Vol 3 Issue 4 October 2015-December 2015

Abstract: Many security primitives are based on hard mathematical problems. Using hard AI problems for security is emerging as an exciting new paradigm, but has been underexplored. Similarly due to rapid growth of internet and multimedia systems now it is easier to copyright the multimedia documents like audio, video, text, images etc. Video piracy has become an increasing problem particularly with the proliferation of media sharing through advancement of Internet services. Since digital video can be easily and perfectly duplicated and illegally distorted, appropriate schemes are needed to protect the rights of content owners or the integrity of the surveillance video. Now days there are many techniques available for providing security to multimedia documents. Video watermarking is an important emerging technique for these issues.  In this paper, we present a new security primitive based on hard AI problems, namely, a novel family of graphical password systems built on top of Captcha technology, which we call Captcha as graphical passwords (CaRP) .In the system we also propose scene change detection (SCD) watermarking algorithm which is the most convenient and efficient method for copyright protection of video. This method is more robust to withstand with different types of video attacks like frame dropping, lossy compression. Here combination of both the Schemes is implemented so as to give enhanced security to the system. This makes our system robust against all type of attacks

Keywords: CaRP, SCD, Visual cryptography, watermark, image security.

Title: Enhanced Security in Online Database System Using Visual Cryptography and Water Marking

Author: Divya Sahay, Monalisa Merchant, Suleiman Sheikh, Rohan Shukla, Shubhangi Suryavanshi

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: Web services provide new types of services leading to an increasing number of services emerging on internet in service and cloud computing. As a result, service-relevant data has become too huge to effectively process and users are finding difficulty to utilize services that match their mash ups. To overcome this challenge technique used is to filter and recognize the similar services under clusters and recommend by calculating their similarities, which is referred from Clustering Based Collaborative Filtering Approach for Recommendation System (ClubRS)   technique. User’s data are always privacy sensitive and can be misused by the service provider while generating recommendations. In this paper we aim to protect the private data from the service provider while preserving the functionality of the system. We propose a cryptographic solution for preserving privacy of customer’s data in recommender system. In short, private information of customer is protected and service provider generates recommendation by processing encrypted data. The proposed method is based on homomorphic encryption schemes. Later in the first stage, the encrypted services are divided into small clusters of similar services using agglomerative clustering algorithm (AHC). In the second stage, collaborative filtering algorithm is applied on one of the clusters and recommendations are made using decryption algorithm. Since the number of services in a cluster is much less than the total number of the services provided, it is expected to also enhance the execution time of collaborative filtering.

Keywords:  big data application, cluster, collaborative filtering, mashup, homomorphic encryption, privacy.

Title: Privacy Enhanced Clustering Based Collaborative Filtering of Services for Big Data Application

Author: Dr. Ramakrishna V Moni, Akshay M Nandagaon

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: Software testing is used to guarantee the quality of software products however, software testing researchers continue to rely mainly on using the documents from the systems specification of the software product. Over the years, software testing researchers attempts to improve the quality of the software products through testing process. Still, modern software testing techniques rely on having a reasonable knowledge of either the codes, systems designs or the modules/units. In this paper, attempt is made to derive Assumed Systems Specifications from the output results generated by the software product being tested. The proposed Assumed System Specifications, will aid the software tester, to capture the functions of software product, using the output results. The merits of the proposed software testing approach are to deriving Assumed System Specifications from output results generated, and this can be used to recommends areas of improvement in the quality of software.

Keywords: Software Testing, Systems Specifications, Software Development Life Cycle, Quality Software.  

Title: Software Testing Based on Assumed Systems Specifications

Author: Osaghae E. O.

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: At present people are using various online platforms for discussion purpose. We can ask questions and get answers to our questions for other people using that platform. Some of these platforms are: Quora, face book pages, Google plus, etc. and these questions are from various fields like politics, science and technology, sports, entertainment, education, etc.. the answers that we get on these platforms are mainly in the form of comments that include text data. To get a complete understanding of what is going on the user have to read all the comments and some of this comments can even have same content. It is wastage of time and energy and still the user doesn't get the answer that he is looking for. A way to provide better solution to these questions is to analyze the text data and generate a report of it. Report will be the summarized version of the entire conversation. We have tried to do this with the help of text mining techniques to extract information from these user responses and generate report having pie charts or graphs. the way asking question is in the form of multiple choice questions. The person posting question will post the question along with some option that he can think of as answers to the his question. The users submitting responses can select from the pre-defined options or can submit a new response. Whether to add that option or ignore it will be up to the owner of the question. Depending on few parameters report will be generated that can be referred by the user at any time. It will save the user from the headache of reading the entire conversation. It will be a free platform developed in android. It will require a working internet connection. At present it will be used as a simple report generation platform but in future depending upon the user base it can extended for survey and research purpose by individuals and many different organization.

Keywords: K-means; NLP; Clustering; Opinion mining.

Title: My Opinion: AnRGE to Enrich Your Knowledge

Author: Shubhankar V. Mujumdar, Manish Kumar, Rohit More, Sudarshan Mane, Mr. Sagar Salunke

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: Nowadays, the banking and financial systems have been totally changed due to the environment and globalization changes and competition of business services. Internet Banking or Web Banking is used to describe banking transactions through internet application. Internet Banking means user can get connected to his bank’s website with personal computer system and web browser. But there are many security problems like fraudulent websites, fake emails from banks, capturing user IDs and passwords, hacking personal bank accounts and steal money etc. To overcome these problems, this gives a solution with a Fingerprint biometric.

Keywords: Secure Internet banking, Finger print recognition, Banking transaction, Minutiae.

Title: Enhancing Security for Online Banking Using Fingerprint Biometrics

Author: Priyanka Mahajan, Supriya Malekar, Anuja More, Amol Wairagade,Prof. B. Mahalakshmi

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: Object detection and tracking are two fundamental tasks in multi - camera surveillance. This paper proposes a framework for achieving these tasks in a nonoverlapping multiple camera network. A new object detection algorithm using mean shift (MS) segmentation is introduced, and occluded objects are further separated with the help of depth information derived from stereo vision. The detected objects are then tracked by a new object tracking algorithm using a novel Bayesian Kalman filter with simplified Gaussian mixture (BKF-SGM). It employs a Gaussian mixture (GM) representation of the state and noise densities and a novel direct density simplifying algorithm for avoiding the exponential complexity growth of conventional Kalman filters (KFs) using GM. When coupled with an improved MS tracker, a new BKF-SGM with improved MS algorithm with more robust tracking performance is obtained. Furthermore, a monitoring-based object recognition algorithm is employed to support object tracking over nonoverlapping network. Experimental results show that: 1) the proposed object detection algorithm yields improved segmentation results over conventional object detection methods and 2) the proposed tracking algorithm can successfully handle complex scenarios with good performance and low arithmetic complexity. Moreover, the performance of both monitoring - and training-based object recognition algorithms can be improved using our detection and tracking results as input.

Keywords: Video analytics, detection, tracking, recognition, Bayesian Kalman filter.

Title: Video Surveillance over Camera Network Using Hadoop

Author: Naveen Kumar, Elliyash Pathan, Lalan Yadav, Viraj Ransube, Prof. Sonia Mehta

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: Vehicle accidents are one of the most leading causes of fatality. The time between an accident occurrence and the emergency medical personnel are dispatched to the accident location is the important factor in the survival rates after an accident. By eliminating that time between an accident occurrence and the first responders are dispatched to the scene decreases mortality rates so that we can save lives. One approach to eliminate that delay between accident occurrence and first responder dispatch is to use An Accident Alert and Vehicle Tracking System, which sense when a traffic accident is likely to occur and immediately notify emergency occurred. In this paper, that system is described the main application of which is early accident detection. It can automatically detect traffic accidents using vibration sensors and immediately notify a central emergency dispatch server after an accident had occurred using GPS coordinates further the central emergency dispatch sever send that location details to the ambulances which are near to that location. This system uses the things i.e. Raspberry Pi, Vibration Sensors, GPS and GSM modules to detect traffic accidents.

Keywords: GPS Module, GSM Module, Raspberry Pi, Accelerometer ADXL 335.4.

Title: Review on Accident Alert and Vehicle Tracking System

Author: Prashant Kokane, Sawant Kiran, Doiphode Piraji, Bhole Imran, Prof. Yogesh Thorat

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: In this paper, we consider the clustering of very large distributed datasets over a network using a decentralized K-means algorithm. Analysis of this data and identifying clusters is challenging due to processing, storage, and transmission costs. Many algorithms has been invented for distributed data clustering, which are applied when datasets cannot be concentrated on a single machine, for instance because of some reasons or due to net-work bandwidth limitations or because of the big amount of distributed data. Low overhead analysis of huge distributed data sets is must for current data centers and for future sensor networks. Our experimental evaluations show that dataset Clustering using K-means can discover the clusters more efficiently with scalable transmission cost, and also expose its supremacy in compare to the popular method LSP2P.

Keywords: Distributed systems, clustering, dynamic system, partition-based clustering, and density-based clustering.

Title: Dataset Clustering Using K-Means Algorithm

Author: Mayur Kulkarni, Karan Jamdade, Krushna Kashid, Akash Dhotre

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: The continuous updating of antivirus database with malware signatures degrades the efficiency of the antivirus system. Existing antivirus researchers are finding ways making the malware signatures database gets currents signatures of merging malware threats. However, the astronomical increase in the number of malware signatures update, thereby reducing the performance of the computer system. Consequently in this paper, an attempt was made to group individual malware signatures into a similar pattern, called the family malware signatures. Each group of family malware signatures, cancels all the individual malware signatures with similar patterns into that group of malware signature. The first advantage of this single grouping of malware signature is, the searching time to attempt matching a pattern of executable file with malware signature will be reduce. Another advantage of this approach is in the reduction of many individual malware signatures into a single group of malware signature, hence, reducing the number of signatures in antivirus database and at the same time, making antivirus database more scalable.

Keywords: Antivirus, signatures database, Malware signatures, malware code.  

Title: Improved Signature-Based Antivirus System

Author: Osaghae E. O.

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: Fingerprint recognition is one of the most adopted techniques for user identification. Many highly accurate systems are commercially available. Poor quality of fingerprint image degrades the performance of existing biometric systems. Recent research has begun to tackle the problems of poor quality of fingerprints. So, we proposed a methodology to improve the matching performance of distorted fingerprints based on a single image. The orientation field of a fingerprint is important for feature extraction and matching. However, the reconstruction of orientation map of distorted fingerprints is a challenging problem. Conventional methods don’t use prior knowledge of fingerprint. Hence, we proposed to use a enhanced feedback based approach using fuzzy logic for the reconstruction of orientation map. Further, the proposed enhanced fuzzy feature matcher (EFFM) takes a restructured ridge orientation map and minutiae map for matching. The proposed methodology enhances the performance and improves the fingerprint matching accuracy.

Keywords: Orientation Field Estimation, Enhanced Feedback, Minutiae, Fuzzy Feature Match (FFM). 

Title: Distorted Fingerprint Matching Performance Improvement

Author: Shubhangi Ingale, Dr. Ayesha Butalia

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: In today’s world, gigantic amount of data is available in science, industry, business and many other areas. This data can provide valuable information which can be used by management for making important decisions. But problem is that how can find valuable information. The answer is data mining. The work focuses on the fundamental concept of the Data mining i.e. Clustering and Classification Techniques. Clustering is done by analyzing k-means algorithm and classification techniques such as J48 and NAÏVE BAYES. Cluster is a group of objects that belongs to the same class. In other words, similar objects are grouped in one cluster and dissimilar objects are grouped in another cluster. The procedure follows a simple and easy way to classify a given data set through a certain number of clusters. Classification is an important data mining technique with broad applications. It classifies data of various kinds. Classification is used to classify each item in a set of data into one of predefined set of classes or groups. This work has been carried out to make a performance evaluation of Naïve Bayes and j48 classification algorithm. Naive Bayes algorithm is based on probability and j48 algorithm is based on decision tree, that to make comparative evaluation of classifiers J48 and NAÏVE BAYES.

Keywords: Classification, Clustering, J48, Naïve Bayes, k-means, unsupervised learning.

Title: Analysis of K-Means Algorithm Using Classification Techniques in Mammographic Dataset

Author: Mrs. K.K. Kavitha, Ms. S. Kiruthiga

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: The E-Voting means the voting process in election by using electronic device. We also described how the android mobile phones are efficient and can be used for voting. The android platform is used to develop an application. Our system support simultaneous voting due to the distributed nature of the database. During election electronic device is used for voting process. A voter may only need to register only once for a particular election and that does all, voter need to cast his /her vote without actually have to present at the voting cell. The registration process must be done at Booth application for once then voter is been given a facility to vote from his/her Android mobile phone irrespective of his/her location. We suppose to propose a new e-voting system, which ensures voter confidentiality and voting accuracy, thus providing an important framework that based on unique identification ADHAAR ID (U-ID) number. An online solution is very useful as the information about the voters and the election committee is also made available to the people in this system. Keywords: Android, E-voting system, Open sources, Web services. Title: Literature Survey on Secure Mobile Based E-Voting System Author: Rajashree Raskar S, Bhagyashree Jaykar V, Akshay Akhare A, Manoj Gadale R, Mrs. D.A. Phalke International Journal of Computer Science and Information Technology Research ISSN 2348-1196 (print), ISSN 2348-120X (online) Research Publish Journals

Abstract: In 2012 Mozilla Corporation demonstrated an open source operating system named Firefox OS through its Boot2Gecko project. A year later, in 2013 the Firefox OS based phones were introduced in market. The widely accepted Android, iOS phones are there since many years, so research has been done on their digital investigation techniques and thus many forensic tools are available for such platforms. But, as the Firefox OS is still in its evolving phase not much research has been done on its forensic procedures, leading to lot of new opportunities and challenges for forensic investigators. Even the existing forensic tools does not fully support the Firefox OS based devices. Many tools detect the device as Android, the reason being the underlying Linux kernel in Firefox OS just like Android. Due to the increasing rate of mobile crimes the forensic analysis of this new, unique OS is urgently needed. The purpose of this paper is to present a method for acquiring forensic data from a mobile device running Firefox operating system. The paper provides a way to start acquisition process by using Android Debug Bridge before proceeding to identify the important files that should be used for analysis. Finally it shows how tools like qtADB, HexEditor and MOBILedit can be used in the acquisition phase of mobile forensics.

Keywords: Firefox OS; B2G; Mobile Forensics; Data Acquisition; Forensic investigation; Gecko.

Title: Data Acquisition Methodology for Forensic Investigation of Firefox OS

Author: Mandar Jadhav, Prof. K.K Joshi

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals

Abstract: An Innovative Product With Societal Acceptance Is The One That Aids The Comfort, Convenience And Efficiency In Everyday Life. Purchasing And Shopping At Big Malls Is Becoming Daily Activity In Metro Cities. We Can See Big Rush At These Malls On Holidays And Weekends. People Purchase Different Items And Put Them In Trolley. After Completion Of Purchases, One Needs To Go To Billing Counter For Payments. At Billing Counter The Cashier Prepare The Bill Using Bar Code Reader Which Is Very Time Consuming Process And Results In Long Queue At Billing Counter. In This Paper, We Discuss A Product “Smart Trolley In Mega Mall” Being Developed To Assist A Person In Everyday Shopping In Terms Of Reduced Time Spent While Purchasing. The Main Objective Of Proposed System Is To Provide A Technology Oriented, Low-Cost, Easily Scalable, And Rugged System For Assisting Shopping In Person.

Keywords: Smart Trolley, Innovative Product, Big Malls, Metro Cities, Big Rush, Billing Counter, Technology Oriented, Low-Cost, Easily Scalable.

Title: Smart Trolley Using QR Code

Author: Arbaaz Khan, Aadil Siddiqui, Zeeshan khan, Jasmine khan, Prof. Amit S Zore

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), SSN 2348-120X (online)

Research Publish Journals

Abstract: Life is the main interesting thing in this existence and we are always searching  to make it better and easy answering our duties as simple as it can , the purpose is easy to understand it cause we are looking for a comfortable life which handles all our routine tasks, for that today we introduce a system which can be plugged on our terminals and which it can make tasks automatically basing on the environment events.

We present a dissertation project details and its application on both sides the web corner and terminal assistant, Multi-Agent System collaborates with PDDL to determinate which functions and procedures the agent has to execute at the time basing on environments events which are playing the rules of parameters in each functions or procedures.

Keywords: Multi-Agent, Agent, PDDL, Engine, Domain, Predicates, Robot.

Title: Multi-Agent System and Events Plan Construction Using PDDL

Author: ABDELLAH REDJAIMIA

International Journal of Computer Science and Information Technology Research

ISSN 2348-1196 (print), ISSN 2348-120X (online)

Research Publish Journals