International Journal of Computer Techniques

6-9

Authors: - IJCT K.Ravikumar, P.Gandhimathi

The spam mails are used by spammers to pilfer the data of the users and organizations online. Quick growth rate of the use of the internet has increased the spam mails. There are several methods employed for filtering spam. Most of the spam filtering techniques are based on objective methods such as the content filtering and DNS/reverse DNS checks. Recently, some cooperative subjective spam filtering techniques are proposed. Objective methods suffer from the false positive and false negative classification. Objective methods based on the content filtering are time con-suming and resource demanding. They are inaccurate and require continuous update to cope with newly invented spammer’s tricks. On the other side, the existing subjective proposals have some drawbacks like the attacks from mali-cious users that make them unreliable and the privacy. In this paper, we propose an efficient spam filtering system that is based on a smart cooperative subjective technique for content filtering in addition to the fastest and the most reliable non-content-based objective methods. The growing use of internet has promoted an easy and fast way of ecommunication. The well-known example for this is E-MAIL. The increasing use of e-mail and the growing trend of Internet users sending unsolicited bulk e-mail, the need for an antispam filtering or have created, Filter large poster have been produced in this area, each with its own method and some parameters are to recognize spam.

14-18

Authors: - IJCT S. Alexander Suresh

Merchants selling products on the Web often ask their customers to review the products that they have purchased and the associated services. As e-commerce is becoming more and more popular, the number of customer reviews that a product receives grows rapidly. For a popular product, the number of reviews can be in hundreds or even thousands. This makes it difficult for a potential customer to read them to make an informed decision on whether to purchase the product. It also makes it difficult for the manufacturer of the product to keep track and to manage customer opinions. For the manufacturer, there are additional difficulties because many merchant sites may sell the same product and the manufacturer normally produces many kinds of products. In this research, we aim to mine and to summarize all the customer reviews of a product. This summarization task is different from traditional text summarization because we only mine the features of the product on which the customers have expressed their opinions and whether the opinions are positive or negative. We do not summarize the reviews by selecting a subset or rewrite some of the original sentences from the reviews to capture the main points as in the classic text summarization. Our task is performed in three steps: (1) mining product features that have been commented on by customers; (2) identifying opinion sentences in each review and deciding whether each opinion sentence is positive or negative; (3) summarizing the results. This paper proposes several novel techniques to perform these tasks. Our experimental results using reviews of a number of products sold online demonstrate the effectiveness of the techniques

31-37

Authors: - IJCT Jaybhaye Chaitali, Dr. P. M. Pawar

Wireless Sensor network is formed of sensor nodes, placed in defy and hostile environment. Many important fields are dependent on WSN for data gathering from human unreachable areas. Sensor nodes have restricted amount of resources like memory, battery. Data aggregation helps to reduce energy usage by minimizing data transmission and removing data redundancy. Research proposed Multiobjective based optimization for secure data aggregation. Multiobjective optimization is used for cluster level aggregation. The research proposed watermarking based mechanism for securing data Watermarking provides data confidentiality and data integrity with less computational cost. Results of the SDAMMO are compared with the MH-EESDA system. Results show that the SDAMMO system is more efficient than the MH-EESDA

45-47

Authors: - IJCT A.K.Mariselvam , A.Akhil , S.Vishnumohan Reddy , B.Rishvita Reddy

A number of methods have been proposed for recognition of a face over the past few decades. Some of them have shown promising results but they work only in well controlled condition. In this method, the face patch and the gallery image are brought to the same size before finding the similarities between the images. In our real world scenarios, a lot of faces get covered by other objects like scarfs, sunglasses, another person face in front of us, the angle between the camera and the person etc., not being able to recognise the persons face. To overcome such cases a method should be developed to identify the persons face. A gallery image and a face patch is taken, first we detect their keypoints and extract their textual features and geometric features. Then we proceed with the robust point set matching where we match keypoints with the textual features and the geometrical features of both the gallery image and the face patch image. Finally, the distance between the features is matched and the face is recognised

55-61

Authors: - IJCT Suganya Devi D, Lavanya S

Wireless Sensor Network (WSNs)provides a new model for sensing and spreading information from different environments, with the possibility to serve numerous and various applications. WSN is prominent area of research nowadays, because of the potential utilization of sensor networks in different applications. Wireless sensor network is a collection of various small micro-electro-mechanical devices. These small devices have sensors, capability of computation, supply of power and the wireless transmitter and receiver. A body area network (BAN), also referred to as a wireless body are network(WBAN) or a body sensor network (BSN), is a wireless network of wearable computing devices. BAN devices may be embedded inside the body, implants, may be surface-mounted on the body in a fixed position Wearable technology or may be accompanied devices which humans can carry in different positions, in clothes pockets, by hand or in various bags. The proposed process focus on Wireless body area sensors are used to monitor human health with energy constraints. Different energy efficient routing schemes are used to forward data from body sensors to medical server. It is important that sensed data of patient reliably forward to medical specialist or server for further analysis. To increase the throughput and reliable communication between sensors and sink, we propose a new scheme. Main advantages of our proposed protocol scheme are to achieve a longer stability time. Nodes stay alive for longer period and consume less energy. The large stability period and less energy consumption of nodes, contribute to a high throughput.

71-78

Authors: - IJCT Ms. Lavanya M, Mrs. Shoba S.A.,

In the past, many researchers used data mining techniques in any area. A lot of amounts of data have been collected from scientific domains such as geo sciences, astronomy, meteorology, geology and biological sciences. Data mining techniques and tools used by researchers in biological and environmental problems also. In biological science data mining used in sequences alignment is based on the fact that all living organisms are related by evolution and in environmental science data mining used in predicting data such as earthquakes and landslide etc. This paper highlights on the wide survey of protein sequences, (DNA, RNA) sequences, cancer prediction, relational and semantic data mining for biomedical research area. Health care data, multiagent framework for bio data mining, predicting earthquakes, landslide and spatial data in distributed data mining alg orithms and tools. This is including in bioinformatics and environmental studies also.

88-95

Authors: - IJCTMs. Ramya P., Saravanan A. M.

Networks are getting larger and more complex, yet administrators rely on rudimentary tools such as ping and trace route to debug problems. ATPG propose an automated and systematic approach for testing and debugging networks called “Automatic Test Packet Generation” (ATPG). ATPG reads router configurations and generates a device-independent model. The model is used to generate a minimum set of test packets to minimally exercise every link in the network or maximally exercise every rule in the network. Test packets are sent periodically, and detected failures trigger a separate mechanism to localize the fault. ATPG can detect both functional and performance problems . ATPG complements but goes beyond earlier work in static checking which cannot detect liveness or performance faults or fault localization which only localize faults given liveness results. ATPG describe our prototype implementation and results on two real-world data sets: Stanford University’s backbone network and Internet2. We find that a small number of test packets suffice to test all rules in these networks. ATPG code and the data sets are publicly available.

104-110

Authors: - IJCT B.Sudha

In Cloud Computing models, Virtual Machines (VM) can be dynamically provisioned according to demand and released when not needed. Efficient Virtual Machine (VM) provisioning and allocation allows the cloud providers to effectively utilize their available resources and obtain higher profits. The existing static VM provisioning may not guarantee economically efficient allocation and thus cannot guarantee maximum revenue for cloud providers. A better solution would be to take into account the users' demand and dynamically provision VM instances. In the recent times, the cloud providers have introduced auction-based models for VM provisioning and allocation which allow users to submit bids for their requested VMs. MM Nejad, L. Mashayekhy and D Grosu have formulated a dynamic VM provisioning and allocation problem for the auction-based model as an integer program considering multiple types of resources. They have designed truthful greedy and optimal mechanisms for the problem such that the cloud provider provisions VMs based on the requests of the winning users and determines their payments. Since Virtual Machines are created by Hypervisors in the underlying physical machines and this would result in overhead and ultimately would impact the overall performance of the cloud. We propose that this overhead could be factored inthe design of Truthful Greedy mechanisms. It is hoped that our proposed idea can achieve better results in terms of revenue for the cloud provider.

115-120

Authors: - IJCT Hayder K. Fatlawi

Cervical cancer threatens the lives of many women in our world today. In 2014, the number of women infected with this disease in the United States was 12,578, of which 4,115 died, with a death rate of nearly 32%. Cancer data, including cervical cancer datasets, represent a significant challenge data mining techniques because absence of different costs for error cases. The proposed model present a cost sensitive classifiers that has three main stages; the first stage is prepressing the original data to prepare it for classification model which is build based on decision tree classifier with cost selectivity and finallyevaluation the proposed model based on many metrics in addition to apply a cross validation.The proposed model provides more accurate result in both binary class and multi class classification. It has a TP rate (0.429) comparing with (0.160) for typical decision tree in binary class task

130-135

Authors: - IJCT L.Dhivyajayadharshini

Conventional approaches to visual search re-ranking empirically take the “classification performance” as the optimization objective, in which each visual document is determined whether relevant or not, followed by a process of increasing the order of relevant documents. In this project, we first reestablish the fact that: the classification performance fails to produce a globally optimal ranked list. Hence, we formulate re-ranking as an optimization problem, in which a ranked list is globally optimal only if any arbitrary two documents in the list are correctly ranked in terms of relevance. This is different from existing Approaches which simply classify a document as “relevant” or not. To find the optimal ranked list, we convert the individual documents to “document pairs”, Each pair is represented as an “ordinal relation.” Then, we find the optimal document pairs which can maximally preserve the initial rank order while simultaneously keeping the consistency with the auxiliary knowledge mined from query examples and web resources as much as possible. We develop two pair wise re-ranking methods, difference pair wise re-ranking (DP-re-ranking) and exclusion pair wise reranking (EP-re-ranking), to obtain the relevant relation of each document pair. Finally, a round robin criterion is explored to recover the final ranked list. Conventional approaches to visual search re-ranking empirically take the “classification performance” as the optimization objective, in which each visual document is determined whether relevant or not, followed by a process of increasing the order of relevant documents. In this project, we first reestablish the fact that: the classification performance fails to produce a globally optimal ranked list. Hence, we formulate re-ranking as an optimization problem, in which a ranked list is globally optimal only if any arbitrary two documents in the list are correctly ranked in terms of relevance. This is different from existing Approaches which simply classify a document as “relevant” or not. To find the optimal ranked list, we convert the individual documents to “document pairs”, Each pair is represented as an “ordinal relation.” Then, we find the optimal document pairs which can maximally preserve the initial rank order while simultaneously keeping the consistency with the auxiliary knowledge mined from query examples and web resources as much as possible. We develop two pair wise re-ranking methods, difference pair wise re-ranking (DP-re-ranking) and exclusion pair wise reranking (EP-re-ranking), to obtain the relevant relation of each document pair. Finally, a round robin criterion is explored to recover the final ranked list.

137-143

Authors: - IJCT Girish , Dr.Phaneendra H.D

Group signature is a method which allows any genuine group member to sign many number of messages on behalf of the group without giving his identity. Whenever dispute arises, the manager of the group can reveal the identity of the signer. If quantum computer comes in to picture group signatures which are based on the traditional cryptography can be broken easily. In this paper, we propose a new identity based group signature. Our signature scheme is based on bilinear maps. This identity based scheme has the security properties of group signatures. In this scheme the size of the group pubic key and the length of signatures are independent on the size of the group. This scheme can be used for very large groups. Same key pair can be used by the group member to sign many messages

154-158

Authors: - IJCT R.Karthikeyan , Dr.T.Geetha , Shanmugapriya M , Vimala M

The Smart Grid is an electrical power infrastructure that makes intelligent decisions about the state of the electrical power system to maintain a stable environment. It is expected that the smart grid will radically add new functionalities to legacy electrical power systems. However, believe that this will in turn introduce many new security risks. In addition, different protocols that are used in these networks use their own set of security requirements. The public key infrastructure (PKI) is a viable solution; it has some difficulties to satisfy the requirements in availability, privacy preservation, and scalability. To complement the functions of PKI, introduce some novel mechanisms so that those security requirements can be met. In particular, propose a mechanism to efficiently resist Denial-of-Service (DoS) attacks, and some suggestions to the security protocol design for different application categories.