IJCST Logo

International Journal of Computer Science and Technology
Vol 6.3 ver – 1 (July to September 2015)

S.No. Research Topic Paper ID Download
1

Survey on Human Activity Recognition Techniques For Video Surveillance

Utkal Sinha, Himanish Shekhar Das, Mayank Shekhar

Abstract

Activity recognition aims to recognize the actions and goals of one or more agents from a series of observations on the agents’ actions and the environmental conditions. Vision-based activity recognition is a very important and challenging problem to track and understand the behavior of agents through videos taken by various cameras. The primary technique employed is computer vision.
Vision-based activity recognition has found many applications such as human-computer interaction, user interface design, robot learning, and surveillance, among others. This report describes the literature served for developing human activity recognition algorithm and a novel approach to it.
Full Paper

IJCST/63/1/A-0516
2

Merge Key Algorithm to Enhance Security for AODV in MANET

Namrata Awasthi, Anurag Jain

Abstract

Various routing protocols and techniques are being included in wireless network and making it an area for further research. Congestion avoidance and security are the major areas in Wireless routing which are having research focus. Major research in area of security is focused on key based mechanisms or third party trust management systems. Improved Routing Security is being proposed in this work which will provide the routing protocol security using validating a node for identification which is being distributed to each node through protocol. AODV (Adhoc on Demand Vector) routing is a proactive routing protocol which uses the neighbors’ database to find the best route. The work is this paper is focusing on security over routing security and simulations are being proposed to show the improved packet delivery ration, throughput, end to end delay and reduced packet drop rate for Ad hoc On Demand Distance Vector (AODV) routing protocol. Attacks are being avoided proactively by including changes in the basic implementation of AODV routing protocol. Further in this work proposal to provide access control technique and unique key based authentication for AODV is being given. In existing work system performance may affect severely due to application of security mechanisms therefore research scope in this area is always available. Since the industry of communication is growing by leap and bounds therefore the need of continuous research in this area is very much needed.

Full Paper

IJCST/63/1/A-0517
3

Fast Dimensionality Reduction for High Dimensional Datasets Supporting Big Data & Cloud Computing

Sonam Malik, Er. Pooja Narula

Abstract

High-dimensional datasets present many mathematical challenges as well as some opportunities, and generally are bound to offer increase to new theoretical developments. Among the issues with high-dimensional datasets is that, in lots of instances, maybe not all the measured variables are “important” for understanding the underlying phenomena of interest. While certain computationally expensive novel practices can build predictive models with high accuracy from high-dimensional information, it is still of interest in several applications to lessen the dimension of the original data ahead of any modeling of the info. Modern Cloud platforms have steadily increased their use of convoluted, High dimensional data.

Full Paper

IJCST/63/1/A-0518
4

Mining of Text Documents using Side Information with High Efficiency

Ashuta Mishra, Prateek Gupta

Abstract

In web and text documents lot of important information is available within the side information. This side-information are available in documents as provenance information, the links in the document, user-access behavior from web logs, or other non-textual attributes which are embedded into the text document. This side information is used in document clustering of such documents bunch. This information will include noisy information as well; hence it is difficult to use it efficiently. Therefore proper algorithms are required to use such information in document clustering to avoid noise from the extracted clusters. We need a properly ruled way to perform the mining process to optimize the advantages got from using this side information. In this paper, we improve the design of an algorithm which will combine classical partitioning algorithms with probabilistic models in order to create an effective clustering approach. We provide experimental results on a Cora data set to illustrate the advantages of using such an approach.

Full Paper

IJCST/63/1/A-0519
5

Association Rules Based Efficient Web Usage Mining

Mohit Dubey, Reetesh Rai

Abstract

Increasing growth of the Internet has made the thousands of the websites and web user around the world. Increase of web users and websites is also causing increase of huge data per second on the web. This increase is causing users difficult to find the relevant information from the web. Therefore we need to create algorithms for the easy and fast web mining tools for the users. Since it is majorly a responsibility of the website owners to track and provide relevant information to the users as per their usage history therefore web usage mining is particularly focus of this work. Web usage mining aims at discovering useful information or knowledge from usage data registered in log files, based on primary kinds of data used in the mining process. This paper uses a web usage mining technique to fetch knowledge from web server log files where all user navigation history is registered.

Full Paper

IJCST/63/1/A-0520
6

Diagnosis of Hypertension using Adaptive Neuro-Fuzzy Inference System

Rimpy Nohria, Palvinder Singh Mann

Abstract

This research paper presents Adaptive Neuro-fuzzy inference system (ANFIS) technique for diagnosis of hypertension. For training the ANFIS system patient’s data set are collected under the supervision of physician in clinical trials on hypertension patients in hospital. This paper presents the methodology of ANFIS was used to diagnose and presents the comparison of proposed system with existing fuzzy expert system on the basis of performance matrices i.e. Accuracy and Sensitivity. This paper proved that ANFIS system has better performance than existing model. We obtained that the accuracy and sensitivity of proposed model results is 94.63% and 97.50% resp.

Full Paper

IJCST/63/1/A-0521
7

Outlier Detection Using Hybrid Genetic Algorithm and Bacterial Foraging Optimization Algorithm

Dr. Amit Verma, Er. Parminder Kaur, Sharnjeet Kaur

Abstract

Data Mining is the process of extract useful information from the large data by using different mining techniques. Clustering and classification manage the large amount of data into different clusters according to their properties. Sometimes data arranged in clusters contain outliers that degrade the performance of the system. The outliers detected by K-Mean genetic algorithm also contain information; to detect this information and outlier’s properly K-Mean genetic bacterial foraging algorithm is applied. This paper also presents the comparison between K-Mean genetic algorithm and K-Mean genetic bacterial foraging algorithm.

Full Paper

IJCST/63/1/A-0522
8

Popular FOSS Tools, Apps and Data Sources in the Geoinformatics Realm

Harpinder Singh, Amardeep Singh, Dheeraj Gambhir, Sagar Taneja

Abstract

In recent years there has been a sudden increase in the utilization and popularity of free and open source software (FOSS), applications and data sources in the field of geographic information system (GIS) and remote sensing (RS). The tools and functionalities provided by FOSS are comparable to their proprietary peers. In addition to being free of cost these software have a small learning curve and are customizable. Except for some specialized functionalities, FOSS software provide almost all the tools required by a researcher for a general GIS and RS project. Proprietary software, instruments like global positioning system (GPS), camera etc and Satellite data are usually costly and cover a significant part of the project expenses. Large amount of time is also consumed in acquiring these resources. This paper reviews and lists important FOSS software with plug-ins, Android apps and data sources necessary for a researcher or a student to undertake a research work /project with least cost and time.

Full Paper

IJCST/63/1/A-0523
9

Algorithm Design and Comparative Analysis for Outlier Detection using Genetic Algorithm and Bacterial Foraging

Dr. Amit Verma, Er.Parminder Kaur, Sharnjeet Kaur

Abstract

Data mining derive its name for searching necessary information from a large database and utilizes this information in better way. To arrange the data in proper manner there is a need of clustering and classification techniques. With respect to other algorithms K-Mean clustering algorithm attains attention to arrange the data in clusters . The optimization algorithms such as Genetic algorithm and Bacterial Foraging Algorithm used to optimize the result of K-Mean clustering algorithm and detect outliers from the result of K-Mean Clustering algorithm. In this paper a hybrid genetic and bacterial foraging algorithm is proposed to find outliers from the data.

Full Paper

IJCST/63/1/A-0524
10

An Enhanced Database Authentication Techniques Using Batch DSA

P.Prabhakaran, C.Thiyagarajan

Abstract

Multicast is an efficient method to deliver data from a sender to a group of receivers. Authentication is an important issue in multicast communication. Conventional block-based multicast authentication schemes overlook the heterogeneity of receivers by letting the sender choose the block size, divide a multicast stream into blocks, associate each block with a signature, and spread the effect of the signature across all the packets. The approach of signing and verifying each packet independently raises a serious challenge to resource-constrained devices. In mobile environments, the situation is even worse. The instability of wireless channel can cause packet loss very frequently. The smaller data rate of wireless channel increases the congestion possibility. The congestion will lead to packet loss.

Full Paper

IJCST/63/1/A-0525
11

A Constructive Analysis of Time Synchronization in Wireless Sensor Networks

S. Karthik, Dr. A. Ashok Kumar

Abstract

Wireless communication is defined as a ubiquitous environment where millions of people are involved in communication via portable devices. In Wireless Sensor Networks (WSNs), nodes are dispersed at different locations for various network operations. A wireless service imparts much error rates on data transmission and takes more time to generate signals when compared to wired services. Most importantly, transmission of packets to the target node without any collision or damage will be a challenging task in WSN. So, here we discuss about time synchronization which helps us in attaining the successful data transfer at higher rates and imparts higher acknowledgement to the base node using timestamp method. Time Synchronization (TS) is a key in achieving the above mentioned task which could be applied for real time applications in an effective manner. In this paper, we have analyzed few techniques, protocols of time synchronization in WSN.

Full Paper

IJCST/63/1/A-0526
12

Fuzzy Logic based Requirement Prioritization (FLRP) – An Approach

Ruby, Dr. Balkishan

Abstract

Requirement prioritization helps to determine which set of requirements should be implemented in certain release within limited constraints and provide higher customers satisfaction. In this paper a fuzzy logic based requirement prioritization (FLRP) approach is introduced that prioritize the requirements on small, medium and large scale, and take the advantage of both customers and developers preference to prioritize the requirements. The proposed approach prioritizes the requirements on the basis of customer importance, cost, time, risk and fuzzy decision making. Fuzzy decision making helps to solve the uncertainty and negotiation concept in requirement prioritization. The approach is implemented on the tool like Matlab. Paper defines a priority function which helps to determine the priority of requirements using various input parameters.

Full Paper

IJCST/63/1/A-0527
13

Visual Cryptography Schemes for Secret Images Encryption and Decryption

Vandana Shastri, R.S.Shekhawat

Abstract

Visual cryptography is one of the techniques used to encrypt the images by dividing the original image into transparencies. The transparencies can be sent to the intended person, and at the other end the transparencies received person can decrypt the transparencies using the tool, thus gets the original image. The proposed Visual cryptography provides the demonstration to the users to show how encryption and decryption can be done to the images. In this technology, the end user identifies an image, which is not the correct image. The basic principle of the Visual Cryptography Scheme (VCS) was first introduced by Naor and Shamir. VCS is a type of clandestine distribution scheme that focuses on sharing clandestine imagery. The idea of the chart cryptography model proposed in is to split a secret image into two random shares (printed on transparencies) which separately reveals no information about the secret image other than the size of the clandestine picture. The secret picture can be reconstructed by stacking the two share. The fundamental process of this scheme is logical operation OR.

Full Paper

IJCST/63/1/A-0528
14

An Efficient Routing Algorithm to Enhance the Performance of Distributed Mobile Computing Environment Using Cluster

Faizul Navi Khan, Kapil Govil, Prof. R. K. Dwivedi

Abstract

Mobile Computing Environment (MCE) provides the facility to make the information and data available anytime and anywhere to its users by using mobile computing devices (i.e. laptops, Smart phones, notebook and tablets) with the help of wireless network. It provides flexible mode of communication to network users. Transmission between mobile computing devices and mobile network application occurs with the help of routing algorithms. Network routing plays a vital role to enhance the performance of Mobile computing. An efficient routing algorithm is needed to select transmission path to transmit data from one point to another point in Mobile computing environment. Routing algorithms support each other by exchanging services or information in mobile computing environment. This research demonstrate data transmission over the network to execute on available processing nodes with optimize transmission cost with the help of a cluster based routing algorithm from source to destination point. Presented algorithm is also tested in Matlab environment and shown better results over to past algorithms.

Full Paper

IJCST/63/1/A-0529
15

Comparative Study of Hive and Map Reduce to Analyze Big Data

Nisha Bhardwaj, Dr Balkishan, Dr. Anubhav Kumar

Abstract

Big data is the combination of large datasets and the management of this large dataset is very difficult. So, we require some new techniques to handle such huge data. The challenge is to collect or extract the data from multiple sources, process or transform it according to our analytical need and then load it for analysis, this process is known as “Extract, Transform & Load” (ETL). In this research paper, firstly implementation of hadoop in pseudodistributed mode is done and then implement hive on hadoop to analyze the large dataset. In this paper, we consider the data from Book-Crossing dataset and take only BX-Books.csv file from dataset. Over this dataset we perform query by executing hive on command line to calculate the frequency of books which are published each year. Then, comparison of hive code is done with the mapreduce code. And, finally this paper shows that how hive is better than map reduce.

Full Paper

IJCST/63/1/A-0530
16

Big Data Analysis – A Big Approach

Sanjam Singla, Priyam Kaur Sandhu

Abstract

In the term Big Data Analysis, two entities have come together i.e. “Big Data” and “Analysis”. Data is classified as Big Data when its three Vs, namely “Volume”, “Variety” and “Velocity” exceed the capability of the conventional systems. Analysis of Big Data includes collection of different tool types including data mining, Hadoop technology and artificial intelligence and so on. Data in ig Data can contain both structured and unstructured data i.e. very large traditional databases are not capable to process it. In a single dataset, the size of the Big Data varies from dozen terabytes to a lot more petabytes in a single data set. While there are many other methods through which the data can be handled but one of them is Hadoop, although this technology has some drawbacks that will be discussed later i.e. it is not fast, it does not support adhoc queries, etc. This paper gives a brief overview on Big Data, how that Big Data is managed, stored and analyzed in order to find out different patterns and relationship.

Full Paper

IJCST/63/1/A-0531
17

A Hybrid Approach Using GA and ACO for Risk Assessment in Bank Loan Process

Er. Iqbaldeep Kaur, Rajvir Kaur

Abstract

Risk assessment has been one of the major areas of research in banking. The main purpose of assessing credit risk is to differentiate between a potentially default customer and non-default customer. The work presented in this paper focuses on calculating risk assessment value by using Genetic Algorithm and Ant Colony Optimization Algorithm. The modelling of bank loan process that a banking organization follows is done by using BPEL (Business Process Execution Language). The outputs of this method are important for reusability purpose.

Full Paper

IJCST/63/1/A-0532
18

Link Lifetime based Border Node (LLBN) Protocol for Vehicular Ad-Hoc Networks

Manisha Chahal, Sanjay Batish, Sanjeev Sofat, Amardeep Singh

Abstract

Vehicular ad hoc network (VANET) attracts rising attentions of researchers for safety related and other communication applications. These networks have many challenging characteristics such as high mobility, fast changing topology and limitation of bandwidth. We proposed LLBN (link lifetime based border node protocol) by taking advantage of BMFR for bandwidth utilization and link life time to avoid link failure.

Full Paper

IJCST/63/1/A-0533
19

Detailed Review of Image Based Steganographic Techniques

Surbhi Singhal, Rajkumar Singh Rathore

Abstract

Security of the confidential information has almost become a challenge now-a-days because of large amount data being exchanged on the internet. For the protection of these data from unauthorized access various methods have been developed and are in practice today. This paper discusses one of the techniques named Steganography, which is a tool for hiding information inside an image. For hiding secret data in digital images, large varieties of steganographic techniques are available, some are more complex than others, and all of them have their respective pros and cons. This paper tries to give thorough understanding of some recently developed image steganography algorithms. The study shows that steganography has played a very beneficial role in various applications. It would be very useful and will provide a better platform for the beginners who want to work in steganography.

Full Paper

IJCST/63/1/A-0534
20

A Parallel Optimization Approach for Test Suite Minimization

Zeeshan Khan, Prabhat Verma

Abstract

Test case Minimization techniques try to schedule test cases in an execution order according to given criterion. The main purpose of this Minimization is to increase the likelihood that if the test cases are used for regression testing in the given order, they will more closely meet the objective than they would if they were executed in some other order. For instance, testers might schedule test cases in an order that achieves code coverage at the fastest rate possible, exercises features in order of expected frequency of use, or increases the likelihood of detecting faults early in testing. In this work we continue the research direction in favor of parallel processing of test case s in order to minimize the overall test suite execution time.

Full Paper

IJCST/63/1/A-0535
21

Enhanced Intrusion Detection Using Feature Extraction and Adaptive Boost With SVM-RBF Kernel

Maninder Singh, Sanjeev Rao

Abstract

With the quick increment of web innovation, the malevolent exercises on the system are likewise expanding. So the utilization of a productive technique is must to distinguish the intrusion. Security for all systems is turning into a major issue. In this paper we compared the existing machine learning algorithms and proposed a new hybrid approach of classifier which is Adaptive boost with SVM-RBF.

Full Paper

IJCST/63/1/A-0536
22

Efficient XML based Structured Reporting of Bio Medical Images

Deeksha Kanwal, Ranjeeta Kaushik

Abstract

Recently, the digital imaging and communications in medicine (DICOM) standard introduced rules for the encoding, transmission, and storage of the imaging diagnostic report. This medical document can be stored and communicated with the images in picture archiving and communication system (PACS). It is a structured document that contains text with links to other data such as images, waveforms, and spatial or temporal coordinates. Its structure, along with its wide use of coded information, enables the semantic understanding of the data that is essential for the Electronic Healthcare Record deployment. In this work we present DICOM Structured Report (SR) and discuss its benefits. We show how SR enables efficient radiology workflow, improves patient care, optimizes reimbursement, and enhances the radiology ergonomic working conditions. As structured input significantly alters the interpretation process, understanding all its benefits is necessary to support the change.

Full Paper

IJCST/63/1/A-0537
23

A Hybrid Online Genre-based Recommender System

Dr. Amit Verma, Harpreet Kaur Virk

Abstract

Recommendation is a sort of web intelligence technique used for filtering information to provide relevant data to the people on daily basis. There are various approaches used for recommendation. In content-based filtering, the systems examine items previously chosen by the actual user, whereas in collaborative filtering, recommendations are based on the information of similar users or items. The recommendations are also influenced by the factors such as age, gender and some other user profile information. In this paper, both content and collaborative techniques along with some demographic information are combined into a hybrid approach, where additional content features are used to improve the accuracy of collaborative filtering. And the genetic algorithm along with kNN approach and correlation is used to provide recommendations to the user. The system is evaluated using precision and recall parameters.

Full Paper

IJCST/63/1/A-0538
24

Indian Sign Language Animation Generation System for Gurumukhi Script

Dr. Amit Verma, Sandeep Kaur

Abstract

Sign language is used in deaf and dumb communities to communicate with each other. Deaf and dumb people use sign language as their mother tongue. Research work on Indian sign language is very limited. The main reason behind lack of Research is that there is no proper grammar of ISL. Another reason is that Sign language is not universal; it varies from country to country and region to region. This paper presents a system which generates animation corresponding to inputted Punjabi word. For generating animation, system covers all basic words used in daily routine. This system generates animation corresponding to 200 Punjabi words. Firstly, HamNoSys corresponding to 200 words are generated. These notations are generated according to signs used in India. To check the accuracy of these notations, JA SIGML PLAYERAPP’s are used. Accuracy of the animated signs is tested with the help of “Indian Sign Language Dictionary” and results are very encouraging.

Full Paper

IJCST/63/1/A-0539
25

A Hybrid Recommender System for Performance Improvement using Personal Propensity

Dr. Amit Verma, Harpreet Kaur Virk

Abstract

Recommendations basically are the suggestions that one experience in day to day life. Actually we all depend on suggestions from others to do our daily work for example choosing dress to wear we ask others for suggestions. In selection process such as filtering, the systems select items previously chosen by the actual or true user, whereas in collaborative filtering or selection , recommendations are based on the data or entity of similar users or items. The recommendations are also influenced by the various factors such as age, gender and some other user profile information. In this paper, Hybrid Recommender System for Performance Improvement using Personal Propensity is presented . The Mutate and standard deviation are the key component used in this paper. The system is evaluated using various parameter like precision and recall.

Full Paper

IJCST/63/1/A-0540
26

Review and Algorithm Design for Content based Classification Using Multilayer Perceptron

Dr. Amit Verma, Shruti Mittal, Arpna Dhingra

Abstract

Spam is any junk message or fake message sent by spammers to lure the legitimate users. Spam classification in Twitter has been performed by using various techniques like Naïve Bayes, Support Vector Machine, URL analysis etc. In this research, classification for various tweets have been performed as spam and fam or non-spam. This research first extracts the live tweets and then preprocess them to refine the tweets. Then, features are extracted and Multilayer Perceptron Algorithm is applied.

Full Paper

IJCST/63/1/A-0541
27

A Hybrid Recommender System using Genetic Algorithm and kNN Approach

Dr. Amit Verma, Harpreet Kaur Virk

Abstract

Recommender Systems are now popular both commercially and in the research community, where various approaches have been adopted and validated on large scale. In content-based filtering, the systems examine items previously chosen by the actual user, whereas in collaborative filtering, recommendations are based on the information of similar users or items. We also analyzed that recommendations are also influenced by the factors such as age, gender and some other user profile information. In our work both content and collaborative techniques and some demographic information are combined into a hybrid approach, where additional content features are used to improve the accuracy of collaborative filtering. Also we are using the genetic algorithm and k-NN algorithm to provide recommendations to the user. To evaluate precision, recall and F1-measure performance parameters are used.

Full Paper

IJCST/63/1/A-0542