IJCTA-Volume 8 Issue 2 / March-April 2017
Title/Author Name
Page No
Performance Analysis of Neural Networks for Intrusion Detection System
-Pinal J. Patel,Dr.J.S.Shah,Jinul Patel
Existence of an Intelligent Intrusion Detection System relies only on availability of an effective dataset. A dataset with a sufficient amount of quality data can help to train an Intelligent Intrusion Detection System. The goal of this work is to introduce different methods of Neural Network for Intrusion Detection. Different types of Neural Network are implemented to check their performance on KDD’99 intrusion Dataset. We have train dataset using different training functions also. We used MATLAB software to train and test dataset using neural network
Keywords: Neural Network, Intrusion Detection, Attack
LFoC-Interpretability of Linguistic Rule Based Systems and its Applications To Solve Regression Problems
-Cat Ho Nguyen,Thai Son Tran,Van Thong Hoang,Van Long Nguyen
The low-interpretability of fuzzy rule based systems (FRBSs) is required to based on the inherent semantics a word-set of a variable, written shortly as LFoC. LFoCs are fundamental semantic basis and the interpretability of their fuzzy representation is examined based on how an interpretation-assignment of fuzzy sets to their words can convey discovered key properties of LFoCs to their fuzzy set representation. The study proposes two additional constraints imposed on the desired interpretation to maintain discovered features of LFoCs saying that the word-semantics of a variable should be defined in the context of the entire variable-domain, which implies that the fuzzy set semantics of the words of LFoCs must be unchanged when they are enlarged. To show the role of this interpretability of their trapezoid fuzzy set representations, the performance of FRBSs interpretably designed in the new sense to solve regression problems is examined based on a comparative simulation study.
Keywords: fuzzy rule based systems, order-based semantics of words, hedge algebras, low-interpretability, regression problems
Open source tools: empowered the e-learning pedagogy in distance education
-Dr.Mohammed Alamgeer,Iqrar Ahmad
E-learning is defined as pedagogy empowered by digital technology which involves learning done at a computer, usually connected to a network, giving us the opportunity to learn almost anytime, anywhere. Today, E-learning allows to share and manage knowledge and skills of the professionals and to get the right information to the right people whenever they need it.. There are many software systems available that provide distance learning environment. This software is in both forms, commercial and open source software (OSS). Moodle is one of the Open source tool that have been increasingly gaining worldwide popularity in e-learning system. This paper is highlights, how the open source tools empowered e-learning. In turn, the popularity of open source tools day by day increases. A huge number of Students are linked and gaining degree in feasible mode.
Prevention of Collusion Attack By Implementing Dynamic Hash Table in Cloud
-Venkata Aditya Chintala,Tulasi Krishna Doradla,TYJ Naga Malleswari
The data in the cloud is encrypted so as to secure the data and protect the privacy of the user. So as to access the encrypted data by the users who are authenticated, a practical management system for the group keys is imperative for the data sharing presumably in the cloud. The current management mechanisms for the sharing of the keys of the group presume that the server is trusted. The cloud storage module system is a very mammoth scale and it is also an open application, the user group in the current systems is also dynamic, so that’s why the server cannot be trusted. Hence to preserve the privacy of the user AES, a simple Symmetric Encryption algorithm is used. Also, the existing systems have to compute the private key each time a user is revoked; To solve this problem the usage of Dynamic Hash Table is proposed. It also avoids collusion attack and phishing attack. Also, RSA Tokenizer is used to generate tokens to verify the authenticity of the user which provide additional security for the users.
Keywords—Dynamic Hash Table(DHT), Cloud Computing, Anti-collusion attack, Tokeniser.
A New Intrusion Detection and Prevention System for Ad-hoc network Using AODV Protocol
-Subhajit Rauth,Yashwanth Chowdary,R.Brindha
Due to the accelerated advancement of wireless ad hoc Networks in terms of limited power and economical data‐relaying has ďeen partially aĐhievedďeĐause of the accelerated progress in radio transceiver designs and integrated circuits and technology. Due to this, the wireless devices are able to cluster information, process them if required and send them to the next device. The resource strained ad hoc wireless network is functional yet exposed to many attacks. The communication Framework with limited networks may connect with the delicate data in the inimical environment where the nodes may fail and new nodes may join the network, which may lead to the awareness to many kinds of security attacks. An intruder can snoop on all the messages within the transmission area, by operating in dissipated mode. So, it is crucial that the Security of the network routing from the attacker for the wireless ad hoc network must be approve for important missions. There are many devastating attacks, predominant in now a days are wormhole attack, tampering of data, Selective Forwarding, Sybil Attack, Hello Flooding Attack. In this paper, the Intrusion made by the attacker in ad hoc networks has been implemented and also the number of sentinel nodes has been determined and achieved. Operation of the sentinel nodes is like local inter-node collaborative data merge and decision merge to detect, isolate and prevent any further attacks is to be achieved. Simulations has been performed under disticnt situation and from the results of simulation we have noticed that our scheme is capable of providing the security in resource strained wireless ad hoc networks
Emotion Detection through Tweets
-Shivang Arora, Sumant Mishra
Twitter is a popular micro blogging service where users create status messages or small text-based Web posts called tweets. Twitter currently receives about 190 million tweets a day, in which people share their comments regarding a wide range of topics. A large number of tweets include opinions about products and services. Analyzing these tweets to extract opinions or sentiments help us determine the student behaviour. This project is aimed at building a sentiment analyzer tool for analyzing tweets which can be used to accomplish the above goal of determining the student behaviour.This project mainly focuses on classifying the tweets as to belong to student one of positive, negative or neutral category using pre classified tweets as training data. We have used the Naïve Bayes algorithm for implementing the sentiment analyzer tool. The Sentiment analyzer tool developed can give an approximate estimation of the success of student . The tool has been developed using java programming language for the back end and JSP and HTML for the front end and the tool has been presented as a web application.The application retrieves tweets real time based on the user’s query, analyzes and classifies them as to belonging to one of positive, negative or neutral category, summarizes the result and presents the result in a format such as a pie chart, graph which in turn helps in determining the popularity of a student. The algorithm’s efficiency is mainly dependent on the quality of the training data, for the training data chosen for this project we obtained an accuracy of close to 42% with precision and recall standing out at 45.65% and 67.74% respectively.
An Efficient Design for a Ranked Search of Multi-keywords in Encrypted Cloud
-Shrey Chauhan,Akhil Pathak,D. Vanusha
The data which is sensitive has an imperative need to be encrypted so as to avoid pruning eyes, which antiquate data driven algorithms such as keyword-based retrieval of required documents. The Secure ranked search of multi-keywords scheme which is implemented on the cloud data which is encrypted, which synchronously has the ability to support update operations such as insertion, deletion dynamically of documents. The kNN algorithm which is secure has been used to as to encrypt various vectors such as the index vector and query vector, on top of this they also ensure calculation of scores between index vectors and query vectors which is encrypted. For the purpose of protecting the vulnerabilities of statistical attacks, certain phantom terms are necessarily added to the existing index vector for the purpose of blinding of the search results. Since we have used a special structure of a tree-based indexing, this design can accomplish the efficiency of linear search and it can additionally deal with insertion and deletion of the required documents in a really flexibly and dynamic way. Comprehensive experimental conditions have been simulation to denote the competence of this particular scheme.
Interference Mitigation Approach Towards 5G network
-Ekta Srivastava,Dr A V Kulkarni

In the recent years, the demand for higher data rates has been continuously growing to satisfy consumers’ desire for a faster, safer, and smarter wireless network. Since current wireless systems are facing a bottleneck in spectrum resources which make it difficult to enhance performance in the limited available bandwidth. To overcome this problem new Generation of mobile communication known as fifth generation (5G) comes into the picture. The goals of 5G network design are to achieve not only large network capacity, but also ultra-low latency and heterogeneous device support. 5G adopt multi-tier architecture where several low-power Base Stations (BSs) inside small cell are deployed within the coverage area of the macrocell. However, interference by the simultaneous usage of the same spectrum in these cells creates severe problems which reduce system throughput and network capacity. Hence, the resource management is an integral part of 5G HetNets so that interference between different devices can be minimized. The main aim of this work is comprehensive study about the different Interference mitigation approaches in 5G network.
Keywords— 5G cellular wireless, Interference management, HetNets, multi-tier networks, Heterogeneous Networks, co-tier interference, cross- tier interference

Analysis of Feedforward and Recurrent Neural network Algorithms in Predicting the Significant Wave height from the Data Sets of Moored Buoys in the Bay of Bengal
In meteorological and marine engineering applications prediction of Significant Wave Height (SWH) plays a major role for forecasting cyclones, earthquakes & tsunamis that may occur in the ocean and warn the society for appropriate action. Recently, researchers are exploring the use of soft computing techniques to predict SWH. In this work, the wind and wave data obtained from moored buoys of Bay of Bengal is used to predict the SWH using Artificial Neural Networks (ANN). The Recurrent and Feed Forward Networks were analyzed using Levenberg Marquardt (LM), Conjugate Gradient (CG) and Bayesian Regularization (BR) algorithms. Results indicate that Recurrent Networks with BR algorithm has higher correlation and less error.
Index Terms : Artificial Neural Networks (ANN), Significant Wave Height (SWH), Recurrent Network, Feed Forward Network, Levenberg Marquardt (LM), Conjugate Gradient (CG) and Bayesian Regularization (BR).
Analysis of Web Service Efficiency
-Nirmal Sam,Kavin Parithivel,Mohamed Azeemudeen
Web service standards used nowadays are Extensible Markup Language based and the important technology in communication between heterogeneous applications are over Internet. Thereby selecting an efficient web services among numerous options satisfying client requirements has become a challenging and time consuming road block. The path for the optimal execution of all the user request is done using the Hidden Markov Model (HMM). The results have shown how our proposed methodology can help the user to select the most reliable web service available. Our analysis is about creating a cost effective servicing mechanism for web services, if effectively implemented this concept will reduce the need for network engineers in maintenance of web services. As a result of the parallelism technique used in this analysis significant reduction in RT and increase in composition speed has been observed.
Keywords: Hidden Markov Model (HMM), Extensible Markup Language, Web Services, Service Quality Architecture (SQA)
Challenges and Issues of Graph and Subgraph Isomorphism
-Rachna Somkunwar,Dr.Vinod Moreshwar Vaze
The process of matching graphs and subgraphs has number of issues and challenges. Some of them are related to space and time complexity which reduces the performance of the system. Matching different graph is very critical task for the researcher and to check whether an input graph is a part of the main graph is another difficult task for the researcher. This paper proposes challenges and issues of the graph and subgraph isomorphism The main objective of this paper is how we can solve the problems related to graph and subgraph isomorphism.
Review of Bio-inspired Algorithm in Wireless Sensor Network: ACO, ACO using RSSI and Ant Clustering
-Niharika Sharma,Prof.S.D.Chavan
Biological inspired routing or bio-inspired routing is a new heuristic routing algorithm in wireless sensor network, which is inspired from biological activities of insects. ACO is ants’ inspired routing algorithm ACO, which has the ability to find shortest path and re-establish the new route in the case of route failure. In order to improve the network performance i.e. increase network lifetime and reduce transmission overhead, localization and clustering technique can be used in network. To locate the sensor node in network RSSI localization technique is used, which has an edge over other techniques. Another technique is clustering which groups the similar object in the network can also be used. For a large number of data objects clustering is very useful.
In this paper we have studied about bio-inspired algorithm Ant Colony Optimization (ACO), ACO using localization technique (RSSI) and bio-inspired clustering approach Ant clustering.
Keywords— WSN, ACO, RSSI, ACO using RSSI, Ant Clustering, Clustering
Survey on Wireless Sensor Network Protocol to Solve Rushing Attack
-Sudha Panwar,Sunil Malviya
Wireless sensor network (WSNs) is network formed by a huge amount of sensor nodes which is furnished with a sensor to explore physical incident, such as light, heat, sound, pressure, etc. and to communally transfer their data by the network in the direction of main location. For sensor nodes, the factors that required to be worked upon are limited energy and memory. Wireless sensor networks are required to deliver communication with security necessities like confidentiality, reliability and accessibility. The intent of this paper is to analyze five WSNs protocols DSR, SDSR, SAODV, ARAN and Trust Oriented Secure AODV against Rushing attack in wireless sensor network. This paper concentrates estimate the performance of WSNs protocols when Rushing attacks involve in wireless network
Index Terms— Wireless sensor network (WSNs), DoS Attack. Rushing Attack, Routing Protocol.
Ensembling Classifiers for Detecting User’s Intentions behind Web Queries
-Ritesh Kumar,Aniket Srivastav,P. Mahalakshmi
Clients input their solicitations by entering a short succession of question terms, which are further deciphered via web crawlers keeping in mind the end goal to give pertinent answers .So client didn't get right expectations from the web search tools. This paper uses another approach of k-means bunching calculation .This makes web crawlers enter players in comprehension and naturally recognize the client aims and give the legitimate outcomes auto proficiently settling a huge number of questions.
In this paper we utilizing k-implies bunching and an element rich portrayal for client goals distinguishing proof its used to cases are then used to consequently classify new questions by means of correct terms coordinating .Its perform managed learning is a machine learning errand of construing a capacity from marked preparing information from the client expectations.
Keywords – Ensemble of classifiers, Query Interpretations, Heuristic Patterns, k-means clustering, feature-rich representation, Ranking and Listing.
Analysis of Neural Network based Integration of e-Commerce and Social Media
-Kanumuri ravi teja,Leela,veera,surya,madhu,Ushasukhanya.S
The enlarging refinement of long range social correspondence is giving new opportunities to relationship in electronic exchange. It is making recalling the bona fide objective to get abilities to customer correspondences to fulfil more basic budgetary regard. This representation is proposed as social business. The rising of Web 2.0 change and the impact of long range agreeable correspondence goals in like way propose the possible results and conceivable consequences of enduring social exchange E-business. This paper game-plans to take a gander at the troubles related with social business and strategies to have a beneficial choice. A couple related and especially depicted structures, advice and aides to bankrupt down to portray a general control to build up the accomplishment change of social exchange choice in E-business. Systematic Literature Review (SLR) to clear up probability of social business.. For Applying a consider examination to these papers, it was possible to gather the present confirmation about the E business and review some open troubles.
Keywords: E -Commerce, Social media, online shopping, E-business
Keyword Search in Multi- Dimensional Data Sets
-Gopagoni Sai Santosh Kumar,Kamujula Srikar,S.Ushasukhanya
Keyword query in multi-dimensional datasets is a noteworthy application in information mining .In existing framework, an algorithm named Promish in record of calculating sporadic projections and hashing the calculation is actualized. This finds a perfect subset of arrangements and actualizes PromishA which looks for close perfect results with higher effectiveness. Outcome of this implementation demonstrates that Promish is quicker when compared to other techniques and it has different number of orders of magnitude performance improvement .An added advantage is that our techniques can match with both real & synthetic data sets. The present works concentrate on the sort of inquiries where the co-ordinates are known. Regardless of the way that it can make k limits of its which is almost similar to k esteem in nearest keyword request, these changes has no effect on their execution. So we implemented the cluster calculation by using sql queries techniques hich are used to search nearest neighbour for nearest keyword based search. These techniques can be used to predict the nearest keyword set search.
KEYWORDS:Keyword search ,Knn-Algorithm , Sql server ,Ranking Algorithm.
Prisoner Localization and Escape Prevention Inside the Prison
-Sivaraja B,Mukhil Bharathi R D,Purushothaman D,Kavitha M
Existing researches on location tracking focus either entirely on indoor or entirely on outdoor by using different devices and techniques. Several solutions have been proposed to adopt a single location sensing technology that fits in both situations. This paper aims to track a user position in both indoor and outdoor environments by using a single wireless device with minimal tracking error. RSSI (Received Signal Strength Indication) technique together with enhancement algorithms is proposed to cater this solution. The proposed RSSI based tracking technique is divided into two main phases, namely the calibration of RSSI coefficients (deterministic phase) and the distance along with position estimation of user location by iterative trilateration (probabilistic phase). A low complexity RSSI smoothing algorithm is implemented to minimize the dynamic fluctuation of radio signal received from each reference node when the target node is moving. Experiment measurements are carried out to analyze thesensitivity of RSSI. The results reveal the feasibility of these algorithms in designing a more accurate real-time position monitoring system.
KEYWORDS:Global robot localization, Kalman filtering,radio frequency identification (RFID) technology, redundantsensors
Maximizing Throughput of Cognitive Radio Networks through Secondary User Power Consumption
-Shanmugam G,Mohan Prasath M S,Nirmal Sam
A cognitive radio outline requires the QoS of the primary coil winding substance an exploiter maintained while the spectrum allocated to the primary exploiter is used by the Secondary exploiter. The primary user in such cognitive radio system may use a queue to computer memory package boat during transmittance. In such a system the stableness of the queue has to be ensured. This sharing of available spectrum may happen through any of the style s of process such as articulation interweave-underlay meant fashion or only-interweave mode. In reefer interweave-underlay mode, secondary coil user transmits when primary user is absent and transmits adaptively when the primary user is present tense. In the Spiff Interweave-underlay mode, the role of the intercession restraint on the secondary user transmit business leader is of vital importance. Hence in this project, we use the technique of adaptive packet public exposure in meter to increment achievable packet throughput along with interference constraint on secondary user transmit power. We present the method playacting acting and provide the pretense results and analyze them for different duct conditions.
Reduction In Energy Consumption and HandoverDelay with SDN Concept
-M Karthikeyan,Deepak Singh,Divyanshu Agrawal
Cellular networks lead to lot of traffic, Mobile traffic do have a major component and what we call as Wi-Fi. Many carrier grade Wi-Fi have been employed, still Wi-Fi networks do have a huge amount of arrears, such as supporting seamless handover between APs, automatic network access and unified authentication, etc. through project, and SWN which is software based network is proposed by us. Partition of control and data plane is the highlight of SDN. Global view analyzed by control plane, thus it uses NAS to identifies a network state, and bundles the perceived information and network management operations into northbound Application Programming Interface (API) for upper applications. In the data plane, we construct software access point (SAP) is constructed to abstract the connection between user equipment (UE) and access point (AP). These APIs and the SAP abstraction are used by Network operators which can design network applications to manage and configure the whole network, thus adding flexible, user-friendly, and scalable to the Wi-Fi networks.
Keywords—SDN,OpenDaylight,Relay,LATEX,paper, template
News Aggregation Social Bot using Data Mining
-Aditya Sanjay Ekbote,Asst.Prof. A.Murugan,Rahul Garg
The purpose of the system is to showcase the tech-nology developed for building a fully automated real-time news provider. It mainly focuses on using data extraction, data mining and article summarizing algorithm to build the product. The main target of the research is to develop a fully automated platform which can provide news on latest trending topics on social media without wasting much time on different platforms and reading whole article.
Keywords: Text Mining, Web parsing, Text Summarization, News Aggregation
IJCTA © Copyrights 2015| All Rights Reserved.

This work is licensed under a Creative Commons Attribution 2.5 India License.