OFDM method. The performance of the proposed LWC-MIMO-OFDM is taken by means of area, delay, power, number of slices, flipflops and LUTs. The LWC-MIMO-OFDM method is compared with Advanced Encryption Standard (AES) to evaluate the efficiency of LWC-MIMO-OFDM. The delay of the LWC-MIMO-OFDM for Virtex 5 device is 13.3ns, it is less when compared to the delay caused by the AES.

77

Authors: Hicham Bouzakri, Ahmed Abbou, Khalid Chennoufi

The optimal installation of a photovoltaic panel in fixed mode, allows us to receive the maximum of solar radiation at noon, when the sun is in front of the panel, although the output drops in the morning and afternoon. In order to maintain this efficiency as much as possible, the photovoltaic panel must be oriented by a solar tracker. However, the introduction of such a mechanism makes the installation expensive and complicated during maintenance, especially since a part of the energy produced by the panel will itself be consumed for its operation. This paper proposes a simple and less expensive system allowing the photovoltaic panel in fixed mode to maintain its efficiency almost at the maximum all day long with the introduction of a simple reflection system based on two reversible mirrors at noon. After a theoretical study and on the basis of a simulation, the proposed system will have an efficiency of more than about 35% compared to that in fixed mode and more than 90% compared to that in tracking mode, these results will be validated via an experiment at the end of the document with a prototype produced.

105

Authors: M. Ganesh Karthik, M. B. Mukesh Krishnan

In recent decades, the Internet of Things (IoT) is a growing technology in smart applications, where it is highly susceptible to security breaches and the resources are constrained in nature. Hence, the growth of IoT devices allows the hackers to take benefit of the communication capabilities to detect different types of attacks such from NSL-KDD such as Denial of Service (DoS), Probe, Remote to Local (R2L) and User to Root (U2R) attacks. The existing deep networks used larger data in training because the size of non-predictive parameters for learning required huge potential to attack detection. In order to overcome the problem occurred in the existing models, an efficient Post Pruning Decision Tree-Synthetic Minority Over-Sampling Technique (PPDT-SMOTE) is proposed in the research work. The proposed PPDT-SMOTE eliminates or pruned the non-predictive parameters from the huge data samples and SMOTE solves the data imbalance problems as they use the prominent samples from the non-pruned regions. Thus the PPDT models improves the learning rate of the system for huge data and SMOTE will overcome the problem of class imbalance problem as the learning rate is improved. The results obtained from the proposed PPDT-SMOTE approach effectively detects the DoS, Probe, U2R, and R2L attacks in terms of accuracy as 98.04% better when compared to the existing shallow models of 95.22%, Routing Protocol for Low Power and Lossy Networks (RPL) of 91.5 % and Multi-Convolution Neural Network (CNN) fusion model of 64.81% in the IoT environment.

123

Authors: Phyu Myo The, May The` Yu

Hand gesture detection and recognition is a way to communicate between deaf-mute person and rest of society. In this study, the static sign and the dynamic sign recognition system based on hand gesture is presented. The dynamic sign gesture is more challenging than static sign gesture which contains the movement. The vision based sign language recognition using hand gesture consists of the following main steps: acquisition the sign data, detection the hand region, extraction the features and recognition. The data acquisition stage, the data acquired by using web camera without using any data gloves and special markers. In hand region detection process, this study proposed hybrid hand region detection process by combining CbCr channel from YCbCr colour space and motion detection by using mean filter background subtraction to definitely segment the hand region from the background. In hand feature extraction, the scale, shape, texture and orientation features extracted by using GIST (Generalized Search Tree), HOG (histogram of gradient) and HOG-LBP (Local Binary Pattern) methods. Finally, Quadratic Support Vector Machine (QSVM), Cubic Support Vector Machine (CSVM) and Linear Discriminant recognized the static and dynamic hand gesture. The recognition accuracy achieved 86.7% and 99.22 % acceptable accuracy on self-construct dynamic Myanmar sign word dataset and MUDB dataset.

146

Authors: Bhagirathi Bai Vittal Rao, Hanume Gowda Chandrakanth

Presently, industrial automation has become a popular field because of its various advantages includes higher production rates, more efficient use of materials, better product quality, improved safety, and requires minimum labours. This is achieved by utilizing Local Networking Standards (LNSs), remotely monitoring and controlling industrial devices by utilizing Raspberry Pi and Embedded Web Server (EWS) technology. This proposed research provides an idea of utilizing Internet of Things (IoT) for monitoring and controlling the automation process through Wi-Fi or wireless medium by utilizing Raspberry pi as a server system. Additionally, the prediction and error detection utilized in the machine learning process for this Improved Random Forest (IRF) method. The proposed IRF uses Out of Bag (OoB) bagging estimation technique to randomly select sub-datasets for overcoming the optimization problem. The OoB estimation is the technique used to find prediction error in IRF as every samples are not used when each tree in IRF is trained. So for all those bags unused samples estimate the prediction error for a particular bag in prediction process. IRF recursively partition the data into categories and the derived RF model and maps the results. The proposed IRF method overcomes the overfitting problem, receives the command and applies action according to it. The IRF method uses only eight devices to control and monitor more than thirty devices and this entire process is represented as IoT based Machine Learning (ML) Automation Algorithm. The experimental results show that the IRF method provides better outcomes in terms of accuracy, precision, F-measure, and recall. The RMSE values obtained for the proposed IRF model shows 0.0386 lesser error values when compared with the existing models that had achieved RMSE as 3.607 for Long Short Term Memory- Recurrent Neural Network (LSTM-RNN), Artificial Neural Network (ANN) as 1.226 and Fuzzy Gain Scheduling (FGS) of 0.4247.

166

Authors: Vasudha Bahl, Anoop Bhola

WSNs embedded device can be extended to a wide range of implementation in reality. Clustering is efficient way to lessen the energy utilization and improve WSN's lifespan. To improvise network lifespan numerous clustering approaches, implement various parameters for election of CH. An effective clustering algorithm depends upon the number of factors such as number of CHs, uniform cluster size, CHs distribution, energy of the CHs etc. In our research we strengthen our methodology for election of cluster head in HWSN depending on multiple node parameters such as distance, density and residual energy. This paper aims to optimize energy and improve network with Energy Balanced Cluster Technique (EBCT). In our research we reformulated for probability estimation to identify the CHs in each round characterized by node parameters: distance, density, residual energy and node dormancy mechanism. Mathematical analysis and simulations show the proposed method extends the service life by around 8% to 53% relative to the other protocols and optimizes energy utilization of HWSNs.

190

Authors: Abdelmoula Saggai, Samir Zeghlache, Loutfi Benyettou

This work presents a reliable fault-tolerant tracking control system (FTTCS) for actuator faults in a octorotor unmanned aerial vehicle. The proposed FTTCS is designed based on adaptive fuzzy global fast dynamic terminal sliding mode control (AFGFDTSMC) that guarantees the global asymptotic stability of a octorotor system. To mitigate the negative impacts of model uncertainties and enhance system robustness in faulty operation, an adaptive fuzzy system is incorporated into the global fast dynamic terminal sliding mode control (GFDTSMC) scheme for adaptively identifying the model uncertainties online and compensate the actuator faults effect. Lyapunov stability analysis proofs that the developed control method design maintains the stability in the closed loop dynamics of octorotor UAV in faulty operation. Simulation results show that the proposed (FTTCS) enables the octorotor to track the desired reference commands in the presence of actuator faults with satisfactory performance.

217

Authors: Aris Marjuni, Oky Dwi Nurhayati

Image watermarking is a technique to embed a piece of information or data, which also can be an image, into the host or medium image for specific purposes, such as authentication, security, or copyright protection. A good image watermarking method should have a good perceptibility after adding the watermark. It should also robust against attacks and the watermark should be able to be recovered although the watermarked image has distorted. This paper presents the proposed lifting image watermarking scheme based on wavelet transform through singular value and Schur decomposition. This method is proposed to improve the perceptibility and the robustness of the image watermarking scheme, especially against non-geometrical attacks. The lifting scheme is performed by lifting wavelet transform to produce a better imperceptibility. The use of singular value decomposition is used to increase the intactness of the embedded watermark from the distorted image, which is performed on the triangular matrix of low-frequency sub-band produced by Schur decomposition. In perceptibility performance, experimental results show that the proposed method could produce a watermarked image very close to the host image with more than 99% of similarity level. In robustness performance, the proposed method can recover the embedded watermark even though the watermarked image has been distorted by non-geometrical attacks with more than 90% of normalized correlation average.

241

Authors: Pradeep Krishnadoss, Gobalakrishnan Natesan, Javid Ali, Manikandan Nanjappan, Parkavi Krishnamoorthy, Vijayakumar Kedalu Poornachary

Services are delivered promptly and efficiently managed over the Internet using cloud computing. Data and storage services are offered to users through cloud computing without regard to the actual physical location of the users. The cloud environment includes huge number of tasks and computing resources. Identification of appropriate virtual machine (VM) for allocation of resources to complete any submitted task is done through the task scheduling algorithms, which play a significant role in the cloud computing process. Task scheduling techniques improvise the makespan of the schedule and account for considerable reduction of cost expended. In this paper, an efficient hybridized scheduling algorithm that replicates the parasitic behaviour of the cuckoo and food gathering habit of the crow bird, named Cuckoo Crow Search Algorithm (CCSA) had been presented for improvising the task scheduling process. The crow bird always stares at its neighbours looking for a better food source than the one it currently possesses. At some circumstances the crow even goes a step further and steals its neighbour’s food. The CCSA had been inspired from such characteristics of these birds and had been designed to be applied in the cloud environment for identifying a suitable VM for carrying out the task scheduling process. The proposed CCSA accounts for reduction in makespan and cost and its performance had been experimentally verified by comparing it with other existing algorithms like Multi objective - Ant Colony Optimization (MO-ACO), ACO, Min-Min and the results had been presented. The proposed CCSA technique had produced an improvement of 3.14%, 10.70%, and 21.60% for makespan and had reduced the overall cost to the tune of 4.56%, 11.19% and 19.35% when compared with MO-ACO, ACO, Min-Min algorithms respectively when used with 10 VMs. The detailed description of results obtained had been described in result section.

264

Authors: Sethakarn Prongnuch, Suchada Sitjongsataporn

This paper presents a spherical magnetic robot named ‘CE-R1’ propelled by an inner drive unit and a magnetic balancing unit rolling without slipping on a plane. A spherical magnetic robot ‘CE-R1’ consists of two parts as a domed head and a ball-shaped body with contactless and controlled via Bluetooth. A non-holonomic model and wheel-drive locomotion are interested in a spherical mobile robot. Sphere movement is controlled by a differential drive from the inner drive unit inside the sphere. Firstly, the magnetic interaction in a magnetic balancing unit is applied to the connection of a domed head and a ball-shaped body with contactless. Secondly, a differential drive of proposed spherical magnetic robot ‘CE-R1’ with an inner drive unit inside is derived and deployed with the multi-single computer board controller. Hardware of prototype ‘CE-R1’ is designed including the multi-single board computer for wireless charger, voice recognition system and controlling the ‘CE-R1’. The explicit experiments of proposed ‘CE-R1’ spherical magnetic robot are tested as a faculty-assistive guide robot controlled through application at the faculty lobby area in the Suan Sunandha Rajabhat University. This work proposed here is a step towards the overall goal of an autonomous spherical mobile robot.

289

Authors: Ayad Mohammed Jabba

The wide amount of data in the modern applications available on the Internet make it very complicated to deal with the knowledge behind these data. The data classification task is a useful tool that used to deal with a huge amount of data by classify these data into coherent groups. The data size decreases the performace of the classification technique, especially when contain uninformative data (i.e., irrelevant, noisy). The stochastic local search algorithm is an optimization approaches employed to find the informative data to build the classification model. These methods are used to search for the optimal patterns to construct the classification algorithm. Thus, this research introduces a stochastic local search algorithm for rule induction called Iterated-Miner (Iterated local search-based rule induction). The purpose of the algorithm is to construct classification rules from data. This classification algorithm is inspired by the concepts and principles of stochastic local search and rule induction. The performance of the proposed classifier evaluated with a well-known and state of art classification algorithms called, Ant-Miner, ACO/PSO2, PART FURIA, and ACO/GA on 10 UCI datasets. The results demonstrate that our classifier is superior compare with all classifiers respect to classification accuracy; and the model sizes by our classifier are considerably competitive with those discovered by other classifier.

312

Authors: Hilal Rahali, Samir Zeghlache, Loutfi Benyettou

In this study, an adaptive fuzzy type-2 backstepping controller (AFT2BC) has been proposed for an industrial PUMA560 manipulator robot with a variable load and actuator faults. The method realizes an accurate tracking of the trajectory at the end effectors of the manipulator and improves its robustness toward model uncertainty (payload variation) and actuator faults when controlling the position of the robot. By using the advantages of the backstepping control, the convergence speed of the control algorithm has been improved, and its steady-state error has been reduced. Also, integration of the continuous approximation law has been used to eliminate the real-time chattering during the control process without affecting the robustness of the system. A type-2 fuzzy adaptive control law has been designed in order to guaranty faults and uncertainties compensation, small tracking error, and fast transient response, where the prior knowledge of uncertainties and external disturbances is not required, without disappearing the tracking precision and robustness property. Finally, the stability of the controller has been proved by the Lyapunov theory, and comparative simulations in faulty operation are conducted to show the superiority of the developed control strategy.

337

Authors: Aradea Aradea, Rianto Rianto, Husni Mubarok

Internet of Things (hereafter, IoT) service is a complex system because it should meet miscellaneous domain forms represented physically and virtually. The main challenge of IoT is to provide an inference model to resolve the dynamic context on a run-time basis. The system should have the ability to catch instances or concrete IoT services. On the other side, it should have the capability to adapt to the newest evidence of contexts. This paper introduces an inference model consisting of an IoT structure service artifact, a subsystem of contextual knowledge, and a subsystem of run-time adaptability reasoning. The results of model implementation on monitoring system of coronavirus disease revealed that the ability to adapt continuously and provide various alternative solutions to handle uncertain contexts, which is refered to sensor, network and server failure. The example of experiment result when a sensor failure occurs, the data received by the main server from the node is the average of the three previous data.

373

Authors: Nahlah Abdulrahman Alkhalidi, Fouad A. Yaseen

Traffic classification is referred to as the task of categorizing traffic flows into application-aware classes such as chats, streaming, VoIP, etc. Most systems of network traffic identification are based on features. These features may be static signatures, port numbers, statistical characteristics, and so on. Current methods of data flow classification are effective, they still lack new inventive approaches to meet the needs of vital points such as real-time traffic classification, low power consumption, ), Central Processing Unit (CPU) utilization, etc. Our novel Fast Deep Packet Header Inspection (FDPHI) traffic classification proposal employs 1 Dimension Convolution Neural Network (1D-CNN) to automatically learn more representational characteristics of traffic flow types; by considering only the position of the selected bits from the packet header. The proposal a learning approach based on deep packet inspection which integrates both feature extraction and classification phases into one system. The results show that the FDPHI works very well on the applications of feature learning. Also, it presents powerful adequate traffic classification results in terms of energy consumption (70% less power CPU utilization around 48% less), and processing time (310% for IPv4 and 595% for IPv6).

393

Authors: Bishoy Sameeh Sawiris, Sayed Abd El-Gaber, Manal A. Abdel-Fattah

Centralized big data is an emerging field that manages information in one common repository. Currently, Internet of Things (IoT) devices are increasing massively for sensing and transmitting data to the big data environment. This information is sensitive and easily traced by attackers due to the centralized data management. To address the security and privacy issues and avoid the system in a single point of failure, in this paper the researchers proposed a new model, namely, BOBS CRABID i.e. Blockchain Based Secure Centralized Big Data model using distributed computing approach. In big data, Hadoop MapReduce is one of the distributed computing approaches which handles all information in a distributed manner. The BOBS CRABID model is designed by three ideas as Multi-Factor authentication, IoT Devices Data Collection and Processing, and Optics based MapReduce for Data Clustering. In multi-factor authentication, a lightweight camellia key generation algorithm (LCKGA) is used for their authenticating all devices based on ID, IP address, MAC and PUF. These credentials are saved in Blockchain Security Entity for authentication verification. All transactions are hashed and stored in the blockchain using Keccak hash algorithm which performs better than the traditional hashing (SHA) 2. Then the researchers conducted IoT devices data collection and pre-processing using Cross Correlation and Min-Max Normalization for redundant data pruning and normalization, respectively. Finally, clustering is performed in a distributed way for optimizing the storage and scalability performance. For that MapReduce is used in which OPTICS algorithm is applied for clustering based on the data size, type of context and nature of data (sensitive or non-sensitive). In this way, data is stored in the big data environment. Experiments were conducted for the proposed BOBS CRABID model and compared with the well-known methods using Hadoop 2.7.2. The proposed BOBS-CRABID model achieves better performance in terms of throughput (increase 1500mbps), response time (reduce 150ms), energy consumption (reduce 35%), attack detection rate (increase 30%), accuracy (increase 20%), computation overhead (300 kb), and time reduce 4s).

422

Authors: Anuruk Nosan, Suchada Sitjongsataporn

In this paper, a proposed absolute sort delta mean (ASDM) method obtaining the speech feature extraction for noise robustness is developed from mel-frequency cepstral coefficients (MFCC) named ASDM-MFCC, in order to increase robustness against the different types of environmental noises. This method is used to suppress the noise effects by finding a rearranging average of power spectrum magnitude combined with triangular bandpass filtering. Firstly, the spectral power magnitudes are sorted in each frequency band of the speech signal. Secondly, the absolute-delta values are arranged and then a mean value is determined in the last step. The purpose of proposed ASDM-MFCC algorithm is to require the noise robustness of the feature vector extracted from the speech signal with the characteristic coefficients. The NOIZEUS noisy speech corpus dataset is used to evaluate the performance of proposed ASDM-MFCC algorithm by Euclidean distance method with the low computation complexity. Experimental results show that the proposed method can provide significantly the improvement in terms of accuracy at low signal to noise ratio (SNR). In the case of car and station at SNR=5dB, the proposed approach can outperform in comparison with the conventional MFCC and gammatone frequency cepstral coefficient (GFCC) by 80% and 76.67%, respectively. Obviously, some experimental results of the proposed ASDM-MFCC algorithm are more robust than the traditional one.

452

Authors: Krit Chantarasamai, On-Uma Lasunon

We describe a Differential Evolution (DE) algorithm for solving U-shaped Assembly Line Balancing Problems – Type 2 (UALBP-2). The minimum cycle time in a just-in-time production line for producing a single product with a certain number of workstations was investigated by developing solution methods and testing on 15 problem sets (101 instances). The problems were divided into 10 medium-scale problem sets (50 instances) and 5 large-scale problem sets (51 instances). When comparing the results with those obtained from the rule-based heuristic; three rules and two rules, the DE algorithm could generate 14 better solutions (28%) in the medium-scale problems, and 3 better solutions (about 6%) in the large-scale problems. In terms of computational time, DE algorithm was faster than rule-based heuristic methods in medium-scale problems. In large problems, our DE algorithm used computational times that were shorter by 73% (for two rules) and 98% (for three rules).

472

Authors: Fatemeh Ahmadi Zeidabadi, Sajjad Amiri Doumari, Mohammad Dehghani, Om Parkash Malik

There are numerous optimization problems in various sciences that need to be solved using the appropriate technique. One of the most widely used techniques for solving optimization problems are population-based optimization algorithms. The innovation and contribution of this paper is to design a new optimizer called Mixed Leader Based Optimizer (MLBO) to solve optimization problems. The main idea in the proposed MLBO is to create a new member as a leader by mixing the best population member and a random member to guide the algorithm population. The main advantage and feature of the proposed MLBO is that it has no control parameters and therefore no need to adjust the parameter. The proposed MLBO algorithm is mathematically formulated to implement in solving various optimization problems. The capability of the proposed optimizer in optimizing and providing appropriate solutions has been tested on a set of twenty-three standard objective functions. These objective functions are selected from three different types including unimodal, high-dimensional multimodal, and fixed-dimensional multimodal in order to analyze different aspects of optimization algorithms. Also, in order to analyze the obtained optimization results, the performance of the MLBO is compared with eight other well-known algorithms including Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Teaching Learning-Based Optimization (TLBO), Gravitational Search Algorithm (GSA), Gray Wolf Optimizer (GWO), Emperor Penguin Optimizer (EPO), Shell Game Optimization (SGO), and Hide Objects Game Optimization (HOGO). The obtained optimization results from the MLBO show the proper performance of the proposed algorithm in solving various optimization problems. On the other hand, comparing the performance of the MLBO with the other eight optimization algorithms indicates the superiority of the proposed optimizer over the compared algorithms.

500

Authors: Seyed Alireza Bashiri Mosavi

The corrective action for power systems over transient space requires the transient stability prediction (TSP) in an accurate and timely manner. To this end, applying data mining techniques to achieve high performance on TSP is an inevitable approach. One of the influential factors on TSP performance is extracting the most relevant features (MRFs) of transient data. In this paper, first, we constructed the transient dataset in the form of reactive power-based two-variate time series (RP2vTS). Next, for exploiting the optimal feature subset, the MRFs of RP2vTS are obtained by the 1-persistence parallel fragmented hybrid feature selection scheme (PFHFS) based on filter and wrapper methods. Finally, to evaluate the efficacy of the proposed framework, experimental comparisons on inter-connected New England and New York systems (NETS-NYPS) were applied using the support vector machine (SVM) classifier. The results showed that the proposed framework by selecting MRFs of RP2vTS offers high-performance capacity on TSP.

524

Authors: Shahbaa I. Khaleel

Estimating models in software engineering are used to estimate some important and future characteristics of the software project, such as estimating the developed project effort, and that failure in the program is mainly due to wrong project management practices, so estimating the software effort is a very important step in the software management process for large projects. In this research, a software estimation tool was built to find an efficient and accurate method for estimating the effort. The average COnstructive COst MOdel COCOMO was used, which is classified as one of the best traditional methods among arithmetic estimation models. Four methods of swarm intelligence were used, the first of which is the Glowing worm Swarm Optimization GSO method and the second is the Bird Swarm Algorithm BSA and the third is the first proposed method hybrid BSA-GSO Method1 BSA-GSOM1, where the GSO and BSA algorithms were hybridized, and the performance of the third method was improved to form the fourth method, represented by the second hybrid new method, which called hybrid BSA-GSO Method2 BSA-GSOM2. The new tool was implemented with all its methods on the NASA data set and satisfactory results were obtained by the first and second swarms intelligence, and excellent results were obtained in the first proposed method, but the results of the second proposed method were better and more accurate than the previous ones. many measurements of performance were used for all the methods, the second proposed method yielded the best results from everyone.

548

Authors: Muhamad Asvial, Agatha Cracias, Muh. Asnoer Laagu, Ajib Setyo Arifin

In this paper, we create a network for IoT applications that able to reduce power to a minimum. The system uses a combination of objective function zero and the modified E-Trickle (ME-Trickle) algorithm to reduce power usage. The design and simulation are conducted with the Contiki OS, Cooja simulator. The system performance is evaluated based on the parameters of power consumption, packet delivery ratio, latency, and convergence time on the network. The proposed system in the simulation result can reduce the convergence time by 15%, power consumption by 8%, and latency by 9%. This system also continues to provide efficient performance with an average PDR of 93.8%. The simulation concluded that MRHOF provides better performance in network quality, whereas OF0 has a faster network convergence and consumes less power than MRHOF. Therefore, the use of ME-Trickle and OF0 will be appropriate for network scenarios with limited resources.

571

Authors: Nurul Zainal Fanani, Adri Gabriel Sooai, Khamid, Festa Yumpi Rahmanawati, Alex Tormasi, Laszlo T. Koczy, Surya Sumpeno, Mauridhi Hery Purnomo

The increase of the classification accuracy level has become an important problem in machine learning especially in diverse data-set that contain the outlier data. In the data stream or the data from sensor readings that produce large data, it allows a lot of noise to occur. It makes the performance of the machine learning model is disrupted or even decreased. Therefore, clean data from noise is needed to obtain good accuracy and to improve the performance of the machine learning model. This research proposes a two-stages for detecting and removing outlier data by using the covariance estimator and isolation forest methods as pre-processing in the classification process to determine fine motor skill (FMS). The dataset was generated from the process of recording data directly during cursive writing by using a digitizer. The data included the relative position of the stylus on the digitizer board. x position, y position, z position, and pressure values are then used as features in the classification process. In the process of observation and recording, the generated data was very huge so some of them produce the outlier data. From the experimental results that have been implemented, the level of accuracy in the FMS classification process increases between 0.5-1% by using the Random Forest classifier after the detection and outlier removal by using covariance estimator and isolation forest. The highest accuracy rate achieves 98.05% compared to the accuracy without outlier removal, which is only about 97.3%.

1

Authors: Girija Vani Gurram, Noorullah Shariff Chowdary, Rajkumar Laxmikanth Biradar

Security of Wireless Sensor Node (WSN) is an important issue due to inherited nature and uncontrolled operation, which makes it vulnerable to many attacks Sybil attacks, denial of service attacks, etc. The main issue related to WSN is the key management system, as it plays an important role in providing security services for WSNs. However, developing a distribution/establishment scheme for WSNs is challenging due to limited sensor sources. In this study, three schemes namely proxy-based path key establishment (PPK), friend-based path key establishment (FPK) and disjoint path key establishment (DPK) are developed. The main problem of path key explore is minimized by passing the nodes with high value of attack coefficient. In PPK model, the Grey-Wolf Optimization (GWO) is used to optimize the threshold value for k-key fragment number and categorization of nodes. The nodes having the common key with destinations are identified by FPK and the best path to reach the destinations is selected by DPK. According to the path compromise ratio and probability of key recovery, the resistance to the effectiveness of three proposed schemes are verified. The results proved that the proposed PPK-GWO achieved probability of key recovery 0.8 for 100 captured nodes, where the DPK achieved the probability of path compromise 0.95 for 10 captured nodes.

23

Authors: Hady Abdel-Maksoud

DC motor drive is still widely employed in manufacturing and transportation systems. The control schemes require an accurate model for recognizing the motor performance. Some or completely the motor parameters may be unknown for the designer. In this research, the application of farmland fertility optimization (FFO) algorithm to identify the parameters of DC motor is presented. This paper purposes to develop an efficient DC model to characterize its performance under various working conditions. The experimental system to validate the proposed method in identification the motor parameters has been built. The parameters of DC motor have been estimated based the actual motor speed response. Then, the proposed technique has been examined at different motor operation as starting, load change and speed change. Subsequently, the experimental results are compared with results obtained from the FFO. It is found that a good convergence between the actual and estimated results is appeared and the convergence rate is continuously decreased until highly acceptable value. Furthermore, a comparison between the results of FFO and those based recent optimization technique has been occurred. It can be assured that the FFO performs well and has good capability of recognizing the unknown DC motor parameters with fewer errors.

43

Authors: Widi Aribowo, Supari Muslim, Bambang Suprianto, Subuh Isnur Haryudo, Aditya Chandra Hermawan

A power system is a collection of equipment that has characteristics. Power system stability is maintaining how the system can return to its position. Power system stabilizer (PSS) is a device used to maintain the stability of the power system due to load changes. The archimedes optimization algorithm (AOA) is a metaheuristic method based on the force that occurs in the fluid due to a load. The method is based on the laws of physics. A neural network is a concept to duplicate the work function of the human brain. In this study, the archimedes optimization algorithm (AOA) will be proposed to improve the performance of the feed forward neural network (FFNN). This hybrid method is called AOA-NN. The hybrid method is used to improve the performance of power system stabilizers. To test the ability and effectiveness of the AOA-NN method, a comparison with the conventional PSS, feed-forward neural network (FFNN), Cascade-forward neural network (CFNN), Distributed time-delay neural network (DTDNN) and Sine Tree-Seed Algorithm – Feed-forward Neural network (STSA-NN) method are applied. From the research, it can be concluded that the method proposed by AOA-NN has the best ability. The AOA-NN method has the ability to reduce the overshoot speed with an average value of 85.43972% and the overshoot rotor angle with an average value of 38.9278%.

72

Authors: Yashaswini Doddamane Kenchappa, Kwadaki Karibasappa

Recently, the multimedia complexity is noticeably increased due to the growth of internet and multimedia technologies and it has become one of the prime research topics in multimedia content retrieval. In this research, a dual phase model is proposed to enhance the performance of content based image retrieval. Initially, normalization technique is employed to improve the visibility level of the images that are collected from Wang and COREL datasets. Next, color moments and Alex-Net convolutional neural network are used to extract the feature values from the normalized image. By combining low and high level features, the semantic space between the extracted feature values is less that helps in retrieving the relevant content images significantly. Finally, the Manhattan distance measure is applied to measure the distance between the query image and an image dataset to retrieve the relevant images. The experimental result showed that the proposed dual phase model achieved better performance in content based image retrieval by means of recall, precision and f-score. From the comparative analysis, the proposed dual phase model showed a maximum of 0.43 precision value and 0.06 recall value improvement in content based image retrieval compared to the existing works like Bi-layer system, spatial and frequency domain features with Euclidean distance, and local directional pattern with color histograms.

97

Authors: Somkiat Khwanpruk, Chalida U-tapao, Kankanit Khwanpruk, Laemthong Laokhongthavorn, Seksun Moryadee

The Bangkok–Nong Khai high-speed railway is the first high-speed line in Thailand and is currently under construction. The project is due to be completed by 2023, with ticket prices starting at 80 baht plus 1.8 baht per km which can be considered as a straight-line fare set with a fixed rate increase. Different from the current price policy, this paper examines the application of dynamic pricing to Thailand’s HSR. Dynamic Pricing Optimizer proposed in this paper is a unique system of dynamic pricing by considering the changes in the amount of service requirements and the time of purchasing tickets designed to solve the problem of pricing that is appropriate. First, The Dynamic Pricing Optimizer (DPO) system will integrate the user data from the latest trip or historical information. After that, there will be a forecast for users of high-speed trains based on historical data in order to create a demand function that is linear. After that, the price will be determined by the optimization method to find the most suitable price in order to maximize the revenue. Lastly, we compare both fare policies and found that The Dynamic Pricing policy can increase revenue up to 6.5%.

119

Authors: Aye Aye Aung, Zin Mar Win

Bone age assessment methods have been replaced from manual to automatic evaluation in medical field. Bone age is also needed for pediatricians and endocrinologists to compare chronological age for growth disorders and endocrinological problems. Recently, bone age estimation has gained remarkable ground in academia and medicine. Automated bone age assessment (BAA) methods are needed to improve the accuracy of the bone age because manual BAA methods are time-consuming and inefficient. Hand radiograph is the most widely applied to calculate the bone age of children. There are several challenges in bone age estimation such as the less of accuracy to segment region of interest (ROI), to identify boundaries and to classify bone age. The aim of the proposed system is to improve robustness and accuracy of the BAA system. The system applies Histogram of Oriented Gradient (HOG) features for middle finger and Hu moments features for carpal bones. Then, the system uses Support Vector Machines (SVM) by applying extracted features from middle finger and carpal bones for bone age classification. The proposed system uses the dataset from University of Southern California. Combination of HOG features from Epiphysis/Metaphysis region of interests (EMROIs) and Hu moments features from carpal region of interest (CROI) achieved better results than other methods. Running time of combining HOG and Hu moments features is faster than other methods and this method reduces feature dimensions than using HOG features from EMROIs and CROI. As the experimental results, the accuracy of the system achieves 89% of Asian (ASI) and African American (BLK) female radiographs and 86% of ASI and BLK male radiographs in age range from one year to ten years old. In the experiment, this method obtains 1.2717 years of Mean Absolute Error (MAE) compared with radiologist 1 and 1.1997 years of MAE compared with radiologist 2 in testing ASI and BLK females. Then, the system also obtains 1.3681 years of MAE compared with radiologist 1 and 1.3311 years of MAE compared with radiologist 2 in testing ASI and BLK males.

139

Authors: Dheyaa Jasim Kadhim

Massive multiple-input-multiple-output (MIMO) technology offers high spectral efficiency (SE) using simple schemes while it is also especially expected to offer high energy efficiency (EE). Then to capture promising gains expected from the 5G networks, this work proposes two improved versions of famous linear precoding schemes, i.e., MRT (maximal-ratio-transmission) and ZF (zero-forcing). To enhance and evaluate the energy efficiency performance in massive multi-user mimo (MU-MIMO) system, we considered here imperfection base station (BS) channel state information. Additionally, our proposed improved versions of linear precoding schemes are taking into account power circuit consumption and transmit power. Simulation results have shown that the improved ZF precoding scheme achieves higher EE as compared to improved MRT precoding scheme, while the results showed that improved MRT precoding outperforms ZF precoding within small BS number of M-antenna deployment comparatively to number of K terminals (from 20 to approximately 40 antennas). Previous results had been further shown that without circuit power consideration, the energy-efficiency was increasing without limit. In contrast to this belief in the literature, our model shows that energy efficiency is not unbounded when BS antennas increase very-large. We found that around 100 active antennas serving tenth of autonomous users simultaneously showing that energy-efficiency cannot be optimal at low signal to noise ratio (SNR) region, but in region where an appropriate interference-canceling processing, ZF precoding is extremely suitable over an interference-limited processing, MRT precoding. Finally, the results demonstrated and confirmed that an efficient-energy can be achieved by allowing so more and more users to access the channel simultaneously, thanks to massive MIMO that illustrates 5G promising gain expectations.

160

Authors: Irwan Budi Santoso, Yudhi Adrianto, Anggraini Dwi Sensusiati, Diah Puspito Wulandari, I Ketut Eddy Purnama

High performance in the epileptic electroencephalogram (EEG) signal classification is an important step in diagnosing epilepsy. Furthermore, this classification is carried out to determine whether the EEG signal from a person's examination results is categorized as an epileptic signal or not (healthy). Several automated techniques have been proposed to assist neurologists in classifying these signals. In general, these techniques have yielded a high average accuracy in classification, but the performance still needs to be improved. Therefore, we propose a convolutional neural network based on multi-segment of EEG signals to classify epileptic EEG signals. This method is built to overcome data limitations in the convolutional neural network training process and add the ensemble combination process. The multi-segment of EEG signal is formed by splitting the signal without overlapping each channel and converting it into the spectrogram image based on the short-time Fourier transform value. The spectrogram image is then used as input for the convolutional neural network in in-depth training and testing. The convolutional neural network model of the training results is used to classify each EEG signal segment on each test channel before entering the ensemble combination stage for the final classification. To evaluate the performance of our proposed method, we used the Bonn EEG dataset. The dataset consists of five EEG records labelled as A, B, C, D, and E. The experiments on several datasets (AB-C, AB-D, AB-E, AB-CD, AB-CDE, and AB-CD-E) which were arranged from the dataset showed that our proposed method (with segment) performs better than without segment. Our proposed method yielded the best average of classification accuracy which is 99.33%, 100%, 100%, 99.5%, 99.8%, and 99.4% for the AB-C, AB-D, AB-E, AB-CD, AB-CDE, and AB-CD-E. By these results, the proposed method can outperform several other methods on the same dataset.

189

Authors: Nuchanart Santalunai, Samran Santalunai, Piyaporn Meesawad, Chanchai Tongsopa, Samroeng Santalunai

This research focusses on the methods for expanding the working space to radiate the electromagnetic fields of 2.45 GHz within a dielectric load by using only one square patch radiator together with a plus-like electromagnetic band gap (EBG). One port of square patch generates a high transmission power to distribute the symmetrically electromagnetic field distribution. Electromagnetic band gap structure has a unique band gap characteristic where it can suppress the surface wave and continuously increase electric fields distribution on dielectric load under test with dimension of 300×300 mm2 when the thickness dimension is optimized. A dielectric heating applicator using microstrip patch with EBG structure has a low-loss dielectric because of the structure of system which has no gap between applicator surface and dielectric load. This paper provides a comparison of the simulation of electric distribution in microwave chamber when the applicators are copper plate, square patch microstrip and square patch generator surrounded by EBG. In the simulation results, an electric distribution in the dielectric load was compared when the dielectric constant was 4.7. The maximum percentage of electric fields distribution that was evenly distributed in dielectric load was about 99.97% at middle surface with a thickness of 5 mm by using square patch generator surrounded by EBG as the applicator. In addition, the simulation results were confirmed with the results of the dielectric heating measurement of the dielectric load as rice which is a popular grain in the commercial application of dielectric industrial drying.

210

Authors: Chandrashekhar Ghuge, Vudatha Chandra Prakash, Sachin Ruikar

The popularity of video recordings either on mobile devices or video surveillance has contributed to the demand for video data applications. As a result, the management of video has become significant in the object retrieval process. However, the video object retrieval is considered as a major issue in the video objects management. This paper proposes an integrated optimization-based classifier for video object retrieval. Here, the input video is subjected to optical flow estimation using a spatio-temporal feature descriptor, named Histograms of Optical Flow Orientation and Magnitude (HOFM) for extracting the events and establishing the HOFM indexed database using the detected objects. Hence, object tracking is a major step, which detects and tracks the video objects using a hybrid model named Nearest Search Algorithm-Support Vector Regression (NSA-SVR). For the video retrieval, the training of Naive Bayes (NB) classifier is performed using the proposed Lion-Salp Swarm Algorithm (LSSA) on the indexed database of the tracked objects. Then, the recognized events interrelated to the query are subjected to the Naive Bayes classifier to retrieve the required video. The performance of the proposed integrated LSSA-Naive Bayes Classifier is found to be better than the existing methods with maximal precision of 82.985 percent, recall of 87.451 percent, and F-measure of 87.847 percent.

242

Authors: Dhari Y. Mahmood, Mohammed Kdair Abd, Kanaan A. Jalal

Wind Energy Output Prediction (WEOP) has an important impact on the integration of wind energy systems in the power grid, the management, power systems dispatching, safe and stable operation. Short-term probabilistic WEOP is a perfect choice to increase the stability of the power grid. Still, it has a high error because of uncertainty factors such as wind speed, and very important to find out a method to increase the fineness of predicting. Therefore, in this paper, the main contribution is to design an intelligent model capable of wind energy generation prediction. This aim is achieved by adopting a new Developed Particle Swarm optimization algorithm based on proposes a new inertia weight called DPSO and Back-Propagation Neural Network (DPSO-BPNN). The proposed models were compared and verified through MATLAB software test to prove the superiority over other models. The obtained results of the proposed model show a fast convergence and high prediction accuracy. Also, the error rate was improved compared with the published research by up to 98%.

267

Authors: Darmawan Napitupulu, Rudi Yacub, Aditya Halim Perdana Kusuma Putra

Telehealth has become an emerging technology since they offer remote healthcare services for infected patients or those who are worried about being infected. Furthermore, Telehealth can provide routine care without the risk of contracting Covid-19. Although Telehealth offers many potential benefits, the challenge was to encourage people to use them. Research related to Telehealth adoption in Indonesia is still limited, so there is still a lack of understanding regarding their behavior intentions in using Telehealth. One of the most widely used theories to explain Telehealth adoption is the Unified Theory of Acceptance and Use of Technology (UTAUT). However, doctors' opinion and computer anxiety variables have never been applied in the UTAUT model to analyze Telehealth use behavior, especially for cases in Indonesia during the Covid-19 outbreak. In addition, many researchers point out the importance of both variables in the healthcare context increasing the need for examining this feature. This paper discusses it for the first time. The survey was conducted by distributing online questionnaires via Google Docs to 118 Indonesian respondents who have used Telehealth. The results show that performance expectancy, effort expectancy, and facilitating conditions significantly affect behavior intention to use Telehealth. Social Influence is not significantly associated with behavioral intention since early adopters are considered reluctant to other views. Moreover, performance expectancy is significantly affected by doctor’s opinions and effort expectancy is strongly influenced by computer anxiety.

294

Authors: Hayder Naser Khraibet AL-Behadili

Internet usage is increasing rapidly worldwide, allowing numerous connected computer objects or devices to run and communicate with mass digital information. As Internet usage becomes pervasive, attacks against them are also rising aiming to penetrate the target network and remain undiscovered. Therefore, analyzing the behavior of Internet traffic manually is not possible due to its complexity and the large number of user activity. Incoming and outgoing Internet traffic are controlled using a firewall through an automated Internet security system using a predefined set of rules. Machine learning algorithms are used for Repeated Stemanalysis of the activities on firewall devices and to control traffic on the basis of the results. However, the output models (i.e., classification models) lack explanatory power insight into the relative influence of the main factors in the classification and thus have low accuracy. In this study, a decision tree classification algorithm with a tree-structured model is used for firewall activity analysis, which produces high classification accuracy. Empirical results on firewall access with different actions were compared against six benchmark classification algorithms, namely, SVM, OneR, ANN, Multi class classifier, PSO and ZeroR, in five popular evaluation metrics. The experimental results show that the performance of the proposed classifier in all evaluation metrics is higher than the state-of-the-art classification algorithms, such as SVM, ANN, Multi class classifier, PSO, and the most related classification algorithms that provide comprehensible models (i.e., OneR and ZeroR). The proposed classifier offers interpretation ability by presenting the classification model into a tree representation, which is a further advantage.

319

Authors: Mostafa Gamal, Ahmed El-Sawy, Ahmed H. AbuEl-Atta

The daily massive flow of information requires automated summarization methods to extract the most important information. Manual summarization of large text documents is very complicated and time-intensive for human beings. Numerous methods rate all of them decadently based on the sentence scoring that labels ratings for input sentences. The higher-rated sentences are used as a part of the summary. In an extractive-based automated text summary, locating the relevant sentences is an essential problem for the researchers. Therefore, to deal with such problems, evolutionary algorithms are applied as a solution. This paper presents a hybrid approach (CSOGA) based on the effectiveness and convergence to the solution of a (CSO) chicken swarm optimization and a (GA) genetic algorithm for text summarization to ensure the optimal solution. The evaluations of the proposed algorithms are done on the standard dataset from CNN / Daily Mail and are measured by the Recall-Oriented Understudy for Gisting Evaluation (ROUGE). The performance of the proposed method is then compared with other methods. The results show that the new approach hybrid (CSOGA) has the best performance on text summarization quality. The proposed method was capable of generating a better accuracy than other algorithms on the ROUGE-1, ROUGE-2 and ROUGE-L. The highest increase in the accuracy of the proposed method was in ROUGE-1 with a rise of 4.4%, ROUGE-2 with a rise of 12.01%, and ROUGE-L with a rise of 9.8% comparing with the highest accuracy of the other extractive models.

346

Authors: Arfinda Setiyoutami, Diana Purwitasari, Wiwik Anggraeni, Eko Mulyanto Yuniarno, Mauridhi Hery Purnomo

Since local human movements can influence dengue spread, a network-based prediction model considers the dynamic relation between dengue case incidences and their location over time. Some approaches often generated the networks in a certain period with a single spanning time until several months or in one year, called static networks. However, one annual-based dengue spread model could have different simulations depending on the selected months to show different dynamicity. Other approaches do not involve any networks for generating the spread models but employ them for validating the models with simulations. Considering the evolution of dengue circumstances that could quickly change between periods, we proposed a Dengue Spread Dynamic Network (DSDN) model with some timespan and location boundaries variants. DSDN includes five network models with nodes representing localities and links showing dengue spread which varied every day depending on infections presence and environmental conditions in a certain period. With our proposed method, daily dengue spread from one location to another can be predicted based on the location-based incidence historical data as outbreaks prevention initiative. We also analyzed how dengue spreads differently in outbreak and non-outbreak periods using Dynamic Network Link Prediction (DNLP) method. From our experiments result, Neighbor Network which modeled that dengue only spreads between neighboring localities produced an accuracy of 92.54% for the entire period. When applied only to outbreak data, there was a performance increase of 3.39 points, which suggested that link prediction performs better when dengue is rapidly spreading. In addition to that, our experiments concluded that dengue potentially spreads to a location with no current infections if local incidence often occurred in the past.

374

Authors: Giva Andriana Mutiara, Othman Mohd, Nanna Suryana, Ahmad Naim Che Pee

An effort to minimize illegal logging by utilizing advanced technology such as Wireless Sensor Network (WSN) are highly required. WSN is a communication solution between remote sensors located in isolated wide forest areas. However, the WSN challenge in a remote and wide-area has the energy resources limitation. This research proposed a weights-based energy-efficient WSN data communication protocol with firefly synchronization, namely Firefly Synchronization Multi-hop - Geographical Energy-Aware Routing (FSM-GEAR) Protocol. This protocol aimed to increase the network lifetime of each node by shortening the transmission distance. Firefly synchronization between sensors is conducted to reduce the data transmission waiting time. A weights-based approach in cluster head (CH) selection phases and energy consumption analysis are formulated according to the comparison between nearest distance to Gateway and the maximum residual energy of each node based on received signal strength indicator (RSSI). Weights-based FSM-GEAR performance analysis is carried out based on the parameters of network lifetime, residual energy, and throughput with the Modified-GEAR and enhanced Modified-GEAR protocols as a comparison. The simulation result stated that FSM-GEAR protocol showed better network lifetime, residual energy, and throughput than both protocols.

403

Authors: Mohammed Najm Abdullah, Khulood Eskander Dagher

A model avionics system is a type of electronic system and equipment specifically designed for use in airborne computers. This article proposes an improvement of the output performance of Unmanned Aerial Vehicle (UAV) system based on multi-PID-controller with on-line swarm optimization algorithm. The main goal of this research is to design six PID controllers to control the high nonlinear UAV quadcopter system by using an on-line Particle Swarm Optimization (PSO) algorithm that uses to learn and tune the eighteen control gain parameters based on multi-objective function. The task of the proposed on-line multi-PID-PSO feedback path-tracking controller is to obtain precisely and quickly the robust rotation speed of rotors which are used to control the attitude and altitude of the UAV quadcopter system. The results of the proposed strategy show that the on-line multi-PID-PSO controllers are accurate in terms of the UAV quadcopter take-off and follows the Spiral and Cyclone desired paths quickly through fast obtaining of the multi-controller’s parameters and smooth rotation speeds actions generating for UAV system with a minimum number of multi-cost function evolutions that minimized the tracking translation location error around ±5 cm and the overshoot of altitude did not exceed 1cm. Finally, we confirm the effectiveness of the simulation results of the proposed controller through comparison of other types of controller simulation results.

425

Authors: Kelly R. Sungkono, Riyanarto Sarno, Fransisca D. Amadea, Ariqoh F. Irbah

Process discovery is a study to form a process model based on an event log. The output of process discovery is a process model that contains linked activities based on performed business processes. There are several kinds of relationships and additional invisible tasks in forming a process model. The current invisible tasks, called invisible prime tasks, are divided into three conditions, i.e. skip, redo and switch. However, the need for invisible task is not only in those conditions but also in a situation, which is hybrid choice-parallel relationships. Hybrid choice-parallel relationships occur when there are choice relationships, namely XOR, pile up with parallel relationships for all branching activities (AND) or parallel relationships for several branching activities (OR). This research proposes graph-based Invisible-Task Prime and Non-Prime (ITPNP) algorithm that mines invisible prime tasks and invisible non-prime task in the cased of hybrid parallel relationships. Graph-based ITPNP algorithm creates rules for detecting invisible task needs based on relationships occurring in the graph model. The evaluation is implemented with seven processes and four comparison methods, i.e. Alpha#, Alpha Mining, Fodina, and Graph-based Invisible Task. Among all algorithms, graph-based ITPNP obtains one of fitness for all event logs and one of precision in four event logs. On the other hand, Fodina algorithm has 0.6 precision which is lower than graph-based ITPNP in the event logs requiring invisible non-prime tasks of hybrid XOR-OR. Besides, Graph-based Invisible Task has low fitness, i.e. 0.67 in Hazardous Waste Handling processes that depict choice relationships with additional invisible tasks for hybrid XOR-AND relationships.

447

Authors: Md Shoaibuddin Madni, C. Vijaya

Currently, hand gesture recognition becomes a promising system for the digital entertainment field due to the recent development of sensor innovations and machine learning. However, hand gesture recognition is a challenging task in most of the existing models, due to background clutter, motion blur, illumination variations, and occlusions. In this research, a dynamic hand gesture recognition system is proposed to improve the performance of hand gesture recognition. At first, the normalization method is used to enhance the visibility level of the gesture images which are collected from the Indian sign language dataset. Then, semi vectorial multilevel segmentation method is employed to segment the exact gesture regions from the normalized images. Further, improved reliefF algorithm and K-nearest neighbour classifiers are used to select the optimal features and to classify the 16 gesture classes or symbols. In the experimental phase, the proposed improved reliefF- K-nearest neighbour model performance is analysed in light of Matthew’s correlation coefficient, accuracy, sensitivity, specificity, and f-score. For overall 16 distinct alphabets; A, B, D, E, F, G, H, K, P, R, T, U, W, X, Y, Z, the proposed improved reliefF-K-nearest neighbour model achieves average accuracy of 98.95%. Hence, the proposed improved reliefF-K-nearest neighbour model showed maximum of 7.97% and minimum of 0.15% improvement in recognition accuracy compared to k means-neural network model, and discriminant correlation analysis based unimodal feature-level fusion model.

468

Authors: IBG Manuaba, Muhammad Abdillah, Ramon Zamora, Herlambang Setiadi

A disturbance such as a small load change in the electrical power grid triggers the system to oscillate and may lead to performance degradation of the control system in the generation source unit. To overcome this problem, a power system stabilizer (PSS) is added as a supplementary control in the power system grid to damp the oscillation caused by the disturbance. This paper proposes a kernel extreme learning machine (K-ELM) method to adjust the parameters of PSS for a wide range of operating conditions. The proposed control scheme in this research work is PSS based on K-ELM called KELM-PSS. To examine the robustness of KELM-PSS, a single machine infinite bus (SMIB) is utilized as a test system. The simulation results showed that the KELM-PSS provided a satisfactory result in both the training and testing phases. In the training process, KELM-PSS has a smaller mean square error (MSE) value, mean absolute error (MAE) value, and sum square error (SSE) value in terms of accuracy criterion compared to PSS based on machine learning such as extreme learning machine (ELM), support vector machine (SVM), and least square support vector machine (LS-SVM). Also, in terms of computation time, KELM-PSS has faster CPU time than other machine learning methods to obtain the parameters of PSS. The parameters of K-ELM including the regularization coefficient Cr and kernel parameter σ obtained from the training phase are utilized in the testing process. In a testing phase, the K-ELM approach will obtain the appropriate parameters of PSS (Kstab, T1, and T2) when there are changes of generator real power output P, reactive power output Q, terminal voltage Vt, and equivalent reactance Xe. In this stage, KELM-PSS could decrease the overshoot and compress the settling time better than other approaches utilized in this study. Moreover, KELM-PSS has the best performance index based on integral time absolute error (ITAE) compared to other techniques utilized in this research work.

493

Authors: Musaab Riyadh, Basim Jamil Ali, Dina Riadh Alshibani

The rapid growth of computer networks has led to massive flow of data every second. Some of these data flow is a malicious activity and cannot detect by anti-malware and firewalls. Therefore, an intrusion detection system is an urgent issue aims to distinguish between non relevant and relevant data in order to maintain data availability and integrity. Due to this, a hybrid intrusion detection system is proposed in this study based on machine learning techniques to tackle various challenging issues in data set such as mixed type data, incomplete, and uncertain data. The proposed system has achieved its objectives by supporting: Firstly, the density based clustering approach due to its robustness to noise removal. Secondly, the K means and K-nearest neighbour algorithms to transform n dimensional data into one dimensional data in order to deal with mixed type data and minimize the running time. Finally, a special type of dissimilarity measure has been supported to tackle the problem of missing data. The experimental results illustrate that the proposed classifier has better or similar classification accuracy (92.9%) as compared with support vector machine (91.5%), CANN(92.2%), Random forest(93.3%), and EIDS-ACC- OD(91.9%) classifiers in KDD Cup99 data set. However, the proposed classifier has the best performance (92.2%) and (91.2%) when randomly removing 5% and 10% of the KDD Cup99 data set in spite of decreasing the overall accuracy classification for all the classifiers.

516

Authors: Vincentius Raki Mahindara, David Felipe Celeita Rodriguez, Margo Pujiantara, Ardyono Priyadi, Eduard Muljadi, Mauridhi Hery Purnomo

Integration of inverter-based resources (IBR) as a distributed generator (DG) in a microgrid has been a trend. Considering that DG integration transforms the short circuit behavior in terms of magnitude and direction in the system, the conventional protection scheme becomes no longer relevant. The major concern of the traditional protection scheme in a microgrid is the relay operating time to clear the fault in a coordinated manner. This paper presents a novel voltage-current model for directional over-current relay (DOCR) that gives the relay capability to reduce the overall relay time operation. The state-of-the-art of the proposed relay architecture relies on the adaptive pickup current threshold, which varies with the voltage magnitude, to anticipate the small fault current caused by the IBR. The improvement of relay sensitivity offers faster tripping time in a coordination scheme. The proposed method is assessed in the IEEE 13 bus distribution system with the integration of DG. The outcome of this study indicates that the total relay time operation using the proposed method is 2.363s. Compared to the conventional method based on the standard equation, the proposed technique significantly reduces the total relay operating time up to 47.67%.

537

Authors: Tanongsak Kongsin, Sakon Klongboonjit

This study aims to introduce a new method for modular design, which is called Connection Correlation Method (CCM), to clustering machine parts with inter-part connection conditions to classify independent modules for the Industrial Printer Powertrain set with 105 components. With the CCM technique, module members would be appropriately assigned for each module so that machine modules are independent of their function. The CCM technique should first calculate the distance coefficient matrix with surface contact conditions of bonded, rough, separation, and frictionless for machine components. Finally, this distance coefficient matrix is used to generate the machine dendrogram with the dependency coefficient between 2.4945 and 3.5593. At the dependency coefficient of 2.7913, the 105 components of the Industrial Printer Powertrain set are clustered into 7 modules: Module 1 with 75 interconnection surfaces and 8 parts, Module 2 with 136 interconnection surfaces and 14 parts, Module 3 with 176 interconnection surfaces and 17 parts, Module 4 with 61 interconnection surfaces and 8 parts, Module 5 with 200 interconnection surfaces and 21 parts, Module 6 with 209 interconnection surfaces and 19 parts, and Module 7 with 18 interconnection surfaces and 18 parts. The results show that the CCM technique can apply to design a modular machine like DSM technique, and multitudinous connectivity factors can also be analysed together with general factors.

555

Authors: Girish Nagaraj, Nandini Channegowda

In recent decades, frame duplication is a common inter frame tampering operation in the digital videos. To find the duplicate frames with better computational time, multi scale local oriented feature descriptors are proposed in this paper. Initially, histogram equalization is used to improve the visual quality of the images or videos, which are collected from surrey university library for forensic analysis dataset. Then, feature extraction is accomplished utilizing binary robust invariant scalable keypoint, speeded up robust features, and maximally stable extremal regions to extract feature vectors or key points from the enhanced images. After identifying the keypoints, matched keypoints are evaluated by hamming distance and k-means clustering algorithm from the source and moving video frames. The multi scale local oriented feature descriptors with two step feature matching significantly decreases the computational time and effectively determines forged and non-forged frames in the video sequences. Simulation results showed that the proposed model achieved better performance in passive video forgery detection in terms of accuracy, sensitivity, f-score and specificity. Compared to the existing approaches like spatio temporal context learning, inter-frame forgery detection algorithm and adaptive parameter-based visual background extractor algorithm, the proposed model obtained better detection accuracy of 95.10%, and average detection time of 1.04 seconds per frame.

580

Authors: Saritha Ibakkanavar Guddappa, Rajeshwari Mahabhaleshwar Hegde

Cyber-Physical Systems (CPS) is an emerging technology of intelligent systems with the united computational performance and physical capacities. Due to the growing system complexity and system directness the CPS is affected by the malicious attacks which lead to improve the CPS performances with an efficient data transmission method. In this paper , CPS making use of the Orthogonal Frequency-Division Multiplexing (OFDM) to achieve the high data rate requirement of modern communications. The Priority Aware Frequency Domain Polling (PAFDP) protocol is proposed to detect the priority based node which identified by using the throughput and data rate from the group of nodes in the network and also the priority based nodes are identified as malicious nodes which are avoided to minimize the packet loss in the existing network. The performances of the PAFDP-OFDM-CPS are analysed in terms of various QoS parameters such as delay , energy , Packet delivery ratio , packet loss ratio , through put and overhead so that the proposed method is assessed by three existing methods such as Adaptive Duty Cycle Control Based Opportunistic Routing (ADCCOR) protocol , Trust Aware Routing Framework (TARF), Trust Aware Wireless Routing Protocol (TAWRP) and It was found that the performance of the PAFDP-OFDM-CPS is improved much compared to the above said three protocols.

1

Authors: Naim Ahmad

Cloud computing is a matured platform to host the information and communication technology (ICT) infrastructure and services. All the top leading companies have made investments in cloud infrastructure including US Department of Defence’s contract of USD 10 Billion to Microsoft for cloud migration. Cloud computing helps in conserving the resources and improving the performance of ICT services, hence making the computing more sustainable. Migration to cloud needs careful planning, strong business case, credible migration strategy and robust migration frameworks. This research has reviewed the area of cloud computing migration to identify important factors. The author keywords of 299 research articles downloaded from the web of science database have been analyzed using co-word analysis and resulted in 10 most significant factors in the form of cloud migration tuple. This study applies the interpretive structural modeling (ISM) technique to study the interrelationship on these factors. Application of ISM followed by Matriced’ Impacts Croise’s Multiplication Appliquée a UN Classement (MICMAC) analysis has categorized these factors into independent and dependent sets. The resultant model demonstrates that organizations need to incorporate the dimensions of service orientation, virtualization, multi utilization, dynamic allocation and mobile support into their data, application and security stacks. Migration of these stacks with sound strategy, policy and frameworks will lead to post-migration performance improvements in a synergistic fashion. The adoption of cloud migration tuple with understanding of interrelationships will help organizations make strategies and plan to migrate to cloud in an efficient manner.

22

Authors: Aravind AR, Rekha Chakravarthi

Wireless sensor network (WSN) employs the mobile sink to collect the information from the sensor nodes that are deployed in the network environment in such a way to reduce the energy hole issues. However, the placement of mobile sinks optimally in the network without degrading network lifetime result in a great challenge. Thus, the adjacency-based cell score factor is adopted in this research in order to place the mobile sink optimally in WSN. Here, the network is transformed into the cell using the Voronoi transformation. For each uniform-sized cell, the cluster head (CH) is selected using Sparse-Fuzzy C-means clustering. The optimal placement of the mobile sink is carried out using the adjacency-based cell score, which effectively places the mobile sink using the constrained factors, such as energy, Euclidean distance, and fairness. The proposed adjacency-based cell score obtained better performance with a minimal distance of 22.4615m, maximal energy, fairness, network lifetime, and throughput of 97.2182, 92.6064, 145, and 95.8852, respectively, while considering 400 nodes.

43

Authors: Pradeep Kumar, Anita Kanavalli

Clustering plays a major role in the data mining application, because it divides and groups the data effectively. In the pattern analysis, two major challenges occur in real-life applications that includes handling the categorical data and the availability of correctly labeled data. According to the characteristics of homogeneity, the clustering techniques are designed to group the unlabeled data. Some important issues such as high memory utilization, time consumption, overhead, computation complexity and less effective results are present in various existing algorithms of numerical data. Therefore, the research study implemented clustering techniques based on the similarity of categorical data. Simultaneously, the attributes of inter and intra-clusters’ similarities are identified, and then the performance of proposed method is improved by integrating those similarities. The noises are also removed by performing the pre-processing techniques, so the similarity between noise-free elements are estimated. Once these similarities are identified, the insignificant attributes are removed and the relevant attributes are chosen from the pre-processed elements. The overhead is reduced by developing the Similarity-based K-means Clustering (SKC) approach for clustering the attributes that depends on divergence distance. The efficiency of SKC is tested in the experimental analysis by means of precision, f-measure, accuracy, error rate of clustering and recall. The results state that the developed study achieved 98.45% accuracy for the publicly available dataset when comparing with the existing techniques: variations of Particle Swarm Optimization (PSO) and semi-supervised clustering system.

67

Authors: Sammaiah Seelothu, Konakanchi Venugopal Rao

The automatic Face Expression Recognition is very important in computer vision since it has widespread applications in real time. However, the main issue is the design of an efficient descriptor that describes the appearance changes on the face. Towards such design, in this paper we have introduced a novel binary compact coding called as Edge Adaptive Local Directional Binary Pattern (EALDBP) that encodes the texture features of facial image. The proposed EALDBP is resilient to noise and illumination variations as well as it encodes the directionality of expressions. The directionality is discovered by the computation of edge responses through different compass masks. Further to encode intensity variations, we have considered the relationship between gradient pixels in the local neighborhood. Finally, a compact face descriptor is constructed with the help of Histograms computation over the encoded face image. To evaluate the performance, we tested out approach on two datasets; they are CK+ and JAFFE. The average recognition accuracy of these two datasets is observed as 90.3898% and 93.5922% for JAFFE and CK+ respectively. These accuracies demonstrate that the proposed approach outperforms the several state-of-art methods.

95

Authors: Srinivasa Suresh Sikhakolli, Asha Kiran Sikhakolli

A program design can be understood by the developers through detecting the design patterns in object-oriented programming source code. The main advantage of the design pattern detection includes maintainability, understand-ability and reusability of object-oriented programs. An object-oriented program is used as the input to the design pattern detection techniques, which judges the candidate roles of the patterns. The existing pattern detection techniques use the machine learning (ML) techniques, namely Support Vector Machine (SVM) and Decision Tree with all feature metrics (more than 70 metrics) to identify patterns, which increases the computation time. To minimize the issues of using all feature metrics, an effective feature selection technique is used in the research study. The most important relevant features are selected by proposed Bacterial Foraging Algorithm (BFA) and given as input to ensemble classifiers namely SVM, Decision Tree and Random Forest (RF) classifier to design pattern detection. By using BFA technique, the method used only five metrics for the identification of patterns, where existing techniques uses 70 to 80 metrics for same pattern detection. The simulations are conducted to test the effectiveness of BFA with ensemble classifiers on the Python Software platform in terms of average accuracy and precision. The results stated that BFA-RF achieved 88.57% of average accuracy, where BFA-SVM technique achieved 81.43% of average accuracy, which shows the RF achieved better results among other ensemble classifiers.

121

Authors: Shimaa Ouf, Noha Hamza

Objectives: this paper aimed to first, exploring the relationship between multiple physical measurements. Second, predicting the treatment course of hospitalized patients with COVID-19 disease from physical measurements. Third, investigating the primary symptoms and the average duration of each symptom’s disappearance. Fourth, provide the physicians with the prediction model to help them determine the best combination of drugs for Covid-19 patients’.　Methods: this paper first, apply correlation analysis on the dataset to identify the relationships between the dataset attributes and selecting model features more efficiently to improve the accuracy results. Second, implements seven machine learning algorithms with four cross-validation techniques to predict the most appropriate treatment course for COVID-19 hospitalized patients. Two treatment courses were identified; course-LR was patients treated with Lopinavir-ritonavir and course LR + AR were patients treated with Lopinavir-ritonavir combined with arbidol for antiviral treatment. Results: by applying correlation analysis between dataset attributes, we found that there is no relationship between the presence of chronic diseases or the patient’ age and the Covid-19 clinical classification. The prediction model results show that 10 fold cross-validation with Naïve Bayes and neural network achieving the highest accuracy of 85.71%.　Conclusion: this paper has exploited correlation analysis and machine learning based approaches to identify relevant attributes in the COVID-19 patients’ dataset and predicting the most appropriate treatment course.

147

Authors: Manohar Manur, Alok Kumar Pani, Pankaj Kumar

Heart disease is a serious disease that causes sudden death among 80% of the people around the world. The traditional models performed predictive analytics using machine learning techniques to make a better decision. For better decision making in heart disease prediction, the big data analysis shows the great opportunities to predict the future health status from health parameters and provide best outcomes. However, the traditional decision making models had traffic data or contained noise and uncertainty was unpredictable as the data ambiguity emerged. In order to overcome such an issue, the big data is used to ensure the medical service which is mostly needed in a timely manner and for accurate diagnosis. The pre-processing of the medical data acquired from Cleveland heart disease UCI datasets has a vast number of attributes which helps to predict the heart disease. The data are contaminated with the noise and some of the data are missing, so the pre-processing using Min max Normalization is performed to remove contaminated noise acquired in the data which is taken from the UCI repository dataset. The proposed Fuzzy Deep Convolution Network (FDCN) permits the input features for fuzzification process that uses transformed features. The fuzzification process eliminates the redundant or irrelevant fuzzified features and overcomes the system complexity problems. The proposed FDCN obtains accuracy of 95.56 % and 92 % of F-score shows better results when compared with the existing KNN-DT, Naive Bayes, and Random Forest algorithms.

167

Authors: Easpin Brumancia, Justin Samuel

Clustering is the most frequently used physical architecture in data fusion, causing a problem as it has random distribution within local clusters, we propose a density peak based clustering algorithm utilizing a hierarchical approach. As a proficient method to joint decision-making of multi sensors, data fusion contains several benefits in data mining. In this work, we use the Salp Swarm Optimizer (SSO) algorithm based Hierarchical Density Peak Clustering (DPC) method to cluster the multiple sensor nodes as per the data similitude for the process of data fusion. Here, the higher density, distance and the cut off values are determined by using the SSO algorithm and then the chosen optimal fixed values were used to further process. Hierarchical Density Peak Clustering estimates the density and distance of each point. It is validated experimentally and compared it with the datasets and the error prediction with the test case numbers. The experimental results of the method are 96% accuracy determined from different datasets.

190

Authors: Wathiq Laftah Al-Yaseen, Ammar Jehad, Qusay Abdullah Abed, Ali Kadhum Idrees

Breast cancer has been recently considered as one of the broadly spread diseases that causes death among women. Early disease diagnosis is a critical aim in building the treatment policies and is extremely related to safety of patient. Therefore, there is a necessity for computer aided detection (CAD) in order to provide accurate and rapid diagnosis for breast cancer. Recently, many classification models utilizing machine learning approaches have been adopted and modified to diagnose breast cancer disease. Moreover, the performance of each model depends on different compositions such as the number and type of data features and the parameters of model. In order to enhance the performance of classification model, this research proposes a model using modified K-means algorithm to create a new training dataset of breast cancer which can highly improve the performance of support vector machine model. A modified K-means algorithm is also proposed to build a high quality training dataset that contributes significantly to reduce the training time of classifiers, and improve the performance of classifier. The proposed model handles the noise and irregularity in data and produce high quality dataset which represents all the cases of disease. The two recognized datasets Wisconsin Breast Cancer (WBC) and Wisconsin Diagnostic Breast Cancer (WDBC) have been used to examine and appraise the performance of the proposed model. The experimental results show that the proposed model has a significant performance compared to other previous works and with accuracy level of 98.067%, sensitivity of 100%, specificity of 94.811%, precision of 97.011% and finally with area under the curve related to the receiver operating characteristic of 97.406%.

211

Authors: Wayan Firdaus Mahmudy, Mohammad Zoqi Sarwani, Asyrofa Rahmi, Agus Wahyu Widodo

A supply chain process in manufacturing industries involves a distribution process that ensures finished goods distribute to their customers properly. The process may involve several parties such as manufacturing plans, distributions centres, and retailers. To maximize profit, the companies need to optimize the distribution process by minimizing costs. The multi-stage distribution problem contains several constrains so obtaining optimum solution using exact methods may require excessive processing time. Genetic algorithm is proposed the solve the complex problem. The genetic algorithm is improved by modifying chromosome representation and related genetic reproduction operators. The new chromosome is designed to address different number of distribution stages in manufacturing industries. Thus, the proposed approach could be applied for various distribution problems. Several genetic operators are tested to obtain the most suitable reproduction operator for the multi-stage distribution process. The numerical experiments prove that the improved genetic algorithm is suitable for the optimization of the multi-stage distribution process and produce better result compare to those achieved by original genetic algorithm, simulated annealing, and random search as baseline method. The improved genetic algorithm produces solution with cost of Rp 20,167.8, lower than cost of Rp 21,860.4, Rp 23,354.30 and Rp 34,328.0 obtained by original genetic algorithm, simulated annealing, and random search, respectively.

236

Authors: Michael Moses Thiruthuvanathan, Balachandran Krishnan, Madhavi Rangaswamy

Online teaching and learning has recently turned out to be the order of the day, where majority of the learners undergo courses and trainings over the new environment. Learning through these platforms have created a requirement to understand if the learner is interested or not. Detecting engagement of the learners have sought increased attention to create learner centric models that can enhance the teaching and learning experience. The learner will over a period of time in the platform, tend to expose various emotions like engaged, bored, frustrated, confused, angry and other cues that can be classified as engaged or disengaged. This paper proposes in creating a Convolutional Neural Network (CNN) and enabling it with residual connections that can enhance the learning rate of the network and improve the classification on three Indian datasets that predominantly work on classroom engagement models. The proposed network performs well due to introduction of Residual learning that carries additional learning from the previous batch of layers into the next batch, Optimized Hyper Parametric (OHP) setting, increased dimensions of images for higher data abstraction and reduction of vanishing gradient problems resulting in managing overfitting issues. The Residual network introduced, consists of a shallow depth of 50 layers which has significantly produced an accuracy of 91.3% on ISED & iSAFE data while it achieves a 93.4% accuracy on the Daisee dataset. The average accuracy achieved by the classification network is 0.825 according to Cohens Kappa measure.

262

Authors: Shreela Dash, Madhabananda Das, Dayal Kumar Behera

The proposed work suggests a high capacity steganography technique using dual steganography approach. It utilizes a Multi pass encryption system to scramble the mystery message before hiding. The scrambled mystery message is divided into two parts and concealed utilizing two duplicates of the same cover image. One copy of the image uses a modified Quotient value differencing approach to hide 1st part of secret message and another copy uses difference expansion method to embed the 2nd part of secret messages. It provides more secrecy because the suggested work uses the benefits of both encryption and steganography. The investigation shows that the two stego-pictures provide high embedding capability (EC) of 1848153 bits with Peak signal to noise ratio PSNR1 of 47.2 dB and PSNR2 of 51.1 dB. Further, examination with other existing methodologies shows that the suggested methodology is proven to be better in Structural similarity Index (SSIM), EC and PSNR.

285

Authors: Evy Hendriarianti, Aryuanto Soetedjo

This paper presents the Internet of things (IoT) technology for real-time monitoring of wastewater phytoremediation. Phytoremediation is a technique to remove pollutants from the wastewater using the plants. A conventional method to monitor the phytoremediation performance is by taking the samples of the contaminants from a site and measuring them at the laboratory. This method needs many works for data preparation and analysis. A recent development on the IoT technology may eliminate such tasks by a real-time monitoring system. In the proposed real-time monitoring system, several phytoremediation models are implemented on embedded hardware and connected to the Thingspeak IoT platform. The proposed system aims to provide a real-time monitoring system to better model the phytoremediation by examining the monitoring data time interval and fitting techniques. From the experiments, the proposed monitoring system achieves a data transfer reliability of 81.4% when the period of the data transmission is one minute, which is suitable for one-day interval real-time monitoring system. The proposed monitoring system can build a phytoremediation model using a polynomial fitting with a higher fit than the existing methods using an exponential fitting. Further, the proposed method promises a better solution in terms of the best model, the low cost, and the acceptable accuracy.

306

Authors: Shaaban M. Shaaban

Groundwater (GW) quality evaluation includes a variety of biological, chemical and physical parameters. The fundamental problem with water quality assessment is the difficulty with which a large number of parameters are evaluated. If all criteria have been used to evaluate the quality of GW, then computational difficulty will certainly increase. In this paper, a new hybrid three-stage assessment approach based on Feature Extraction Algorithm (FEA), standard deviation (SD) and Complex Proportional Assessment Method (COPRAS) was proposed. In the first stage the redundant criteria for GW quality assessment is removed using FEA. Secondly, the weights of the reduct parameters are evaluated based on SD. Finally, GW sites are ranked using (COPRAS). Sixteen GW samples were gathered from several GW wells. The collected samples were investigated for 12 various physicochemical water quality criteria to evaluate GW quality. The results reveal that sulfates (SO4), nitrate (NO3), Fluorides (F), sodium (Na), and Escherichia coli (E. coli) are the main parameters for GW quality assessment. Furthermore, the optimal concentrations of physicochemical parameters: (SO4), (NO3), (F), (Na), and (E. coli) are 18.9(mg/L), 8.18(mg/L), 0.222(mg/L), 21(mg/L), 1.9(MPN/100mL), respectively, with 40 WQI.The suggested approach is compared to three MCDM methods to validate the performance of the proposed methodology. The assessment results gained by the FEA combined with COPRAS and SD significantly minimize computational difficulty, reasonable and accurate. The approach presented in this study improves the system for evaluating GW quality.

323

Authors: Udayakumar Nandagopal, Subbulakshmi Thirumalaivelu

Malwares (Malicious Software’s) has increased rapidly in the recent years over the internet, In-order to detect the malwares many anti-malware strategies also been introduced but most of them relay on signature based techniques. But signature based technique failed to detect the new and unknown malwares, so signature based method failed to provide the better result. So it is important to find an technique that could able to detect the both new and unknown malwares without fail. Dynamic malware analysis with sandbox setting provides the best way to analyse the malware in the isolated environment. In this work, inside the cuckoo sandbox malwares were executed for observing behavioural accomplishments of the each and every Malicious Application. After executing the malware inside the sandbox, it will record the activities and finally the report will be provided in the JSON format. System calls needs to be generated from JSON file, MIST file is generated from the JSON file format. This will be organised in the N-Grams after extracting the system call, which helps construct the classifier using the Gained Information (GI) as the selection technique for the submission. Then the overall based on the N-Gram value and Byte length, various classifiers are evaluated, in that random forest has provided the best result.

344

Authors: Yeon Ji Choi, I Nyoman Apraz Ramatryana, Soo Young Shin

The technology of Unmanned Aerial Vehicles (UAV) has been the subject of pioneering research in recent years. Also, the subject has been used in various situations for both outdoor and indoor. In this paper, a cellular communication-based autonomous UAV Navigation is proposed that enables UAVs to maneuver independently in previously unknown and GPS-denied indoor environments with LiDAR sensors. The main idea of the proposed scheme is to implement LTE connection to UAV Navigation and compared it to a Wi-Fi connection. We have proposed that autonomous UAV navigation system for unknown environment like indoor. This system relies on the combination of ROS-based Hector SLAM systems, 2D-LiDAR sensor and LTE connection. The performance of proposed scheme is evaluated as the error distance and exploration time for unknown area like indoor environment. The mapping efficiency of LTE connection is 57.5% greater than the Wi-Fi connection. Computation time for flight is approximately 200 seconds.

361

Authors: Iman Fahruzi, I Ketut Eddy Purnama, Kayo Yoshimoto, Hideya Takahashi, Mauridhi Hery Purnomo

Sleep disorders are challenging to diagnose. The complexity of records obtained from electrocardiogram (ECG) recordings requires manual inspection by experienced medical practitioners. Meanwhile, ECG records are still widely used to diagnose heart problems during sleep. To resolve the issue, the fractal analysis is a promising means to help identify the characteristics of non-overlapping apnea and non-apnea events based on signal scaling behaviour and QRS wave morphologies. Therefore, we propose a new approach to develop automatic sleep disorder classification to minimalize visual inspection and manual scoring. We employed the monofractal and the multifractal analyses to generate new features such as alpha1, residue1, alpha2, residue2, Dqmin, Dqmax, hqmin, hqmid, hqmax, and hqmaxhqmin. To improve the proposed method's performance, we used the ten new features that have been extracted, which are eventually being used as inputs space to a support vector machine (SVM). Through examining the feature set, we designed an optimum SVM model classifier to explore the usability of patterns to predict potential sleep disorder corresponding to apnea and non-apnea events. Hence, our approach through SVM with radial basis function (RBF) kernel is achieved to have accuracy, sensitivity, specificity of 92.16%, 88.24%, 94.12% respectively.

386

Authors: Reshma Siddegowda, Shaila Kalanath, Venugopal Kuppanna Rajuk

The environmental monitoring application is widely used in agriculture, green house farming, floriculture to monitor temperature range in different zone. This application is an embedded application fabricated with wireless sensors, smart phones, protocols, etc. Therefore, with respect to sensor nodes, the battery lifetime and security are the greater challenge in Wireless Sensor Network. In our work, we have designed a protocol using ranking strategy and intelligence similarity function. Initially, the clusters are formed based on the radius range. The SDARIS protocol focused on training the outer nodes of the cluster to acquire balanced cluster. The outer nodes are trained using ranking strategy and are assigned based on Euclidian distance from outer node to cluster head’s and node density of the clusters. This ranking strategy assist to balance cluster size by assigning nodes evenly among all clusters. In addition to balance a cluster size, the SDARIS concentrates on aggregates and secure data during transmission using intelligence similarity function and session keys. The session key is generated for every interval, so the intruder node fails to identify the session key and thus mitigates intruder intervention in the network. Thus, the collaboration of ranking strategy, intelligence similarity function and session key concept in SDARIS protocol improve the overall performance of the network. Hence, SDARIS protocol achieved greater performance with increment of 12% data accuracy, 10% of cluster accuracy, 15% network lifetime and 26% throughput when compared with existing protocol.

410

Authors: Aryo Nugroho, Surya Sumpeno, Kimiya Fujisawa, Mauridhi Hery Purnomo

Campaigns on online social media are becoming intriguing because all parties can compete with each other. Various content can be created to support or defeat one another. Interaction is important to control the battlefield in the face of disinformation. The campaign team must be able to increase interaction and reporting it as a form of achievement. The reality is that the campaign team has not been able to provide a measurable report on the success rate of political campaign interaction on social media. This is because there is no standardization in measuring these interactions. This fact inspires to build a model for measuring campaign interaction on social media. This model will be able to help the campaign team. This study uses a dataset from a governor election campaign in Indonesia which is then tested on a presidential election campaign dataset in America. The model was built using machine learning with the k classification method Nearest Neighbor (kNN) and Distance Weight kNN (DWkNN). The research stages were arranged using the Nassi-Shneiderman Framework. This framework describes the stages and comparison of the use of the KNN and the DWkNN from the training to testing stages in an easy to understand manner. The kNN and DWkNN training stages showed excellent accuracy results of nearly 100%. Furthermore, at the testing stage using the cross-validation method with a variety of fold 5 to 20 variations showed excellent results, namely 99.89% on the governor election dataset and a range of 98% on the American presidential election dataset. This classification model has been tested using several datasets from several candidate social media accounts. We also created a dataset campaign interaction. This study shows that the proposed model can outperform previous studies. So, that it provides novelty in the form of a campaign interaction guide model for the campaign team.

432

Authors: Eman F. Mohsin, Tariq Tashan, Ekhlas H. Karam

In this paper an optimal Immune-Proportional-Integral-Derivative (Immune-PID) scheme is suggested to control to the heart rate of humans and regulate it, based on Yanagihara, Noma, and Irisawa (YNI) as mathematical model of the heart rate. The parameters of the suggested immune PID controller are optimized by using Biogeography-Based-Optimization (BBO) algorithm. In addition, the proposed controller scheme is implemented using Field Programmable Gate Array (FPGA) due to its great data storage capacity, low consumption of energy, and high speed of operation. The Xilinx system generator blocks and Spartan-6 (XC6SLX45T-3FGG484) board are used due to availability with 1% utilization of hardware platform slices. The performance of the proposed controller is compared to traditional PID and immune controllers. That comparative analysis results show best improvement when using the proposed controller with 0% maximum overshoot, a reduction of steady state error and rising time reaches to 0.1 and 0.0023 second respectively and faster response reaches to 0.01 second.

450

Authors: I Gusti Putu Asto Buditjahjanto

The Fuzzy Multi-attribute Group Decision Making (FMAGDM) has been used to assist people in making decisions for sophisticated problems. In Traditional Chinese Medicine (TCM), decision making is used to select the syndrome differentiation of a patient. If the TCM physicians or practitioners are mistaken to determine the syndrome differentiation of a patient can cause wrong in the medication. Conventionally, the problem in TCM is the difficulty in determining a syndrome differentiation of patients. The difficulty is because the numbers of symptoms are numerous and these symptoms are correlated directly to syndrome differentiation. Even some symptoms in one syndrome can emerge from other syndromes. Therefore, the TCM physicians or practitioners must capable to identify the type of symptoms that correlated to the syndrome. The objective of this research is to use FMAGDM to determine the syndrome differentiation in the lung organ from the TCM perspective. Lung syndrome differentiation has five syndromes and 18 symptoms. This research is composed of three methods such as a Fuzzy Linguistic, FMAGDM, and Simple Additive Weighting. The fuzzy set is utilized to capture the experts’ linguistic preference toward the severity level of symptoms, FMAGDM is utilized to build the weight matrix of the experts and Simple Additive Weighting is utilized to select the syndrome differentiation of the patient through selecting the highest preference value of the expert of lung syndrome differentiation. By using data from 9 lung syndrome patients who had different symptoms, the FMAGDM simulation results showed that FMAGDM can determine the type of lung syndrome suffered by each patient. The FMAGDM simulation results for the 9 patients were the same as the diagnosis result of a physician exactly. This shows that FMAGDM can determine the type of lung syndrome as well.

474

Authors: Hussein Almazini, Ku Ruhana Ku-Mahamud

The performance of different mechanisms utilised to perform anomaly detection depends heavily on the group of features used. Therefore, dealing with a multi-dimensional dataset that typically contains a large number of attributes has caused problems to classification accuracy. Not all attributes in the dataset can be used in the classification process since some features may lead to low performance of classifiers. Feature selection (FS) is a good mechanism that minimises the dimension of high-dimensional datasets. Modified binary grey wolf optimization (MBGWO) is a metaheuristic algorithm that has been successfully used for FS. However, the MBGWO algorithm has a drawback in selecting sub-optimal feature sets from an original set of features. This drawback is related to the linearly decreasing value of a parameter where there is no control between the exploration and exploitation processes. This study proposed an enhanced binary grey wolf optimization (EBGWO) algorithm for FS in anomaly detection by controlling the balancing parameter. The new method focused on obtaining a value for a parameter that controlled the trade-off between exploration and exploitation. Evaluation of the proposed method was on the NLS-KDD dataset with different attack classes and compared with other benchmark algorithms, such as binary bat algorithm, binary particle swarm optimization, and four variants of grey wolf optimiser for FS. The experimental results indicated that EBGWO was superior than other algorithms, where it obtained 19 features only out of a total of 41 features with 87.46% classification accuracy. The proposed algorithm can be applied to detect anomaly in network intrusion and outliers in data that are significant but difficult to find.

495

Authors: Adhi Kusmantoro, Ardyono Priyadi, Vita Lystianingrum Budiharto Putri, Mauridhi Hery Purnomo

Microgrid systems are an excellent solution to increasing demand for electricity and are a great way to incorporate renewable energy sources into the electrical energy distribution system. To overcome the fluctuation of renewable energy (PV) based generation, an energy storage system using a battery (BESS) can be used. This paper proposes power management with a modified AC coupling configuration, and a BESS coordinated control strategy based on fuzzy logic. The AC coupling configuration in the microgrid system generally uses batteries to store excess power from the PV. In the proposed AC coupling configuration, two batteries are used, battery 1 (BESS 1) is used to store excess power of the PV array 1 (PVA1) via the inverter battery, while battery 2 (BESS 2) is used to store energy from the PV array 2 (PVA2) ). In this microgrid system, PVA1 is connected to the network, so that the power generated by PVA1 is sent to the AC bus grid using a solar inverter. For two-way power flow on the grid and BESS, inverter batteries are used. The main focus of this article is the analysis of the impact of fuzzy logic (FLC) control strategy and PI control on BESS. When the PV is connected to the grid, control fuzzy logic (FLC) produces an overshoot of 4.73% and a settling time of 14.04 seconds lower than the PI control, overshoot 55.14% and settling time 26.29 seconds. The THD of the voltage and current in this mode is 0.06% and 11.08%. When the microgrid is in island mode, FLC produces an overshoot of 0.54% and a settling time of 13.78, while control PI produces an overshoot of 68.78% and a settling time of 38.11 seconds. The THD of voltage and current in island mode are 0.8% and 11.35%. Therefore, modified AC coupling configuration with fuzzy logic control (FLC) strategy can be used to serve residential loads.

521

Authors: Esther Irawati Setiawan, Hans Juwiantho, Joan Santoso, Surya Sumpeno, Kimiya Fujisawa, Mauridhi Hery Purnomo

Social media development makes it possible for everyone to express their opinions and information through text, speech, video, or images. Multiview sentiment analysis in current studies generally combines two modalities, text and image. It seeks to classify social media posts into two or more polarities, such as positive, neutral, or negative. To improve the performance of multiview Sentiment Analysis, we added another modality, which is concepts derived from text and image. Our proposed model integrates three views into a fusion with an ensemble approach by a meta-classifier. We performed text classification with Deep Convolutional Neural Networks. The input feature is Word2Vec for text representation in order to preserve semantic meaning. Additionally, we analyzed concepts from texts with SenticNet 5 as a knowledge base model and extracted concepts from images using the DeepSentiBank model. We obtained 2089 Adjective Noun Pairs and classified it with Multi-Layer Perceptron. Then we combined predicted probabilities from each classifier for Image, Text, and Concept by Ensemble Learning. A meta-classifier was implemented to predict the final sentiment from a fusion of Image-Text-Concept features. The fusion for multiview sentiment analysis works well and could achieve the best accuracy of 70% by applying the ensemble approach with Logistic Regression as the meta-classifier.

547

Authors: Praveen Kumar Devulapalli, Pardha Saradhi Pokkunuri, Maganti Sushanth Babu

Device-to-Device (D2D) communication can be better option for some applications like military and natural disaster places where we need temporary communication network. Selection of relay nodes and routing schemes are key issues in large scale Device-to-Device (D2D) cooperative networks which affects the performance of the network. The performance may be degraded because of wrong selection of relay nodes. In this paper, we proposed a hybrid cluster based D2D cooperative routing scheme by combining the geographic routing and clustering routing. The main idea of our algorithm is to establish communication between similar mobility devices to reduce the mobility effect since the link between the devices moving with the same (approximately) mobility is reliable. All the equal mobility devices are grouped to form a cluster; one of the devices in this is selected as a cluster head. We selected Cluster Head (CH) based on geographic routing and threshold based cooperative communication is provided within the cluster to improve the network performance. We evaluated efficacy of our proposed scheme over conventional routing approaches discussed in the literature. From the results it was observed that the proposed routing scheme improves End-to-End transmission delay by 39%, Energy consumption by 35% (approximately) and Bit Error Rate performance.