International Journal of Intelligent Engineering and Systems

1

Authors: Oskar Natan, Agus Indra Gunawan, Bima Sena Bayu Dewantara, Junaedi Ispianto

The balance of the aquatic ecosystem is an influential factor in the world of aquaculture, especially in shrimp cultivation. The one that plays a role in that ecosystem is aquatic microorganisms such as vibrio, bacteria, and algae. Therefore, farmers need to know their number and ratio to maintain the shrimp growth. Thus, in this research, models that can estimate vibrio-bacteria ratio and number of algae are developed. These models are formed from aquaculture datasets which are modeled using machine learning algorithms named Gaussian process regressor (GPR) and gradient tree boosting (GTB). Other processing techniques like data pre-processing, feature decomposition, and optimization are also applied to improve model performance. Moreover, these models are also compared to other models which are modeled using another machine learning algorithm like support vector regression (SVR), Lasso, and kernel ridge regression (KRR), so that the best models can be determined. Based on k-fold cross-validation, the GPR model has the best performance in estimating the vibrio-bacteria ratio with mean absolute error (MAE) value of 0.02482 and explained variance score of 0.96515. Then, in the algae estimation, the best performance is achieved by the GTB model with MAE value of 6.55554 and explained variance score of 0.33001.

26

Authors: Vutukuri Sarvani Duti Rekha, Pokkunuri Pardhasaradhi, Boddapati Taraka Phani Madhav, Yalavarthi Usha Devi

A Coplanar waveguide (CPW) fed printed monopole antenna with dual notch band characteristics is proposed for Ultra wideband (UWB) applications. The Ultra wideband antenna operates from 2.48 GHz to 18.35 GHz. Dual notch band is achieved by incorporating double H-slot structure in the radiating element. It achieved notch bands at 3.5GHz (WiMAX) and 8.3GHz (military/radar). The proposed antenna is investigated for the notch bands and operating band as well for various performance parameters like voltage standing wave ratio (VSWR), surface current distribution and far field radiation characteristics. It achieved a peak gain of 5.2dB in the operating band. The proposed antenna is fabricated on fr4 substrate of dimensions 37 x 40 x 1.6 mm3. By analysing the simulated and measured results of the proposed antenna it is observed that the proposed antenna is good for UWB applications by eliminating WiMAX and military/radar application interferences.

46

Authors: Ummadi Sathish Kumar, Edara Srinivasa Reddy

A recent development of a linear dimension reduction (DR) algorithm that is often used in face recognition and other applications has been used to preserve the Locality preserving projection algorithm (LPP). However, in LPP the projection matrix is not orthogonal, so rendering and other apps are not easily reconstructed. Hybridization of Enhanced Orthogonal and Uncorrelated Locality Preserving Projection (HEOULPP) attempts to discover the subspace that best distinguishes distinct face classes by maximizing the gap and reducing the uncertainty within class. In the design of a similarity matrix, HEOULPP algorithm assumes both local data and label data, and needs statistically improved and orthogonal output-base vectors, in order to enhance the OLPP life grade extraction performance. We propose HEOULPP to minimize the locality in an orthogonal projection matrix and to maximize the globality. This investigation is conducted using three datasets, such as YALE face dataset, ORL face dataset, and AR face dataset. Our proposed HEOULPP techniques achieves higher facial recognition rate under three conditions such as occlusion by 70.6%, noise by 69.1% and original by 76.2% than the existing techniques. In addition, dimensionality also reduced in proposed HEOULPP method comparing with traditional methods, including PCA, LPP, OLPP, ULPP and FOLPP. Experimental findings indicated a stronger depiction of data and much greater accuracy in the suggested HEOULPP method.

69

Authors: Sinan Majid Abdulsatar

Coherent Optical Orthogonal Frequency-Division Multiplexing (CO-OFDM) is widely used in optical communication systems, these systems suffer from many problems, including the high Peak-to-Average Power Ratio (PAPR). In this paper, a new method is proposed, it is based on Rotation Odd Constellation (ROC), this is done by dividing the subcarriers into two halves. The first half includes the Odd and the second consists of the Even subcarriers. A phase rotation is added to the first half, and no change is made in the second half. A simulation model has been accomplished by using Matlab platform and VPI transmission maker software to verify the efficiency of the proposed method and the improvment concerning Bit Error Rate (BER), Error Vector Magnitude (EVM) and Quality Factor (QF). The results have been collected for128 subcarriers, 4QAM, 16QAM and bit rate of 10Gb/s, shows the best performance is obtained when /4 rad is the rotation angle.

90

Authors: Ahmed Vall Hemeyine, Ahmed Abbou, Naoual Tidjani, Mohcine Mokhlis, Anass Bakouri

This paper proposes a robust fuzzy logic method based on the Takagi-Sugeno fuzzy controller model. In fact, this one is designed in order to control a Doubly-Fed Induction Generator (DFI-Generator) for the variable wind speed turbine. On the one hand, the main aim of the proposed strategy is the real-time tracking of optimal operating point, extracting the maximum output power, and controlling the pitch. This later is applied in order to limit the output power of the system to its nominal value. On the other hand, the proposed controller can extract the maximum of power under variable wind speed. While the existing methods deals with the case of the fixed wind speed. Thus, in this case, the controller can diverge and cause considerable drop of power, whenever the wind speed is changed. Moreover, the proposed control strategy can increase the efficiency and improve the tracking performance of the system. The model is established based on the Parallel Distributed Compensation technique. The stability and tracking analysis properties are ensured by using the Lyapunov approach theory. In addition, the feedback gains of the strategy controller proposed are determined by Linear Matrix Inequalities. A numerical simulation is carried out under Matlab software for comparing the proposed Takagi-Sugeno fuzzy controller with the conventional PI controller. Effectively, these both controllers are applied to controlling the efficiency of the 3 MW DFI-Generator based on Wind Energy Conversion System in order to test their tracking performances and robustness against rapid changes of the reference signal and for testing the proposed controller strategy as well. Indeed, the simulation results prove the robustness of the proposed controller and show the drawbacks of PI controller, which makes high overshoot under the rapid changes of the reference that can reach 20A in the worst cases. While, the proposed controller strategy tracks rapidly accurately and without any overshoot, the reference signal.

112

Authors: Sethakarn Prongnuch, Suchada Sitjongsataporn

This paper proposes an adaptive step-size approach based on the stochastic gradient-based algorithm for Hammerstein spline adaptive filtering (HSAF).With the good convergence and low complexity computation, a normalized least mean square (NLMS) algorithm is developed with HSAF. A nonlinear Hammerstein adaptive filter consists of memoryless function modified during learning and the spline control point is automatically controlled by gradient-based method. An adaptive step-size approach based on NLMS algorithm with the HSAF is introduced how to derive using the direction of steepest descent method for tracking performance. Hence, the stability and steady-state performance are detailed. The simulation results of trajectories of both update step-size parameters are shown that the initial step-size are reduced, all convergence rates are correspondingly reduced to each own optimum over 100 independent trials of the experiment. Experimental plots of learning curves of mean-squared error of the proposed algorithm for varying the initial step-size parameters are obtained that the proposed algorithm conducts more robust performance and fast convergence compared with the HSAF based on least mean square algorithm, even the large variations of 100-fold initial step-size parameters are used.

134

Authors: Hady Abdel-Maksoud

The main drawback of the switched reluctance motor (SRM) is the torque ripples that restricts its suitability for high-performance applications. Several control techniques have been used to mitigate these torque ripples. Torque sharing function (TSF) is one of the most significant control techniques used for minimizing the torque ripples in SRM. Different TSFs have been developed for reduction SRM torque ripples. However, the difference between the incoming and the outgoing torque responses along with changing of the commutation interval (CI) with changing the motor speed makes the tracking operation difficult. In this paper, a hybrid TSF is proposed. The proposed TSF is formulated based on a linear function for the incoming phase and a nonlinear exponential function for the outgoing phase. Also, the control procedure of the TSFs is developed where the motor speed is become an input along with the rotor position and the reference torque. The CI is accurately adjusted enough for the commutating process. A look-up table model based finite element analysis (FEA) for 6/4 SRM is built. The results of the proposed method are compared with those of conventional TSFs. The effectiveness of the proposed control technique is verified by the simulation results.

154

Authors: Sara H. Muhi, Hadeel N. Abdullah, Bassam H. Abd

This work aims to design an intelligent model capable of diagnosing and predicting the severity of the hepatitis of illness that assists physicians to make an accurate decision. The main contribution is achieved by adopting a new multiclass classifier approach based on a collected real database with new proposed features that reflect the precise situation of the disease. In this work, three artificial neural networks (ANNs) methods, namely, Back Propagation neural network (BPNN), Radial Basis Function Network (RBFN) and, K-nearest neighbor (KNN), used to forecast the level of hepatitis intensity. Real data Collected from the Gastroenterology and Liver Education Hospital of the City of Medicine in Baghdad used as modeling and forecasting samples, respectively, to compare the results of forecasting. The results show that the prediction result by the KNN network will be better than the two other methods in time record to reach an automatic diagnosis with an error rate of less than 1%. Diagnosis accuracy was 99.33% for 2-class and 88% for 5-class, which considered excellent accuracy.

177

Authors: Abdulrahman Sabra, Mohamed Taha, Dina S. M. Hassan

Clustering is considered a critical process to make a fast and reliable connection establishment of ad-hoc networks. In this paper, a new clustering algorithm is proposed, namely Stick Clustering Algorithm (SCA). The proposed SCA is used to overcome the low connectivity, low energy and delay to enhance the performance of large-scale ad-hoc networks. The simulation results showed the efficiency and the reliability of the proposed clustering algorithm against many greatest challenges such as high intensity of nodes under small network area. Also, the results demonstrated that the proposed SCA is very competitive algorithm compared with RCA, CDS, Z-Degree and ZID. The proposed SCA obtained 3.4306 approximation ratio. Also, the efficiency of the SCA is of O(n) for both message complexities and time.

201

Authors: Moine El Hajji, Hassane Mahmoudi, Moussa Labbadi

Air pollution caused by the transport sector has become a very worrying subject. To minimize toxic gas emissions, people start to prefer rail transportation. This high demand has caused high energy consumption in the electric railway. Thus, Moroccan Railway Company was obliged to use different traction locomotives of different technologies which can generate electromagnetic interferences between systems. Reason why the study of electromagnetic pollution has become a necessity to avoid these inter-system electromagnetic disturbances by analysing the experimental measurements on the railway site and by proposing a solution which is the implementation of an active filter capable of eliminating the parasites caused during the connection of the locomotive. However, the authorization to put into operation remains a real subject of study since different parameters must be taken into account, namely the weight, the location, the size and the electromagnetic interferences (EMI) of the filter. This paper discusses the electromagnetic noise by analyzing the impact of the auxiliary power supply quality generated by electric locomotive E1250 on the sound systems when it is coupled to multi-unit Corail cars and by detailing an overview of the conversion chain located in the electric locomotive under test and the sound system installed in Corail car which keeps passengers constantly informed when approaching and arriving at a station, and gives them other indications about the train. Next, static test bench and dynamic experimental tests will be measured in order to determine the Electromagnetic Compatibility (EMC) disturbances in the Common Mode (CM) by studying the harmonics generated by the static converter that feeds the Corail cars, and analyzing their impact on all operating systems especially sound system since it is considered as a very sensitive organ to the least source of noise. In the end, the conversion system will be modeled and a proposed solution will be implemented to mitigate the electromagnetics noises.

222

Authors: Rakesh Kumar Yadav, Rajendra Prasad Mahapatra

Clustering is one of the most significant methods for maximizing the lifetime of the network in wireless sensor networks (WSNs). The technique involves the first grouping of sensor nodes to clusters and second selecting appropriate cluster heads (CHs) for each cluster. In fact, CHs gather data from corresponding nodes and transmit those aggregated data to Base Station (BS). However, the major shortcoming drawn in this technique is to choose optimal CHs. More research works have been already under the investigation on selecting the appropriate CH with respect to different angles. In this scenario, this paper aims to propose a new energy-aware CH selection by hierarchical routing in WSN via proposing a new hybrid optimization technique. More importantly, the selection of CH is carried out under the consideration of stabilization of energy, minimization of the distance between nodes and minimization of delay during data transmission as well. With this, a non-linear objective function is framed, which intent to optimize the CH in such a way that the network lifetime has to be prolonged. Here, for attaining optimal CHs, a new hybrid algorithm is introduced that integrates the concept of Lion Algorithm (LA) and Grey Wolf Optimizer (GWO), which is named as Lion Updated GWO (LU-GWO). Finally, the performance of the proposed model is evaluated by comparing other conventional models with respect to certain performance measures. The simulation analysis shows that the proposed scheme for 50 nodes and 1000×1000 distance was 2.7%, 2.7%, 5.56%, 5.56%, and 5.56% better than GWO, LA, GA, PSO and ABC algorithms with respect to the alive node analysis and the overall analysis proves the superiority of proposed work.

246

Authors: Mohammad Dehghani, Zeinab Montazeri, Om Parkash Malik, Hadi Givi, Josep M. Guerrero

This article presents a new game-based optimization method entitled Shell Game Optimization (SGO). The novelty of this article is simulating the rules of a game known as shell game to design an algorithm for solving optimization problems in different fields of science. The key idea of the SGO is to find the ball hidden under one of the three shells, which should be guessed by players. The main feature and advantage of SGO is that it does not have any control parameters and hence, there is no need to set parameters. SGO is mathematically modeled and implemented on 23 well-known benchmark test functions as well as on a real life-engineering problem entitled pressure vessel design problem. Moreover, SGO is compared with eight optimization algorithms: Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Gravitational Search Algorithm (GSA), Teaching Learning Based Optimization (TLBO), Grey Wolf Optimizer (GWO), Grasshopper Optimization Algorithm (GOA), Spotted Hyena Optimizer (SHO), and Emperor Penguin Optimizer (EPO). The results and data obtained from applying SGO and other mentioned algorithms on unimodal test functions, multimodal test functions, and pressure vessel design problem show that SGO is able to provide better results in comparison with other well-known optimization algorithms. Moreover, results of Wilcoxon signed rank test confirm that SGO achieves more accuracy in comparison with the mentioned algorithms.

268

Authors: Ahmed Aldhahab, Sarmad Ibrahim, Wasfy B. Mikhael

Classification of a brain tumor is a critical step in the design of computer-aided diagnosis systems for Magnetic Resonance Image (MRI) analysis. This work presents an efficient algorithm to classify a tumor in brain MRI images using statistical-based features and deep neural network. Data, within the region of interest, is transformed into two-dimensional discrete Gabor filter and wavelet transform. These filters are combined in this algorithm as directional transformation methods for utilizing all information in all orientations of the MRI input image. MRI Features are extracted based on the first and second order statistics from both domains. Two types of neural network classifiers are employed: Stacked Sparse Autoencoder (SSA) and Softmax Classifier (SMC). Two regularization functions are used in the training of the SA, sparsity regularization and L2-weight regularization. Sparsity regularization controls the firing of the neurons in the hidden layer, whereas L2-weight regularization reduces the effect of the overfitting and improves the performance of the SA. Two datasets are used to evaluate the proposed algorithm. The first dataset consists of 3,064 of T1-weighted MRI slices with three kinds of tumors: Pituitary, Glioma, and Meningioma. The second dataset consists of 200 MRI slices with low-grade and high-grade Glioma tumor collected from the BRATS dataset. The performance of the proposed algorithm is validated using the experimental results in terms of accuracy, specificity, and sensitivity compared to the existing algorithms. For the first dataset, the accuracy obtained is 94.0%, the sensitivity of Meningioma, Glioma, and Pituitary is 87.44%, 97.29%, and 94.27%, respectively, and the specificity of Meningioma, Glioma, and Pituitary is 98%, 96.89%, and 96.78%, respectively. For the BRATS dataset, the accuracy, the specificity, and the sensitivity achieved are 98.8%, 100%, and 100%, respectively.

293

Authors: Lailil Muflikhah, Widodo Widodo, Wayan Firdaus Mahmudy, Solimun Solimun

Liver cancer can be caused by hepatitis B virus (HBV) infection. This genome virus inserts genetic material into its host. Moreover, each gene has numerous HBV DNA sequences due to its high mutation rate in replication. Thus, detecting virus is a difficult task. A support vector machine (SVM) is a robust machine-learning algorithm for detecting liver cancer disease. However, a high data volume can reduce its computation speed and performance measures. Therefore, we propose data simplification using Kernel k-means clustering method to construct the SVM classifier model by minimizing objective function as object distance. Based on experimental results, the proposed method’s performance evaluation was higher than SVM algorithm without kernel k-means, especially for the sensitivity significantly increased. The accuracy rate and AUC of the proposed method were around 98% and 0.95. Furthermore, the performance of proposed method is also predominant of the other machine learning: Random Forest, Naïve Bayes, Naural Network and C5.0.

327

Authors: Kiruthiga Gurumurthy, Mary Vennila Selvadurai

The cloud environment comprised of distributed resources in a dynamic fashion, so this necessitates the need for developing optimal scheduling in cloud environment with the satisfaction of Quality of Service necessitated by the cloud consumer with the maximum profit to cloud providers. But the presence of impreciseness while scheduling cloud resources is the challenging issue of traditional scheduling policies. The main objective of this paper is to treat the vagueness in scheduling of cloud resources by designing Neutrosophic inference system for prioritizing incoming tasks and optimizing the resource usage by applying quantum cuckoo search cache management (INIS-QCSCM). This proposed work represents the parameters involved in resource scheduling by the means of Neutrosophic representation with the aim of reducing the response and execution time with maximize throughput which favours the profit of cloud service providers. Once the job is passed to the virtual machine (local schedulers) it is fine grained at the local host level by utilizing cache segmentation with the knowledge of quantum cuckoo search algorithm. The incoming task request is prioritized using neutrosophic inference model by considering impreciseness as the important factor. In local host memory management specifically cache assignment is done by cuckoo search-based model which selects the best cache allocation policy to optimize the resource utilization with quality of service. To prove the performance of this proposed INIS-QCSCM, its make span, degree of imbalance handling, energy consumption, execution time of each task and the hit ratio of the cache management are used as the evaluation metrics. The efficacy of INIS-QCSCM is proved by comparing it with existing classification models support vector machine, Fuzzy inference system and intuitionistic fuzzy inference system. As the outcome of the simulation results proved that INIS-QCSCM model provides potential resource scheduling with optimized resource usage in cloud computing by treating hesitancy, impreciseness and vagueness in adversarial situation of cloud paradigm.

350

Authors: Sinan M. Abdulsatar, Hadi T. Ziboon, Humam F. Majeed

An essential element and one of the challenging tasks in cognitive radio is spectrum sensing. In this paper, a new method to enhance the sensing is presented and implemented, in this scheme eigen-values are obtained from the sample covariance matrix of the signal that received at the secondary user’s receivers. Random Matrix Theory (RMT) derives the expression of the thresholds required for effective sensing. Low signal to noise ratio cases have been overcome by the new method which bestead the traditional energy detection method as well as the conventional eigenvalues based on single threshold. The simulation results based on randomly generated signals show that the proposed method exhibits a detection probability of (91.3%) when signal to noise ratio is (-19dB) while it’s (78.5%) for classical eigenvalues method. The results also show that the samples’ number that needed for reliable sensing is less than the energy detector and the conventional eigenvalues with single threshold method. An implementation of the the proposed method based on FPGA using myRIO-1900 kit is presented in this paper, the results obtained from the simulation and the hardware implementation are identical.

373

Authors: Salem Abdulla Awadh Ba Hmaid, Vasanthi Varadharajan

Integration of Wireless Sensor Networks (WSN) with the Internet Protocol (IP) has led to the development of Internet-of-Things (IoT) networks with high connectivity and improved data transmission with limited power supplies. Hence, it is necessary to utilize a high performance multi-path routing protocol to avoid energy constraint problems and network connectivity issues. This paper presents an optimal multi-path routing protocol using multi objective algorithms namely Fractional Firefly algorithm with Chicken Swarm Optimization (FFA+CSO) to resolve the energy constraint problem. First the network is clustered and a Cluster Head (CH) is selected to initiate inter-cluster and intra-cluster communication. Fractional Firefly algorithm (FFA) has been developed for this purpose by overcoming the slow convergence and local optima problem of firefly algorithm. FFA selects CH based on energy, delay, link quality and lifetime. Then the routes are formed and optimal route is selected using Chicken Swarm Optimization (CSO) based on energy, inter and intra-cluster delay, link quality, lifetime and hop count. The proposed FFA+CSO routing protocol is evaluated using Network Simulator-2. Results obtained for 100 nodes showed that FFA+CSO provided efficient multipath data transmission for WSN based IoT networks with 24% less delay, 28% greater throughput, 18.7% lesser energy consumption, 21.6% longer lifetime, 20% higher PSNR and 37.5% less number of hops than the existing routing models.

397

Authors: Tenia Wahyuningrum, Azhari Azhari, Suprapto Suprapto

Usability evaluation focuses on how well users can learn and use products to achieve their goals. Evaluation of e-commerce website usability can be done using the Fuzzy Analytical Hierarchy Process (FAHP) method. Usability is considered a multi-criteria problem due to its complex structure. Logarithmic Fuzzy Preference Programming (LFPP), a non-linear programming approach, is one of the FAHP methods for weighting criteria. However, deficiencies in the LFPP method, in some cases, result in a consistency value of 0 from a consistent fuzzy comparison matrix, which should be greater than 0. This study modifies the LFPP method with the Modified LFPP (MLFPP) models. Modifications were made as a development of the LFPP method in improving the accuracy of the consistency index (λ^*) before weighting criteria. The MLFPP makes improvements to the functional constraints, while at the same time improving the determination of criteria weights. The MLFPP model uses arctan to limit functions to improve consistency index. The test results show that the MLFPP can improve the accuracy of the consistency index values in the pairwise comparison matrix given by experts to be more than 0, with an average increase of 14.9%. The amount of consistency index (λ^*) changes has improved the usability score by 6.59%.

422

Authors: Kethavath Prem Kumar, Thirumalaisamy Ragunathan, Devara Vasumathi, Pamulapati Krishna Prasad

In recent years, cloud Load balancing (CLB) is the significant research area because vast data are stored on the servers leads to increase the loads on cloud servers. A trade-off on servers are maintained by LB techniques by distributing less power with equal load. In this paper, Cuckoo Search with Firefly Algorithm (CS-FA) is proposed for LB in a cloud environment. Initially, the capacity and load of each virtual machine are calculated. If the load of the virtual machine is greater than the balanced threshold value then, the LB algorithm is used for allocating the tasks. The CS-FA algorithm selects the best Virtual Machines (VMs) for assigning the tasks and migrate the overloaded VMs tasks to under-loaded VMs task. This algorithm majorly avoids the imbalanced workload performances in a cloud environment. The performance of proposed CS-FA method is compared with existing LB techniques such as Dynamic LB (DLB), Hybrid Dynamic LB (HDLB) and Honey Bee Behavior LB (HBB-LB) to evaluate the capacity and load of methods. The results showed that the existing HDLB method migrate seven tasks whereas the CS-FA method migrates only two tasks. The experimental result shows that proposed CS-FA method migrate only two tasks when number of loads is 40 and existing method migrate 6 tasks.

446

Authors: Sujatha Arun Kokatnoor, Balachandran Krishnan

Online Social Media (OSM) produce enormous data related to the human behaviours based on their interactions. One such data is the opinions expressed and posted for any specific issue addressed in the OSM. Majority of the opinions posted would be categorized as positive, negative and neutral. The lighter group’s opinions are termed anomalous as it is not conforming the regular opinions posted by other users. Though, lot of conventional classification and clustering based learning algorithms works well under supervised and un-supervised environment, due to the inherent ambiguity in the tweeted data, anomaly detection poses a bigger challenge in text mining. Though the data is un-supervised, for the learning purpose it is treated as Supervised Learning by assigning class labels for the training data. This paper attempts to give an insight into various anomalies of OSM and identify behavioural anomalies for a Twitter Dataset on user’s opinions on demonetization policy in India. Through Self-Supervised learning, it is observed that 86% of the user’s opinions did agree to the demonetization policy and the remaining have posted negative opinions for the policy implemented.

1

Authors: Sahirul Alam, Selo Sulistyo, I Wayan Mustika, Ronald Adrian

Vehicular ad hoc network (VANET) is the emerging wireless communication technology in vehicular domain to support the safety and non-safety purposes in transportation. Despite the potential implementation of VANET, some challenges mainly due to the dynamic environment need to be addressed and tackled. In this paper, the horizontal handover especially in vehicle-to-vehicle (V2V) connection is investigated. The dominant handover decision method in wireless network, namely received signal strength (RSS) based method, is actually inappropriate for the implementation in V2V VANET. Therefore, a utility based handover decision method is proposed in this paper. The utility function is formulated based on the value of signal to noise ratio (SNR) and the load of network. Despite the fluctuation of SNR, the proposed method can make the better handover decision i.e. by reducing the ping pong effect and improving the throughput by selecting the network with lower load. In this research, the proposed method is evaluated through simulations. The results of simulations show that the proposed method can improve the performance of V2V handover in terms of throughput and handover rate compared with the RSS based method.

20

Authors: Akram Saeed Aqlan Alhammadi, Vasanthi Varadharajan

Virtual Machine (VM) consolidation is a crucial process in improving the utilization of the resource in cloud computing services. As the cloud data centers consume high electrical power, the operational costs and carbon dioxide releases increases. The inefficient usage of the resources is the main reason for these problems and VM consolidation is a viable solution. VM consolidation includes host overload/under-load detection, VM selection and VM placement processes. Most existing host overload/under-load detection approaches of VM consolidation uses CPU utilization only for the determining host load. In this paper, three resources namely CPU utilization, memory utilization and bandwidth utilization are used for host overload detection and an adaptive regression based model called Multiple Regression Multi-Objective Seven-Spot Ladybird Optimization (MR-MOSLO) is proposed. This model is based on combining the benefits of adaptive threshold based and regression based host overload detection algorithms. This approach of combining these features provide more advantages for threshold setting in dynamic environments with accurate prediction of host overloading. For this purpose, initially, Multiple Regression (MR) algorithm is used which relay on CPU utilization, memory utilization and bandwidth utilization for estimation of the host load conditions. Then a Multi-Objective Seven-Spot Ladybird Optimization (MOSLO) algorithm is introduced to select the upper and lower threshold limits for host utilization. Based on these algorithms, the host overload/underload is detected with high accuracy and less power consumption. The simulations are conducted in CloudSim tool and the empirical results shows that the proposed MR-MOSLO algorithm detects the host overload efficiently. The results obtained for 25 hosts, 30 VMs and 500 tasks, are: SLATAH is 20.0434, PDM is 8.7E-4, SLAV is 3.7E-5 and ESV is 10.962 which are lesser than the other methods. Though the energy of 15.4 kWh and SLA of 0.00757 are negligibly higher than some of the existing methods, the proposed approach provided comparatively better performance.

40

Authors: Amal Fouad, Hossam M. Moftah, Hesham A. Hefny

Brain cancer importance emanates from the importance of the brain as an organ and its functions. It has a great effect on the whole human body. Identification brain cancer according to its type, it refers to a multiclass classification problem in the machine learning world. In the real-world, object detection and classification face numerous challenges. The object has a large variation in appearances. In this research, a Haar Discrete Wavelet transforms hybrid with the Histogram of Oriented Gradients (HDWT-HOG) features descriptors are proposed by the local gradients in MR image as shape information. The whale optimization algorithm (WOA) plays a great role to reduce the numbers of HOG and Harr features from 38,640 to 120 features only which are less than .01% from all features. This reduction doesn't affect the system performance but it saves time in the classification phase. The test image is matched with its learned class by performing a Bagging ensemble learning classifier. Bagging achieves 96.4% in average accuracy but when Boosting is used, it achieves 95.8%.

63

Authors: M. Rao Batchanaboyina, Nagaraju Devarakonda

The hybrid Improved monarch butterfly optimization- mutual nearest neighbor (IMBO-MNN) is proposed for outlier detection in high dimensional data. It is a challenge to detect outliers in high dimensional information. The external behavior of the data points cannot be detected in high-dimensional data except in the locally relevant data sub-sets. Subsets of dimensions are called subspaces, and with an increase in data dimension, the number of those subspaces grows exponentially. In another subspace an information point that is an outlier can appear ordinary. It's essential to assess its outlier behavior according to the amount of subspaces in which it appears as an outermost part to characterize an outlier. Data is scarce in high-dimensional space and the concept of closeness does not preserve meaning. In fact, the sparsity of the high-dimensional data means that every point is nearly equal from the point of view of closeness-based finishes. As a result, for higher dimensional information finding is more complicated and non-obviously significant outliers. An enhanced MBO (IMBO) algorithm is offered for enhanced search precision and run time efficiency by a fresh adaptation provider. Statistical results indicate that the elevated local optimal prevention and quick convergence rate of the improved monarch butterfly optimization (IMBO) algorithm helps to exceed the basic MBOs in outlier detection. Comparatively, IMBO produces very competitive outcomes and tends to surpass present algorithms. Optimal value k remains a task, affecting the efficiency of kNN straightforwardly. We are presenting a fresh learning algorithm under kNN in this paper to alleviate this issue called mutual nearest neighbor (MNN). The main feature of our method is that the class marks of unknown instances are defined by mutually next to one another, instead of by closest neighbor. The advantage of mutual neighbors is that in the course of the prediction process pseudo close neighbors can be identified and taken not into account. The performance of the suggested algorithm has been examined with a number of studies. For 100 data, IMBO-MNN is in 4897 milliseconds, PSO is in 5239 milliseconds, random forest is 5347 milliseconds, PNN is 5278 milliseconds and KNN is in 5166 milliseconds. For 250 data, IMBO-MNN is in 5984 milliseconds, PSO is in 6132milliseconds, random forest is 6145 milliseconds, PNN is 6124 milliseconds and KNN is in 6152 milliseconds. For 500 data, IMBO-MNN is in 6416 milliseconds, PSO is in 6636 milliseconds, random forest is 6634 milliseconds, PNN is 6719 milliseconds and KNN is in 6710 milliseconds. For 1000 data, IMBO-MNN is in 6913 milliseconds, PSO is in 7111 milliseconds, random forest is 7019 milliseconds, PNN is 7134 milliseconds and KNN is in 7162 milliseconds. The proposed IMBO-MNN performs better with minimum time taken. The findings indicate that the technique can identify outliers in high dimensional data efficiently in a decreased calculation moment.

83

Authors: Nattapon Jaisumroum, Jirarat Teeravaraprug

This paper is dedicated to the study of electricity generation expansion planning considering uncertainty. The method used for this problem is utilizing a highly constrained uncertainty optimization issue, as it were be solvable and complete classification. The demonstration uses a toll function to minimize environment effect and operational costs. A Neural Network is utilized to long-term forecast in electricity. The stochastic optimization model consists of finding the optimal long-term electricity generation planning for the construction of new generation capacity subject to various economic and technical constraints. The model can be showed experimental results that enable long-term planning of electricity generation with optimization of the costs, pollution emission and uncertainty of these generation resources in determining the best solution. The findings can help decision the government policy not only adjust the existing electricity generation but also coordinate the conflict interaction among system cost, energy supply, pollutant mitigation as constraint violation risk.

108

Authors: Shahryar Sorooshian, Nor Filiani Aziz, Noor Azlinna Azizan

The Performance Measurement Systems (PMSs) seen as part of a factor that indicates the successful and unsuccessful company; especially in the Small and Medium Enterprises (SME’s). The main objective is to analyze most suitable PMSs for Malaysian SMEs based on applicability. In the methodology, qualitative design with applied Analytic Hierarchy Process (AHP) is used. The expert panel is selected from practitioners linked with PMSs and SME’s in Malaysia. The findings show the Balanced Scorecard (BSC) as first ranked of PMS in SMEs. Followed by BSC are the Theory of Constraints (TOC) and Result and Determinate Framework (RDF). The last three lowly ranked PMS which are the Medori and Steeple Framework (MDF), Dynamic Multi-dimensional Performance Framework (DMPF) and Holistic Performance Measurement Framework (HPMF). At the end of the study, the researcher makes the discussion and conclusion; it is also advised to improve and implement the suitable models in the Malaysian SME’s for further success.

127

Authors: Indra Waspada, Riyanarto Sarno, Kelly Rossa Sungkono

The existing method of graph-based process model discovery has weaknesses in detecting parallel relationship (XOR, AND, and OR). The algorithm only works on a particular graph structure, so it must be reconfigured when applied to other different structures. To answer this problem, this paper proposes an improved method of parallel model detection, which is designed in two phases. The first one consists of three steps; firstly is to count and record the value of relationship frequency into every node in a graph model. Then, the second step implements the algorithm to discover the concurrent relationship. The third step detects all possible split and join relationships. Based on the first phase, then a consistent and robust parallel discovery algorithm can be developed. The first parallel algorithm is to identify the XOR relationship. This algorithm is designed with the rule that the XOR pattern cannot have a concurrent relationship between its branch nodes. Next, the algorithm for detecting AND and OR must detect the existence of any concurrent relationship in its branches. Then, AND and OR pattern is differentiated by their unique characteristic of relationship frequency at branch nodes. To verify the ability of the proposed methods in which the existing method fails, we have designed four scenarios. Scenario 1 and 2 consecutively were arranged with two and three branches parallel model. Scenario 3 located the AND and OR inside the XOR pattern. In scenario 4 the sequence relationships were inserted between split and join of parallel patterns. The experimental results show that the proposed method successfully recognizes and differentiates XOR, AND and OR patterns correctly in all scenarios. It also sounds in all discovered model and get 100% fitness.

151

Authors: Bambang Setiawan, Supeno Djanali, Tohari Ahmad

Imbalanced class is one of the main problems for intrusion detection models that use machine learning methods. The classifiers generally are designed to minimize the global error rates, which have not considered the condition of imbalanced class. The amount of training data for each type of imbalanced attack can cause those intrusion detection models to have high accuracy but can also lead to difficulty in identifying minority class attacks. In this research, we propose an intrusion detection model using a combination of the feature selection method for imbalanced class and weighted support vector machine classifier. We apply a composite performance index in the features selection and the optimization of the weight of the minority class. The experimental result using the NSLKDD dataset shows that this model produces overall accuracy, sensitivity, and specificity that reaches more than 99%, with false alarms below 0.5% and false-negative rates below 0.7%. The sensitivity of the U2R and R2L classes are 56% and 92%.

173

Authors: Dedy Trisanto, Nofita Rismawati, Muhamad Femy Mulya, Felix Indra Kurniadi

Credit card fraud is an issue that has affected Indonesia payment system over a decade. Sometimes, the result of the fraud used for terrorism and other crimes. Financial loss is not the only problem that is affected caused by credit card fraud but also Indonesia images in international trade, e-commerce, and the merchant. Currently, a trusted and secured banking payment system is crucial for both customers and banks. The problem from credit card fraud dataset is the data have many features and imbalanced class, this problem leads the paper to propose undersampling technique and feature reduction methods. In this paper we proposed two stage-feature reduction technique because a stage feature reduction could not find the optimal features. On the other hands, we are also applied Instance Hardness Threshold sampling and Random undersampling to deal with imbalance data. The two-stage feature reduction is chosen to eliminate the ineffective feature that cannot eliminate using only one feature reduction. The model from the implemented machine learning methods is evaluated using accuracy, specificity, recall, and Matthews Correlation Coefficient. We implemented our proposed approaches in the ULB credit card fraud detection dataset. According to the result, the undersampling gives a boost in performances which improve the recall and MCC score, the IHT undersampling provide goods results, and in some cases, the result can predict all the test set correctly. However, the two-stages feature reduction fails to improve the accuracy, precision, recall, and MCC score. In one case, the method reduced the accuracy score to 0.302.

196

Authors: Ryan Setiawan, Tohari Ahmad

The internet is widely applied to facilitate the transferring process of data as a means of communication. Some of this transferred message has confidential information and must be protected. One of the techniques to protect is data hiding. It uses carrier media as a cover to hide information inside it, one of which is audio. However, there are some problems in the use of audio, namely the limited capacity and sound quality after data insertion. In this paper, we address these problems by interpolating the audio samples using Modification Lagrange Interpolation (MLI) and exploring Reduce Difference Expansion (RDE) for embedded the secret. Next, samples are smoothed to improve the quality based on three sampling signal points of the stego audio. The testing of proposed method is carried out on fifteen mono audio data sets from various genres the covers and ten files as confidential information data with various file sizes. The results show that using the proposed methods, the performance increases that the average quality of each audio genre raises by about 4dB and able to increase the capacity of message data that can be hidden by 20%.

216

Authors: Istiyo Winarno, Mochamad Ashari, Heri Suryoatmojo

This paper proposed a Hybrid PI-Fuzzy based Static Var Compensator for minimizing voltage deviations as the integration of wind power generation to the distribution network. Hybrid PI-Fuzzy Logic controller constructed by the cascade-connected conventional PI controller and Fuzzy Logic controller. This algorithm combined the advantages and minimized the disadvantages of each conventional controller. The control method applied to the Static Var Compensator, which functioned as a dynamic reactive power source. The controlled dynamic reactive power source influenced the system with proper reactive power value, which resulted in the voltage regulation. Two scenarios were carried out to investigate the efficacy of the proposed method. The first scenario demonstrates the performance of the proposed system in wind speed variations and the second scenario in load variation conditions. Finally, the comparisons with Fuzzy Logic Controller-based self-tuning PI control were made to investigate the performance of the proposed control method. Simulation results verified that the proposed control algorithm outperformed the FLCbased self-tuning PI controller, in terms of the steady-state voltage deviation minimization by 1.47% and the settling time of the voltage respond decrement by 92.78%.

239

Authors: Aye Mya Hlaing, Win Pa Pa

The main objective of this paper is to improve the naturalness of Myanmar Text-to-Speech (TTS) system without using time-consuming and expensive annotation of training corpus. Recently, word embedding, which has the advantage of training directly from a large amount of raw text data, has been used as the additional input features with the conventional input features or as the replacement of that conventional input features on the acoustic modelling of TTS systems. In this paper, the effectiveness of applying word vectors as the additional input features are investigated for Myanmar speech synthesis on the three acoustic modelling techniques, Deep Neural Network (DNN), Long Short- Term Memory Recurrent Neural Network (LSTM-RNN), and a hybrid of DNN and LSTM-RNN (Hybrid-LSTMRNN). For the purpose of achieving better TTS performance, we built our own word vectors for Myanmar language. We further explore the best modelling method and vector dimension of word embedding for Myanmar TTS systems. Both objective and subjective evaluations are done on DNN, LSTM-RNN and Hybrid-LSTM-RNN based Myanmar TTS systems with and without additional input features such as Part-of-Speech (POS) and word vectors. According to the subjective results, applying additional input features in DNN, LSTM-RNN, and Hybrid-LSTM-RNN based Myanmar TTS systems can improve the naturalness of the synthesized speeches though objective results cannot lead to the significant improvement in LSTM-RNN and Hybrid-LSTM-RNN based systems. To the best of our knowledge, this is the first attempt to apply word vector features in neural network based Myanmar TTS systems.

259

Authors: Dwaram Jayanarayana Reddy, Akepogu David Donald, Kothapalli Seshadri Ramana, Karanam Srividya Lakshmi, Pullala Cheruvu Sai Divya

In the financial stock market, a sequence of prices obtained from the share market with respect to the time series is usually examined. Generally, time series in finance, particularly shows importance in predicting investment in today’s share market. Since there are too many factors such as public opinions, general economic conditions, or political events, vulnerability in the economy are directly or indirectly reflects on the evolution of financial time series. The desire of the investor is to predict the future stock prices neglecting whether the investor is a long-term investor or a day-trader. A major challenge is to develop and design an efficient predictive model that guides investors to make appropriate decisions. In this research work, Long Short Term Memory-Recurrent Neural Network (LSTM-RNN) is developed to overcome such disputes and contributing an efficient technique for predicting the future stock prices financially. In addition to the model, cross entropy is calculated using a Mutual Information feature selection model to minimize the optimization problems that create the time complexity in the system. The proposed LSTM-RNN has achieved best accuracy of 61.33% of prediction accuracy compared to state-of-the-art techniques.

278

Authors: Agung Wiratmo, Chastine Fatichah

Distributed representation of a sentences cannot only be seen with the sequence but also with dependency. In this paper, we proposed an answer assessment model that considers with dependency relational each word. The dependency relational is obtained by universal dependency modelling from CoNLL format data. The dependency relational is used in Long Short-Term Memory (LSTM) architecture by modified the hidden state, which is called dependency tree LSTM. The proposed method has an improvement on QWK and accuracy in 2.38% and 2.05%, respectively, that compared with the LSTM state of the art in the English short essay. Furthermore, the proposed method with an Indonesian short essay shows the evaluation of QWK and accuracy of 68.07% and 82.51%, respectively.

299

Authors: Eman El-Sayed Mahmoud, Ashraf Soliman Mohamed

Search engines’ capabilities are improved continually for enhancing the retrieved results and satisfying users’ needs. Despite the great efforts in the information retrieval field, the user spends long time in changing the search keywords; though the retrieved results may be out of user’s expectation. This may be due to the huge number of web resources, the lack of search keywords for specific domains, and the gap between the user and author in specific domain. This paper proposes a model for improving the search engines capabilities through adding level of awareness to the web resources. The model targets the human view about web resources in addition to author perception. This is through automatically extraction of the user’s perceptions and mapping them to the resources’ based on semantic aspects. The experimental studies realized 36% precision improvement compared to other search engines, and user satisfaction was up to 92%.

317

Authors: Theerawat Rojanapitoon, Jirarat Teeravaraprug

Warehousing can be expensive because order picking requires considerable vehicular movement and labor hours. Although many previous studies have focused on order picker routing, there exists a lack of research on the simultaneous increase of order picking speed and energy reduction in rectangular warehouses with varying levels of traffic in each aisle. This study accordingly developed and evaluated a mathematical model for determining optimal picker routes considering the total travel time and energy consumed. The results were validated using the brute-force search method and benchmarked with the time-staged (TS) model. The energy savings were determined by comparing a time-optimized use-case (T) with one optimized for both time and energy (TE). Both use-cases provided routes up to 44% faster than the TS and avoided more than 50% of congested paths, and TE which represents the full functionality of our model provided a possible energy savings of up to 17% over T.

1

Authors: Gavisiddappa, Shivakumar Mahadevappa, Chandrashekar Mohan Patil

In the present scenario, multimodal biometric authentication is one of the emerging fields, which is applied in different applications like prison security, criminal identification, banking security, etc. The objective of this research work is to develop an effective feature selection algorithm to determine the optimal feature values for further improving the performance of multimodal biometric authentication. Initially, the input images were collected from Chinese Academy of Science Institute for Automation (CASIA) dataset. Then, feature extraction was carried-out by using Local Binary Pattern (LBP), minutiae feature extraction, Histogram of Oriented Gradient (HOG), and Gray-Level Co-Occurrence Matrix (GLCM) features like cluster prominence, Inverse Difference Moment Normalized (IDMN), and autocorrelation. After feature extraction, modified reliefF feature selection algorithm was used for rejecting the irrelevant features or for choosing the optimal features. In modified algorithm, Chebyshev distance measure was utilized instead of Manhattan distance in order to find the nearest miss and nearest hit instances. At last, the optimal feature values were given as the input for Multi-Support Vector Machine (MSVM) classifier for classifying an individual as an authorized or unauthorized person. The experimental result showed that the proposed system improved the classification accuracy up to 7% as related to the existing systems in terms of accuracy.

25

Authors: Mummadi Gowthami Reddy, Palagiri Veera Narayana Reddy, Patil Ramana Reddy

An image that is highly informative is produced by unifying the information from couple of or numerous source images which is referred as image fusion. It has been employing in most of the applications in the medical field like detecting of tumours, treating Alzheimer’s, surgery of brain with the assistance of computer and some other clinical diagnosis. Successful diagnosis of specific diseases necessitates the enhancement in the exactness of fusion algorithms. For examining the body of human, the images obtained from magnetic resonance imaging (MRI) and computed tomography (CT) plays a vital role. Channelizing the maximum info from the source images to the fused image with understate loss of info that must mitigate the presence of artifacts in the fused outcome is the basic idea of any fusion methodology. In this context, a novel medical image fusion approach is implemented, that utilizes integrated guided and nonlinear anisotropic (IGNLA) filtering with image statistics. This approach upholds the info of texture in the fused images more efficaciously. In addition, proposed medical image fusion is extended for color images and applied to MR-Gad, MR-T2 and SPECT-Tc images. Extensive simulation results of proposed medical image fusion are compared with traditional and recent image fusion algorithms and disclose the superiority of proposed approach with respect to image quality metrics.

46

Authors: Sudarsan Prabhakaran, Niranjil Kumar Ayyamperumal

In modern decades, a human socioeconomic lifestyle has been disrupted by their stress levels. Normally, the stress levels were measured based on the Electrodermal Activity (EDA) analysis that analyses the changes in the Skin Conductance Response (SCR) level. As a result, different sensors were designed to analyze the changes in human emotional levels using the EDA signals. During analysis, the unwanted noise components in the signals were removed by using a Compressed Sensing-based Decomposition (CSD) approach. Conversely, this approach has high computational complexity. Therefore in this article, a Modified CSD (MCSD) for EDA signal is proposed that may vary the shape of the impulse response with time and varied noise models. The key aim of this model is to improve the recovery accuracy of EDA signal decomposition and enhance the human stress monitoring system. In this modified model, a computationally efficient decomposition method is proposed by using matrix-free convex optimization modeling that exploits the Toeplitz structure to allow decomposition of EDA signals with provable bounds on the recovery of true SCR events by using the computationally efficient algorithm and guarantee a better recovery accuracy. Finally, this approach is tested on Wearable Stress and Affect Detection (WESAD) dataset to exhibit that the proposed MCSD approach achieves a recovery accuracy of 92.4% which is 2.55% higher than the existing CSD approach.

65

Authors: Amira S. Mahmoud, Sayed A. Mohamed, Reda A. El-Khoribi, Hisham M. AbdelSalam

Fast and automatic object detection in remote sensing images is a critical and challenging task for civilian and military applications. Recently, deep learning approaches were introduced to overcome the limitation of traditional object detection methods. In this paper, adaptive mask Region-based Convolutional Network (mask-RCNN) is utilized for multi-class object detection in remote sensing images. Transfer learning, data augmentation, and fine-tuning were adopted to overcome objects scale variability, small size, the density of objects, and the scarcity of annotated remote sensing image. Also, five optimization methods were investigated namely: Adaptive Moment Estimation (Adam), stochastic gradient decent (SGD), adaptive learning rate method (Adelta), Root Mean Square Propagation (RMSprop) and hybrid optimization. In hybrid optimization, the training process begins Adam then switches to SGD when appropriate and vice versa. Also, the behaviour of adaptive mask RCNN was compared to baseline deep object detection methods. Several experiments were conducted on the challenging NWPU-VHR-10 dataset. The hybrid method Adam_SGD acheived the highest Accuracy precision, with 95%. Experimental results showed detection performance in terms of accuracy and intersection over union (IOU) boost of performance up to 6%.

86

Authors: Fika Hastarita Rachman, Riyanarto Sarno, Chastine Fatichah

Detecting song emotion is very important, however many studies have been done based on song lyrics and song audio separately. This research proposes a method for detecting song emotion based on integrated song lyrics and audio. Synchronizing the right structural segment and lyrics of song can be used for hybrid approach to detected right emotion. Song emotion can be classified into Thayer emotion label. The features of a song lyric are extracted using Psycholinguistic and Stylistic; whereas the features of a song audio are extracted using analyze audio signal waveform using Fast Fourier Transform (FFT) method. A song can be divided into 5 structural segments, which are intro, chorus, bridge, verse and outro. A preprocessing method of audio uses Correlation Features Selection (CFS) and preprocessing text for lyrics. Six classification methods are used for classifying emotion based on lyrics and audio of song structural segments separately. The aggregate method is used to analyze the results of classification before to obtain structural segments that represent emotions. Then, the final process Hybrid approach used to combine audio and lyrics features in emotion detection. Sum of matrix and Majority Voting Concept are used for Hybrid approach. The value of F-Measure is 0.823.

113

Authors: Nouf Saeed Alotaibi

A new algorithm for increasing fingerprint with large noise is enhanced in this paper. The fingerprint image is first pre-improved depending on Gabor filters, and local adaptive thresholds are used to achieve a binary fingerprint. In order to reconstruct these regions, which have incorrectly improved in the first stage, the classification deep Boltzmann machines (DBMs) with a range pattern before are used. The proposed technique completely enhances each other by a traditional technique of improvement relying upon Gabor filtering and deep learning. The FVC 2004, Biometrika, Italdata, Crossmatch, and Swipe databases are conducted by different techniques. Experiments indicate that in contrast with other techniques, the suggested technique achieves stronger outcomes and enhances fingerprint performance. Experimental findings indicate that the use of the proposed technique allows the extremely accurate fingerprint image to be reconstructed. The average performance of the proposed method is 91.31, which is better than the average performance of another method. When there exist 18,000 features, the running time of UniNap1, ATVS, Dermalog, and CAoS are 68, 74, 59, 87 s respectively, while the running time of the proposed method is 45s.The proposed technique is superior for removing noise than other techniques, in particular for improving fingerprints of poor performance.

135

Authors: Sethakarn Prongnuch, Suchada Sitjongsataporn, Theerayod Wiangtong

This paper presents a heuristic approach for workflow scheduling in heterogeneous distributed embedded system (HDES). A genetic algorithm (GA) and ant colony optimization (ACO) modified with the greedy algorithm introduced to the system contains multiple heterogeneous embedded machines (HEMs) working as a cluster. Users can remotely access and utilize their computational power. The communications on different types of buses are taken into account to find an optimal solution. New meta-heuristic information based on forwarding dependency is proposed to build probability for ACO to generate task priorities. Besides, a greedy algorithm for machine allocation is incorporated to complete task scheduling. Experiments based on random task graphs running in the HEM cluster demonstrate the effectiveness of the modified greedy ant colony optimization algorithm which outperforms the others by 33% more result quality.

156

Authors: Van-Tan Bui, Phuong-Thai Nguyen, Van-Lam Pham, Thanh-Quy Ngo

Antonymy and synonymy are basic semantic relations between words. Automatically distinguishing between antonymy and synonymy is an important task in natural language processing. This task is hard because antonyms and synonyms tend to occur in highly similar contexts. Recent studies often focus on exploiting dense-vector representations of words to deal with this problem. In this paper, we present a study on antonymy-synonymy discrimination for the Vietnamese language. We propose a deep neural network model (DVASNet) that can utilize not only embedding representations of words but also co-occurrence contexts and specific patterns of Vietnamese word structure. Our experimental results showed that the proposed method achieved significant improvements in comparison with a number of state-of-the-art methods by 14% to 17% in terms of F1 score for both benchmark datasets.

181

Authors: Nam-Soo Kim

In this paper, we propose an overlay cognitive radio (CR) non-orthogonal multiple access (NOMA) network to improve the performance of the primary receiver (PR) as well as the secondary receiver (SR). To accomplish the purpose, firstly, we adapt combining technology, selection combining (SC) and maximal ratio combining (MRC) which utilizing both signals from an indirect path and a direct path for the PR. Secondly, we select a relay which provides the spatial diversity gain among the candidate secondary sources (SSs) for the SR. The performance of the PR and the SR is derived in closed-form and verified through simulation. Numerical results demonstrate that the performance of the PR with SC or MRC always better than that of the PR of a conventional overlay CR NOMA system which receives from the indirect path only. Also, to maintain the same outage probability of the PR with MRC compared to that with SC at the given condition, 2 dB less transmit power is noticed. Furthermore, the performance of the SR with the selected relay improves with the increase of the number of the SSs. Consequently, we show that the proposed network, which assumes a direct link and adapts a selected relay, has better performance than the conventional overlay CR NOMA network.

203

Authors: Suthep Varnasilpin, Ruedee Masuchun

UL (U-shaped Line) reduces the waste material in a manufacturing process. This paper presents the application of three U-shaped Lines with the minimum workstations which consists of UL-1, UL-2, and UL-3. The uncertain task of three U-shaped Lines determines the time as an interval. Three U-shaped Lines formulate the model with a binary integer linear programming. Each uncertain task reduces the normal time to the optimism time. The normal time reduction relates to the allowable time inversely. This paper indicates the procedure of the normal time reduction that applies MATLAB R2014a. The allowable time of UL-1, UL-2, and UL-3 for each uncertain task is 1.5 times of the optimistic time, 1.2 times of the optimistic time, and 1 time of the most optimistic time, respectively. For the utility of this paper, the uncertain task adjusts the finish time which does not affect the modification of the minimum number of workstations.

227

Authors: Yassine Gmih, Abdelmajid Farchi

This article proposes a new and simple miniaturized tag for radio frequency identification (RFID) in the European and Moroccan UHF band (866-868 MHz). This tag can be mounted on the surfaces of the human body and may be used for identification cards. For human body modelling, we used a four-layer laminated elliptical cylinder with physical parameters for electromagnetic simulations. The radiation characteristics of the antenna placed in proximity to the human model and the effects of absorption caused by the human body were analysed and controlled. Our antenna is composed of a meander dipole and a T-shaped adaptation network structure. We used the CST-MS for the design of the antenna and the human body model and verified the results with the HFSS simulator. The designed tag was manufactured to test the reading range. The measured range is about 6 m when the tag is placed near to the chest and between hands. The new features of the proposed new tag are characterized by its simple structure, reliable stability on the surface of the human body, a long reading range and a miniaturized size of 46 x 20 mm2. Our antenna was designed on a low-cost, single-sided FR-4 substrate with a thickness of 0.8 mm. Compared to the conventional ultra-high frequency patch; our antenna achieves a miniaturization rate of 91%.

248

Authors: Mohssine Chakib, Tamou Nasser, Ahmed Essadki

In this paper, we propose a new control strategy called linear active disturbance rejection control (ADRC) used to control the wind energy conversion system (WECS) based on doubly-fed induction generator (DFIG). This recent control strategy is compared with the conventional RST polynomial control method considered better and more efficient than the classical PI method according to different works. By the extended state observer ESO of the ADRC controller, the internal and the external disturbances on the system are estimated and compensated as a total disturbance in real time. Moreover, the proposed ADRC method is easy to tune and more practical which has only one tuning parameter. The RST controller has three polynomials chosen in order to reduce the effect of disturbance. These controllers have been used to control the stator powers of DFIG in order to extracting the maximum energy produced by the wind by using a maximum power point tracking (MPPT), to regulate the dc-link voltage in order to maintain DC bus voltage at constant value and to regulate the reactive power to ensure a unitary power factor. Then, these both control strategies are designed and their performances are compared in order to deduce the most efficient in terms of reference tracking and robustness. The simulation results prove that the proposed ADRC control technique is more robust and have better performance than the RST control.

270

Authors: Abbas Albaidhani, Ahlam Alsudani

For location-based applications, wireless systems in an indoor environment often operate under non-line-of-sight (NLOS)conditions that may cause ranging errors. A promising technology for location-aware sensor networks is an Ultra-wide bandwidth (UWB) transmission due to its robust operation in harsh environments, fine delay resolution, and power efficiency. However, the existence of walls and other obstacles causes a notable challenge in terms of localization, as they can result in positively biased distance estimates. A non-line-of-sight (NLOS) identification and mitigation technique created for UWB technology. A distance mitigation method is proposed using online line-of-sight (LOS) and NLOS identification by Fuzzy logic control decision. The result of the proposed method shows 99 % of the accuracy of the proposed NLOS identification method and less than 20 cm of the measured distance error.

291

Authors: Noha Y. Hassan, Wael H. Gomaa, Ghada A. Khoriba, Mohammed H. Haggag

With the evolution of social media platforms, the Internet is used as a source for obtaining news about current events. Recently, Twitter has become one of the most popular social media platforms that allows public users to share the news. The platform is growing rapidly especially among young people who may be influenced by the information from anonymous sources. Therefore, predicting the credibility of news in Twitter becomes a necessity especially in the case of emergencies. This paper introduces a classification model based on supervised machine learning techniques and word-based N-gram analysis to classify Twitter messages automatically into credible and not credible. Five different supervised classification techniques are applied and compared namely: Linear Support Vector Machines (LSVM), Logistic Regression (LR), Random Forests (RF), Naïve Bayes (NB) and K-Nearest Neighbors (KNN). The research investigates two feature representations (TF and TF-IDF) and different word N-gram ranges. For model training and testing, 10-fold cross validation is performed on two datasets in different languages (English and Arabic). The best performance is achieved using a combination of both unigrams and bigrams, LSVM as a classifier and TF-IDF as a feature extraction technique. The proposed model achieves 84.9% Accuracy, 86.6% Precision, 91.9% Recall, and 89% F-Measure on the English dataset. Regarding the Arabic dataset, the model achieves 73.2% Accuracy, 76.4% Precision, 80.7% Recall, and 78.5% F-Measure. The obtained results indicate that word N-gram features are more relevant for the credibility prediction compared with content and source-based features, also compared with character N-gram features. Experiments also show that the proposed model achieved an improvement when compared to two models existing in the literature.

312

Authors: Jalaney Jabamony, Ganesh Ramaswamy Shanmugavel

Advancement of the public transport system is important to modern society for reliable performance. Intelligent public transport system can utilize the time very effectively to give better performance to the society. Fast advancement in equipment, programming, and correspondence innovations has encouraged the rise of Internet-associated devices that give perceptions and information gathering from modern reality. By interfacing, an internet-enabled device with the public transport system leads to the intelligent public transport system. This paper proposed intelligent public transport with IoT enabled system to give an accurate prediction to the arrival time of the bus to the particular bus stops. Here, Artificial Neural Network (ANN) is used as a prediction algorithm and ANN is trained with different traffic parameters and environmental conditions. Parameters which are considered in the proposed system includes Distance(D), Waiting Time at Stops (WTS), Red signal Duration at Traffic Signal (RSD), Traffic Density (TD), Turning Density (TRD), Rush hours (RH), Weekends (WE), Weather conditions (WC), Number of passengers in the bus (NP), Public Holiday (PH), Road Type (RT). These parameters and the measured real-time arrival time of the bus in different stops for 10 days is used for training the ANN. This trained ANN is implemented on the server-side. In this paper, the performance of the proposed IOT-ANN-SPTS system is compared with the existing previous related works. From the performance analysis, it is shown that the proposed system produced less error (MAE=134.582, RMSE=197.738, MAPE=28.87) when compared to the other methods.

334

Authors: Bain Khusnul Khotimah, Miswanto Miswanto, Herry Suprajitno

In the case of high dimensional data with missing values, the process of collecting data from various sources may be miss accidentally, which affected the quality of learning outcomes. a large number of machine learning methods can be applied to explore the search area for imputation and selection of features and parameters. ML classification needs preprocessing with self-organizing map imputation (SOMI) before the imputation of missing values is done to improve the accuracy of the model. This study introduces a new approach that combines naïve Bayes classification (NBC) and genetic algorithm (GA) optimization procedures to effectively explore the search space based on a sample of experimental points. GA is a classification model approach based on the selection of features that cause computational problems, such as reduced dimensions, uncertainty and imbalanced data sets with various classes. In the experiment, preprocessing the data using SOMI yielded error results that were up to 10% for various data sets with missing data compared to other methods. In the SOMI-GANB hybrid model, the experimental results show that the proposed method can significantly improve accuracy by up to 90% compared to other imputation methods and without feature selection. SOMI can be used for homogeneous, heterogeneous and mixed data sets. The results from the experiment clearly showed that the proposed method could significantly increase the yield compared to the other imputation methods and without feature selection. The combination of GA and naïve Bayes classification was chosen because they are simple, easy-to-understand methods that are very effective in finding optimal solutions from a set of possible solutions. Naïve Bayes imputation had higher accuracy compared to neural network imputation.

354

Authors: Ronald Adrian, Selo Sulistyo, I Wayan Mustika, Sahirul Alam

The clustering method in the vehicular ad-hoc network provides an opportunity for a cluster head to improve the network connections, but it still remains a problem in the border cluster. The vehicles, in view of their position in the slices of two clusters, receive a weak signal from the cluster head. The intersection area causes a throughput decrease for cluster members. To provide the network connection in the border cluster, we, in turn, proposed an adaptive border node clustering by utilizing the combination of K-Medoids algorithm, modified Genetic Algorithm and modified Tabu Search. We ameliorated the recent model in Enhanced Model of Weighted K-Medoids Clustering Algorithm by adding the fusion process to the best gene pieces into one individual temporary. The effects of this fusion were on the production of a temporary cluster that can make the main cluster environment more stable. This temporary cluster was adapted to the position of the two closest main cluster heads. Our model was found able to increase throughput and to keep the stability of cluster members at any velocities. The increase in throughput represents an improvement parameter in the network quality of service (QoS). We achieved the overall throughput at 93.97% (throughput vs cluster member) compared with the previous methods and the cluster member stable at around 25 vehicles (cluster member vs transmission range) in varied conditions.

379

Authors: Sarmad Ibrahim, Shamam Alwash, Ahmed Aldhahab

This paper presents an Enhanced Water Cycle Algorithm (EWCA) to optimize the network reconfiguration and distributed generation (DG) integration simultaneously for minimizing system power losses and improving voltage stability index (VSI) in the distribution system while considering all operational constraints. For validation, the performance of the proposed method is compared with other methods, which utilized well-known meta-heuristic algorithms. Different cases for network reconfiguration and DG integration are carried out in order to evaluate the performance of the proposed method. The effectiveness of the proposed method is assessed using the IEEE 69-node radial distribution system. According to the simulation results obtained, the proposed method in which the simultaneous optimal network reconfiguration and the DG size and location are implemented can provide a remarkable solution in terms of power loss reduction and voltage profile improvement. The proposed method also proved its superiority compared with other existing methods in terms of power loss reduction (i.e., 84.2% loss reduction compared with the base case).

1

Authors: Kumbala Pradeep Reddy, Gullipalli Apparao Naidu, Bulusu Vishnu Vardhan

Robust Action Recognition under multiple views has gained a significant research interest recently. To enhance the performance of Multi-view Action Recognition, we propose a novel Feature extraction and Feature Selection mechanism that allows building a mutual relationship between the actions sequences of multiple views. The feature extraction considered multiple features which are invariant to scale and orientation. Three different features such as Intensity Features, Orientational Features and Contour Features are used to represent every action. Further, the feature selection is accomplished through self-similarity matrix and is very much helpful in the provision of a perfect discrimination between actions sequences of different views. The proposed method is validated over the standard multi-view IXMAS dataset and experimental results confirm that the proposed method outperforms the conventional approaches with respect to Recognition Accuracy.

25

Authors: Eman K. Elsayed, Naglaa E. Ghannam

Ontology is the core of the semantic applications, so the quality of it has a direct proportion with Ontology. It impacts directly on the long life of the semantic applications. There are different negative effects in the design of Ontology' classes as Blob, Lazy Class, Large Class, and Singleton. These negative effects called bad smells. However, detecting smells is not supported in any Ontology editors. This paper proposes a metric method called ONTOPYTHO. It can detect classes’ smells automatically from Ontology models even if it is a Big Ontology. We programed ONTOPYTHO via Python and SPARQL languages. We evaluated the proposed method ONTOPYTHO by applying it on twelve publicly OWL Ontology projects. The detected smells appeared 117495 times in the twelve projects. The results showed that both the size and the number of classes of OWL Ontology has no effect in the presence of the smells. The results also showed that 69.24% of the classes are lazy classes. This means that big OWL Ontologies are not big in their nature, but because of theses lazy classes. The proposed method is the first method that detects Lazy class smell in the design of big OWL Ontology. In our random sample of big Ontologies, Lazy class smell appears approximately 99.8% of the smells.

47

Authors: Farza Nurifan, Riyanarto Sarno, Kelly Rossa Sungkono

Many restaurant review analysis have been done, however only few analysis have been done for specific aspects of a restaurant. In this context this paper proposes aspect based restaurant analysis which includes Physical environment, Food quality, Service quality and Price fairness. The analysis steps include Aspect Term Extraction (ATE), Aspect Keyword Extraction (AKE), Aspect Categorization (AC) and Sentiment Analysis (SA). ATE employs the modification of Double Propagation method and several Topic Modelling methods, AKE utilizes Term Frequency-Inverse Cluster Frequency (TF-ICF), in AC we propose Hybrid ELMo-Wikipedia (HEW), and in SA we propose Hybrid Expanded Opinion Lexicon-SentiCircle (HEOLS). The results show that our modification of the methods used in ATE could increase the f1measure of the AC by average 2%, then the HEW that we proposed had better f1measure compared to other similar methods by average 6%. Other than that, our proposed HEOLS can expand and redetermine the Opinion Lexicon polarity and can increase f1measure of SA by 6%.

71

Authors: Agus Zainal Arifin, Safri Adam, Avin Maulana Mohammad, Fatoni Anggris, Rarasmaya Indraswari, Dini Adni Navastara

Segmentation of single tooth in dental panoramic images is an important process to extract its features and information. However, it might be challenging when the segmentation process faces an overlapping teeth image. In this research, we introduce a new strategy for detecting overlapping area on dental panoramic radiographs automatically. This research proposes automatic thresholding to obtain marking points for the overlapping area and an automatic selection of overlapping area candidates by using the area orientation and the similarity of neighborhood intensity. The experimental results on 44 images show that our proposed strategy can detect overlapping teeth on the dental panoramic radiograph with accuracy, sensitivity, and specificity of 75%, 66.67%, and 85%, respectively. The evaluation conducted on 24 overlapping teeth images shows that the segmentation results of overlapping teeth area have an average misclassification error of 0.31%.

91

Authors: Youssef Barradi, Naoufel Khaldi, Khalida Zazi, Malika Zazi

This paper presents a network reconfiguration which is a vital analysis process for optimizing and controlling distribution systems. The method is based on genetic algorithm by changing the status of the switches to improve the operational performance. The main objective is to minimize the system power losses and to keep bus voltage profile into limits with radial distribution to provide the consumers with quality electrical energy while minimizing the cost. For this optimization problem, an objective function is developed from an electrical branch to sort the fittest solution. Selection, Crossover and mutation are the necessary three operators in which some improvements are made for the effectiveness of the genetic approach. The method can be successfully applied for loss minimum problem. Numerical example simulated with MATLAB/GUI is demonstrated by 33-bus distribution network and tested using a default mode network. As results, premature convergence is avoided, it shows the validity of the proposed methodology while respecting all the constraints.

114

Authors: Ari Endang Jayati, Wirawan Wirawan, Titiek Suryani, Endroyono Endroyono

Generalized Frequency Division Multiplexing (GFDM) is one of the 5G candidates to overcome the shortcomings of Orthogonal Frequency Division Multiplexing (OFDM), high peak rating power ratio (PAPR), and high Out of Band (OOB) radiation. GFDM has a low PAPR due to the use of a number of subcarriers. The purpose of this paper is to compare the two algorithms to reduce PAPR if applied to the non-linear distortion-affected GFDM system. Partial Transmit Sequence (PTS) and Selected Mapping (SLM) are selected because these techniques do not distort the signal, so they do not change the spectrum of the signal. The simulation result shows PAPR GFDM is not significantly affected by nonlinear distortion. After being given the PTS technique, the PAPR value on GFDM dropped to 5 dB. Meanwhile, after being given a selective mapping technique, the PAPR value in GFDM dropped to 6.2 dB.. However, for a 5G application with thousands of devices, this value should still be reduced. After being given a PTS, the PAPR value drops to the value according to the 5G criterion. Better PTS performance decreases PAPR for GFDM systems that are given nonlinear distortion when compared to SLM.

133

Authors: Osama Sayed Abdelrahman, Hesham Ahmed Hefny

The term "Big Data" is a buzzword which describes new technologies that manipulate very large data sets which are massively generated by heterogonous sources. This new term encourages data scientists to extend their work and modify their techniques to overcome the new challenges come with big data concepts. Granular computing has emerged as a new rapidly growing information processing paradigm inside the community of Computational Intelligence. Theories of Fuzzy sets and Rough sets theory are considered powerful examples of granular computing that can be applied to data mining techniques to extract nontrivial knowledge from huge data. The aim of this paper is to introduce a data mining approach for big data based on integrating fuzzy sets and rough sets theories. The proposed approach provides a novel granular data mining for big data that allow extracting useful knowledge and rules from huge data to enhance the decision making process. The proposed approach has been applied on different types of datasets. The experimental results show that our proposed approach is more efficient and robust when dealing with very big datasets and obtained consistent classification rules with classification accuracy 100%.

157

Authors: Il Bae, Jaeyoung Moon, Shiho Kim

One of crucial concern of autonomous vehicles is that driver may experience discomfort, if the driving pattern of the vehicle is quite different from her/his personal driving style. This study focuses on an autonomous vehicle control based on user’s driving preference adopted to provide a personalized and a familiar driving experience to autonomous vehicle users. For this purpose, we defined a driver’s driving preference metric (DPM) to reflect their own driving style by specifying a preferred lateral and longitudinal acceleration region. Moreover, we also proposed a time optimal speed planning that utilize the DPM control parameters extracted from a GG diagram to provide a DPM-aware maneuvering for a personalized autonomous driving. The performance of the proposed planning strategy and a combined vehicle controller were verified using on-road experimental tests. The experimental result shows that the proposed DPM-aware system can control the lateral and longitudinal acceleration of autonomous vehicle within 28.7% of peak error tolerance from the specified criterion of DPM.

177

Authors: Karima Boudaraia, Hassane Mahmoudi, Ahmed Abbou

This paper presents a novel hybrid technique for tracking the maximum power point of the photovoltaic panel. This approach includes two loops: The first one is the ANN (Artificial neural network) loop that is used to estimate and generate the reference of optimal voltage. While, the second loop consists of the BSM (Backstepping Sliding Mode) controller. Effectively, the proposed controller is designed to track the signal of the desired voltage, which is generated using the first loop, by adjusting the duty cycle of the boost converter. Thus, this loop allows the DC/DC converter to product the maximum power at the terminals of the PV module and the load. Indeed, in contrary to the traditional backstepping, the proposed ANN-BSM technique guarantee zero steady-state error. On the one hand, by using the ANN, the system can quickly predict the desired optimal voltage, also it allows the system to avoid the unnecessary calculations and research of the maximum point of power. On the other hand, the sliding mode and the backstepping controller are used to provide the good performances and robustness against the rapid changes of insolation. In addition, the asymptotic stability of this system is made by applying the Lyapunov approach. To show the effectiveness and the tracking performances of the proposed ANN-BSM technique, a comparative study with the classical methods, P&O and incremental conductance algorithms, and hybrid approaches such as the ANN-Backstepping controllers and the ANN-Integral Sliding mode, is investigated in MATLAB/Simulink software.

207

Authors: Wahyu Pamungkas, Titiek Suryani, Wirawan Wirawan, Achmad Affandi

A communication system of Vehicle to Vehicle (V2V) consists of a transmitter, a receiver, and scatterers which can move in random speeds and directions to generate the Doppler Effect that reduces communications system performance. The Doppler effect that occurs in V2V channel modeling is a combination of the Doppler effect on each of its constituent components. After being combined with the Orthogonal Frequency Division Multiplexing (OFDM) multi-carrier system, the signal received at the receiving end experiences attenuation due to the characteristics of the channel used. To analyze the attenuation factor received due to the influence of channel characteristics, the power spectral density parameter is used. Previous research was limited to modeling the V2V channel with moving scatterers, without knowing the effect of velocity on power spectral density. The purpose of this study was to examine the effects of velocity parameters of V2V’s components on power spectral density where scatterers moved randomly. Moving scatterers speed was classified into two categories, where the scatterers were faster or slower than the sender and receiver. The power spectral density was analyzed after combining Quadrature Phase Shift Keying (QPSK) modulation with multi-carrier OFDM. The results show that components with higher velocity yielded more damped power spectral density on the receiver that reached -0.7099 dB/Hz. Moreover, the number of scatterers that moved around the transmitter and receiver highly affected the magnitude of the Doppler effect. The largest Doppler Shift generated in this study was 13.5 kHz with attenuated power spectral density at -0.8646 dB/Hz.

225

Authors: Gehad Ismail Sayed, Ghada Khoriba, Mohamed H. Haggag

This paper proposes a new hybrid approach based on quantum computing and SSA to boost the performance of SSA through finding the optimal trade-off between exploitation and exploration. QSSA relies on embedding the integration of the quantum operators and interference in the optimization process of SSA and the quantum representation of the search space. QSSA is tested on 29 global benchmark functions of CEC2017 and 8 benchmark images. The simulation results showed that employing the principles of quantum can significantly boost the performance of SSA. The performance of QSSA is compared with SSA and other recent and well-known optimization algorithms. The results on CEC2017 demonstrated the capability of QSSA to find the global optima for most of unimodal and multimodal benchmark functions, especially with complex search space. Moreover, the results revealed that the proposed image enhancement based QSSA is robust and can distinctly enhance the contrast of the color images.

249

Authors: Maddali M.V.M. Kumar, Aparna Chaparala

A large number of small sensors in the Wireless Sensor Network (WSN) can be an efficient tool for data collection in a range of environments. Each sensor transfers the information to the basic unit that transmits the information to the end user. The objective of clustering is to separate the network into sectors with a cluster head (CH). The job is to collect, aggregate and transmit cluster heads to the base station. Energy Efficient Clustering has been used widely for energy conservation in wireless sensor networks (WSNs) and also preserves the restricted energy resources of their sensors. Cluster heads (CHs) perform a major part and exhaust power faster than other member nodes in a distributed WSN. Opposition-based Chaotic Whales Optimization algorithm (OBC-WOA) is meta-heuristic optimization algorithm which has recently been proposed in opposition. It simulates humpback whales ' social behaviour. OBC-WOA produces randomly its population during exploration and exploitation stages, like other population-based systems, which can produce values far from the optimal alternative or block the development of local optima. The revised algorithm known as Opposition-based Chaotic Whale Optimization Algorithm (OBC-WOA) is designed to increase solution precision and reliability. The OBC-WOA uses a technique based on opposition to improve the efficiency of OBC-WOA. The OBC-WOA is screened with the initial WOA algorithm and other meta-heuristic techniques. The performance of the proposed approach is evaluated in terms of energy consumption, throughput, packet delivery ratio and network life time. When compared with the existing method the proposed method is 12% better than Whale Optimization algorithm (WOA), 25% better than Gravitational Search Algorithm (GSA), 40% better than Particle Swarm Optimisation (PSO) and 8% better than Fuzzy K-Means and Centralized Mid-point Algorithm (FKM-CMA).

1

Authors: Surya Prasada Rao Borra, Rajesh Kumar Panakala, Pullakura Rajesh Kumar

Nowadays, all the medical diagnosis is achieved by using the Digital Image Processing (DIP). Because, the usage of DIP is more important in the medical field to identify the activities of the patients related to various diseases. Magnetic Resonance Imaging (MRI) and Computer Tomography (CT) scan images are used to perform the fusion process. In brain medical image, MRI scan is used to show the brain structural information without functional data. But, CT scan image is included the functional data with brain activity. To improve the low dose CT scan, Hybrid algorithm is introduced in this paper which is implemented in FPGA. The main objective of this work is to optimize performances of the hardware. This work is implemented in FPGA. The combination of Discrete Wavelet Transform (DWT) and Principle Component Analysis (PCA) is known as hybrid algorithm. The Maximum Selection Rule (MSR) is used to select the high frequency component from DWT. These three algorithms have RTL architecture which is implemented by Verilog code. Application Specified Integrated Chips (ASIC) and Field Programmable Gate Array (FPGA) performances analysed for the different methods. In 180nm technology, DWTPCA- IF architecture achieved 5.145mm2 area, 298.25mW power, and 124ms delay. From the fused medical image, Mean, Standard Deviation (SD), Entropy, and Mutual information (MI) performances are evaluated for DWT-PCA method which has better performance than conventional methods.

22

Authors: Pundru Chandra Shaker Reddy, Alladi Sureshbabu

India is a country which has exemplary climate circumstances comprising of different seasons and topographical conditions like high temperatures, cold atmosphere, and drought, heavy rainfall seasonal wise. These utmost varieties in climate make us exact weather prediction is a challenging task. Majority people of the country depend on agriculture. Farmers require climate information to decide the planting. Weather prediction turns into orientation in the farming sector to decide the start of the planting season and furthermore quality and amount of their harvesting. One of the variables are influencing agriculture is rainfall. The main goal of this project is early and proper rainfall forecasting, that helpful to people who live in regions which are inclined natural calamities such as floods and it helps agriculturists for decision making in their crop and water management using big data analytics which produces high in terms of profit and production for farmers. In this project, we proposed an advanced automated framework called Enhanced Multiple Linear Regression Model (EMLRM) with MapReduce algorithm and the Hadoop file system. In the proposed model (EMLRM) first, we stored the unstructured weather data in hadoop distributed file system (HDFS), process that stored data by using MapReduce Algorithm and build the rainfall prediction model by utilizing Multiple Linear Regression. We used climate data from IMD (Indian Metrological Department, Hyderabad) in 1901 to 2002 period. The experimental outcomes show that the EMLRM provided the lowest value of Root Mean Square Error (RMSE= 0.274) and Mean Absolute Error (MAE= 0.0745) compared with existing methods. The results of the analysis will help the farmers to adopt effective modeling approach for predicting long-term seasonal rainfall.

43

Authors: Aditi Jahagirdar, Manoj Nagmode

Extracting keyframes for action recognition purposes is a challenging task, as compression is to be achieved without losing the impression of the action. Insufficient or wrong keyframes can lead to confusion in the type of action. Proposed two-level key frame extraction algorithm uses an adaptive threshold technique to identify most dissimilar frames as keyframes. At the first level, global features based on intensity histogram and at second level local features computed from wavelet decomposition are used as similarity measures. A new performance parameter, Compression Ratio-Normalized Fidelity (CRNF) is introduced for evaluating the keyframe extraction algorithm. Average CRNF of 0.81 is achieved by the proposed algorithm, as compared 0.37 achieved by existing histogram-based method for action recognition dataset. CRNF value of 0.95 is achieved for Open Video project dataset, which is more than existing methods. Qualitative results further prove the effectiveness of the proposed algorithm.

63

Authors: Aditya Kumar Sahu, Gandharba Swain

Since the inception of the reversible data hiding (RDH) concept, it has been a compelling topic in the field of data hiding. Being reversible, it has the ability to restore the original image followed by the successful retrieval of the secret data, at the receiving side. The concept of the dual stego-image based RDH technique utilizes two identical images of the original image for concealing the secret data, has gained wide compliance. Therefore, someone with both the stego-images can only extract the concealed data. In this paper, two improved dual imaging based RDH techniques, such as (1) dual stego-image based pixel pair LSB matching with reversibility, and (2) dual stego-image based modified LSB matching with reversibility, are proposed. In technique 1, at first two mirrored images are obtained from the original image. Then, using the pair of two consecutive pixels from the original image, the mirrored images pixels are modified using LSB matching technique. Later, these pixel pairs are readjusted to ensure reversibility at the receiving side. Similarly, technique 2 utilizes each original pixel to generate two distinct stegopixels using modified LSB matching. The experimental result shows that the technique 1 maintains excellent peak signal-to-noise ratio (PSNR) of 51.29 dB and 51.30 dB for the two stego-images with hiding capacity (HC) of 524288 bits. At the same time, technique 2 offers 51.19 dB and 49.44 dB of PSNR while exhibiting the equal HC. Further, investigation with various image quality assessment (IQA) metrics like quality index (QI), and structural similarity index (SSIIM) are proven to be competent over the other existing works considered in this paper. In addition, both the proposed techniques have shown excellent anti-steganalytic ability against RS and pixel difference histogram (PDH) attack.

85

Authors: Mahmood Hamza Al-Mufraji, Thamir R. Saeed, Ansam Subhi Jabbar, Jafar W. Abdul Sadah

In many applications, in industry and laboratory, there is a robot with a human, in close motion, to satisfy the work requirement. Therefore, for safety and monitoring the quality, the classification between them is important. The robot is considered as a semi-rigid object because it has many parts in the half upper, while the lower half is rigid, and a human is a non-rigid object. This paper presents a practical result which investigates the classification between these two objects based on the micro-Doppler signatures. The object's data were acquired using an S-band 2.4 GHz radar. The improved Stockwell transform was used to analyze the radar received signal in the timefrequency domain and feature extraction. The singular value decomposition is used for data filtering based on the Hankel matrix (HSVD) while using the Temporal Radial Basis Function to satisfy the classification. The classification accuracies gained are; 96.4%-100% for semi-rigid, 92%-95% for non-rigid closed objects.

104

Authors: Tejas Mehta, Chandu Bhensdadia

In this paper, we present an adaptive approach in order to match and retrieve near duplicate images at different scales. Matching only local Features does not necessarily identify visually similar images. Global features are fast at matching but less accurate. Many existing methods either use local features or CNN features for image or video retrieval task. In this paper, we combined the use of SURF local points and CNN features extracted around SURF points in order to match near duplicate image pairs. Image pairs are segmented into blocks and CNN features of the image block containing matched SURF features are extracted and matched. Regions around matched image blocks are grown adaptively and matching is carried out until CNN mismatch is observed. To verify our proposed approach, experiments are carried out on benchmarking California-ND and Holiday dataset. Compared to traditional approaches for image retrieval, our approach not only retrieves relevant images but also provides detail of localized matched patch. For California-ND dataset and Holiday dataset, we achieve remarkable mAP (mean average precision) score up to 0.86 and 0.74 respectively.

130

Authors: Nagendar Yerukala, Venu Nalla, Padmavathi Guddeti, V. Kamakshi Prasad

Data encryption play major role in all communications via internet, wireless and wired media. Stream ciphers are used for data encryption due to their low error propagation. This paper presents a new design of stream cipher for generating pseudorandom keystream with two LFSR`s, one FCSR and a non-linear combiner function, which is a bit oriented based on alternating step generator (ASGF). In our design two LFSR`s are controlled by the FCSR . ASGF has two stages one is initialization and the other is key stream generation. We performed NIST test suite for checking randomness on the keystream of length 1500000 generated by our design with 99% confidence level. Keystream of ASGF passes almost all NIST tests for randomness and the results are tabulated. For fixed extreme patterns of key and IV, ASGF is giving random sequence. Throughput of our proposed stream cipher ASGF is 4364 cycles/byte. Throughput comparison of ASGF with existing stream ciphers A5/1, A5/2 and RC4 are presented. It is not possible to mount brute force attack on ASGF due to large key space 2^192. Security analysis of ASGF against exhaustive search, Algebraic attack, distinguishing attack is also presented in this paper.

149

Authors: Ali Mohamed Riyad, Mohammed Saleem Irfan Ahmed, Husni Hamad Almistarihi

An intrusion detection system is one of the network security tools installed to monitor suspicious activity in the network and act as a last line of defense. It normally notifies about the skeptical activity occurred in the network using sensors by sending alarms to the administrator. However, the IDS present in the large network generates not only a large number of alerts but also abundant false alerts. These generated alerts are very difficult to handle as it increases the burden for the network administrator and also pulls down the performance of the defense system. In order to overcome the issue, various countermeasures have been proposed. Commonly, to increase the quality of alerts, the alerts are post-processed in such a way that the false alerts are filtered out thereby refining the performance of the IDS defense. In this paper, we propose an IDS quality framework using alert post-processing techniques to separate out the false alerts generated by various sensors in the network. At low level alert post-processing, the priority scores are assigned based on the quality measures to filter the irrelevant alerts having less significance. At high level alert postprocessing, higher level operations such as alert aggregation, clustering, and hyper alert correlation have been carried out to minimize the number of alerts and the high level report consisting of significant alerts is presented to the administrator. Experiments have been conducted using DARPA 2000 dataset to assess the performance of the proposed system. The system has produced pleasing results than many of the existing methods with 95% of alert reduction rate, 99% of completeness and 100% of soundness towards enlightening the quality of the alerts generated by the IDS.

170

Authors: Amsaveni Muthusamy, Duraisamy Subramani

Attribute reduction for big data is an important preprocessing step in the area of data mining. A multi-step dimension reduction approach was proposed for attribute reduction in big data. It addressed the non-linear relationships within the attributes. The data dimension was reduced through a parametric mapping. The mapping parameters were estimated using low-rank Singular Value Decomposition (SVD). However, the user-defined criterion in multi-step dimension reduction approach has greatly influenced the efficiency of attribute reduction. This approach was proposed for a single machine that means the entire big data must fit in the main memory and the parallelism was limited. So, in this paper, parallel rough set theory based attribute reduction approach is proposed for attribute reduction in big data. Based on two descriptions of lower approximation and upper approximation, a rough set is constructed. Then a reduct is detected using inner importance measure and outer importance measure. The rough set theory is used in MapReduce framework to achieve the parallelism for attribute reduction in big data. Hence, the computation time is reduced by using parallel rough set theory based attribute reduction approach. Finally, the experiments are carried out in Amazon customer review, REUTERS-21578 and International Cancer Genome Consortium (ICGC) on AWS datasets to prove the effectiveness of parallel rough set theory based attribute reduction in terms of accuracy, precision, recall and computation time.

190

Authors: Ari Firmanto, Riyanarto Sarno

Product reviews can be used as suggestion for companies to improve their services in this digitalization era. These reviews can be presented in more detail using aspect based sentiment analysis. In this research, aspect-based sentiment analysis method is proposed using grammatical rules, word similarity, and SentiCircle. This method began with extracting the candidate aspects rules based on phrase detection in constituency parse. The candidate aspects were categorized using word similarity. Word similarity calculated the similarity value between the candidate aspects and the keywords obtained from Wikipedia. To determine sentiment polarity, SentiCircle is used. SentiCircle can capture the contextual sentiment from the data. The result showed that the proposed method was able to categorize aspects correctly, with the highest f1-measure value of 84%, while sentiment analysis produced the highest f1-measure of 87%.

214

Authors: Vemula Vinay Kumar, Kodali Sai Krishna, Sarukonda Kusumavathi

Brain tumour segmentation is a significant task in the Medical Image Processing (MIP). Early diagnosis of the brain tumours plays an important role in improving treatment possibilities and maximizes the survival rate of the patients. Manual segmentation of the brain tumours for the cancer diagnosis form large amount of MRI images that generated in clinical routine, its complex and time consuming task. So, the automatic brain tumour image segmentation process is required. In this research, a tumour portion is segmented from the brain image by using Adaptively Regularized Kernel-Based Fuzzy C-Means (ARKFCM) algorithm. The input images are resized like 256×256 in the pre - processing stage. The pre-processed MRI image segmented by ARKFCM, which is a flexible high-level Machine Learning (ML) technique to locate the object in a complex template. Next, Hybrid Feature Extraction (HFE) performed on the segmented image to improve the feature subsets. The feature selection process was performed by Kernel Nearest Neighbour (KNN) based Genetic Algorithm (GA) in order to acquire the best feature values. The best feature values given to the Deep Neural Network (DNN) classifier as an input, which is classified into Meningioma, Glioma and Pituitary regions in the MRI images. The performance of proposed ARKFCM - HFE - DNN method is validated by T1-WCEMRI dataset. The experimental outcome showed that the ARKFCM - HFE - DNN method improved the average classification accuracy up to 6.19 % than existing classification techniques such as Convolution Neural Network (CNN) and Support Vector Machine (SVM).

232

Authors: Bellam Purushotham, Choda Durga Venkata Subba Rao

Ad hoc networks are the infrastructure-less networks that are easily deployable and self-configuring, with each constituent node performing like a transceiver. The nodes are free to move subjectively in Mobile ad-hoc networks (MANET). Thus, the network’s wireless topology may be haphazard and may alter rapidly. The main important aspect to be handled is choosing a proper path for transmitting data in case of Link Failure (LF). The main concern of the research work is maintaining the path stability in the occurrence of LF in the network. Ant Colony Optimization (ACO) is used for optimizing the Routing Protocols (RP) such as Ad-hoc on demand Distance Vector (AODV). The Prediction of LF (PLF) mechanism is used for choosing the stable/proper path in the occurrence of node LF in the maintenance phase of distance vector RP based on Neighbour Node (NN). Hence, AODV-ACO-PLF is used for efficient transmission of data packets in the ad-hoc networks by maintaining stable path in-case of LF. The path stability is also maintained in presence of LF in the networks by choosing a different path using PLF Prediction Mechanism. The “AODV-ACO-PLF” methodology is used for finding the ideal way to communicate with the source and destination by giving better results in End-to end delay (6%), drop (5%), packet delivery ratio (7%), and energy consumption (7.5%) than existing methodology such as Path Selection Algorithm based on Hamming Distance and delay parameters (PSA-HD) and Predict Failure in AODV (PF-AODV) methodology in MANET.

249

Authors: Panya Jirajaracheep, Tanairat Mata, Pisit Boonsrimuang

In this paper, the overlapping Partial Transmit Sequence (overlapping-PTS) with Artificial Bee Colony (ABC) phase optimization method is proposed to reduce the larger Peak-to-Average Power Ratio (PAPR) which is the major drawback of filter bank multicarrier with offset quadrature amplitude modulation (FBMC-OQAM) system. In this proposed method, we clarify the problem of larger PAPR by reducing the PAPR of each overlapping data symbol with lower combination of the optimal phase factors. Unlike the conventional PTS methods, the proposed method uses the lower combination of phase factors by applying the ABC algorithm with considering the overlap between the current data symbol and the past data symbol to search the optimal phase combination. The proposed overlapping-PTS with ABC phase optimization method can significantly reduce the larger PAPR with low computational complexity for the FBMC-OQAM system which can be verified by the computer simulation results. From the computer simulation results, it can be confirmed that the proposed method shows much better PAPR which can reduce the PAPR by approximately 1.9 dB at CCDF 10-1 as comparing with the conventional PTS method and can perform much lower computational complexity than the conventional methods which are only 4.38% and 0.76% for S phase factor pattern =300 and 50 respectively.

267

Authors: Muthuselvi Gomathinayagam, Saravanan Balasubramanian

Demand Side Management (DSM) provides a better solution in order to manage increased electricity demand in the power system network. The DSM program relieves the stress on the electrical network for maintaining power system reliability during peak hours. This work proposes a new scheduling approach based on an Adaptive Differential Evolution algorithm (ADEA) by considering a new recombination probability factor (CP) and real mutation factor (F) for analysis. This proposed method is analyzed for industrial, commercial and residential network loads. The main aim of this demand control technique is to reduce the difference between the target curve and total load consumption curve. This paper provides a better solution compared to other algorithms like Evolutionary Approach (EA) and Symbiotic Organism Search (SOS) to reduce the peak load and electricity cost in industrial, commercial and residential distribution networks. The proposed method gives 8.2 % peak load reduction compared to EA for residential area and 2.28% reduction in peak load when compared to SOS for commercial sector. Also it reduces the electricity cost at 17.25%, 20.76% and 21.15 % for residential, commercial and industrial sectors when compared to without DSM. The participation factor of a consumer may be increased by price factor and the less violation in their scheduled demand. The described concept provides a relatively accurate fit to the target curve after load shifting. This concept will increase consumer participation in the DR Program.

287

Authors: Archana Nagelli, Ramesh Ragala, Badarudeen Saleena

Nowadays, the Extreme learning machine (ELM) is playing a key role in machine intelligence and big data analytics due to its various advantages such as fast training rate, universal classification/regression and the capability of approximation. The standard ELM uses the Moore–Penrose generalized pseudo-inverse for solving the hidden layer activation matrix and also it identifies the output weights. Because of that, the standard ELM takes more time to train the features from the dataset. In ELM, scalability also considered as a one of the major concern while processing the large dataset. In order to overcome this concern, the Automotive Rank based ELM (AR-ELM) is proposed to obtain an effective tensor decomposition for diminishing the training time. Besides, the Bayesian approach is considered in this AR-ELM to remove the redundancy from the decomposed samples of the tensor. The major objective of this proposed AR-ELM is to process the large amount of dataset without depending on memory capacities. The recognition accuracy is improved by eliminating redundant information. The key idea of the ARELM is to reduce the training time while processing the huge dataset. The implementation and simulation of the ARELM is done in Spark Python 3.7. The performance of the AR-ELM is analysed in terms of accuracy, precision, recall and training time. The proposed methodology is compared with three existing methodologies such as basic ELM, ELM-TUCKER and ELM-PARAFAC. The recognition accuracy of the AR-ELM methodology with Hardlims activation function is 0.8879 for letter recognition dataset, it is high when compared to the basic ELM, ELM-TUCKER and ELM-PARAFAC that are 0.8102, 0.8375 and 0.834 respectively.