International Journal, Scientific Papers, Global Journal, Indexed, Engineering and Technology, Sciences, Applied Research, High Quality, World Class, ISGJESAR. ISGJESAR is an International world calss Scientific global Journal that accets research articles in the field of Engineering, Sciences and applied research Current_issue
Copyright @ Mercury Publications Consortium 2016. All rights reserved.

Current Issue: Volume 3, No. 4 , July - Aug - Sep 2018

Copy Move Forgery Detection on Digital Images – Primer Research
1 Arun Anoop M, 2 S.Poonkuntran
In today’s advanced age the reliable towards picture is twisting a direct result of malicious forgery images. The issues identified with the security have prompted the examination center towards tampering detection. As the source image also, the objective locales are from a similar picture so that copy move forgery is very effective in image manipulation due to its same properties such as temperature, color, noise and illumination conditions. In this article, we added preliminary research methods (existing algorithms). We used hybrid approaches of existing algorithms. The performance is evaluated in terms of normally used parameters precision and recall with improved results. Thus the proposed CMFD can manage all the image processing operations. We added single image forensics analysis method also. Finally made a comparative analysis based on some clustering algorithms and its applications, drawbacks, digital forensics tool analysis. Full Paper Download Please Click Here

Design of Interworking architecture to Hybrid access System for Reducing Handovers

1 S.M.Rawoof, 2 P.Naga Prasad

There are lots of efforts to progress interworking architectures for the purpose of mobile technologies. Current computing and communication devices are omnipresent and operate in a heterogeneous environment. The users have the privilege to stay connected to the Internet by using mobile terminals equipped with multiple networking interfaces. Thus, the users have the ability to use the services of their choice at anytime and anywhere. These processes of switching between different wireless technologies (Wireless LAN, WiMax, Cellular, UMTS, and LTE etc.) are referred to as Vertical handovers /Handoffs. This paper proposes a totally   integrated, hybrid and worth of service attentive interworking architecture that fulfills the potential of users. The Hybrid access system is one such approach that support operators to distribute higher bandwidth and more consistent services to their subscribers. Full Paper Download Please Click Here

1 Shama M, 2 Kungumaswetha
Nowadays data security of each and every technology is a tedious task. We have crossed all the limits of what a classical computer can do. We need an advanced computing like quantum computing to achieve this security. A technology like big data has a big challenge at hand that is to provide security to it. In this paper, we have discussed how creating a big data using quantum computer is beneficial to both its working and security. As a quantum computer allow its qubit to have not only 2 but 3 states, that is 0,1 and both 0&1. The third state is a huge plus point for efficient working of a big data. Using superposition and quantum parallelism concept processing of a big data is much efficient in both time and work done. At present, we provide security to a technology by securing the transmission channels but now using the concept of quantum key distribution (QKD). So QKD encodes the data in forms of keys and if anyone attempts to crack it will be immediately noticed and keys will be changed. The challenges we faced and various techniques we tried are also highlighted. Full Paper Download Please Click Here


1Ananthi.A 2Akila.T  3Geeitha.S

ABSTRACT Big data is one of the most discussed, and possibly least understood, terms in use in business today. Big data is said to offer not only unprecedented levels of business intelligence concerning the habits of consumers and rivals, but also to herald a revolution in the way in which business are organized and run. However, big data is not as straightforward as it might seem, particularly when it comes to the so-called dark data from social media. It is not simply the quantity of data that has changed; it is also the speed and the variety of formats with which it is delivered. This article sets out to look at big data and debunk some of the myths that surround it. It focuses on the role of data from social media in particular and highlights two common myths about big data. The first is that because a data set contains billions of items, traditional methodological issues no longer matter. The second is the belief that big data is both a complete and unbiased source of data upon which to base decisions. Full Paper Download Please Click Here


R.P.Abinaya1   B.Kalaiselvi2    S.V.Rajeshwari 3
The rapid increase in social networking  results in problematic usage of it. Due to that there is an increasing number of social network mental disorders (SNMDs), such as Net Compulsion, Cyber-Relationship Addiction and Information Overload have been identified. Mining online social behavior provides an opportunity to actively identify SNMDs at an early stage for clinical intervention. A machine learning framework have been proposed namely Social Network Mental Disorder Detection (SNMDD), that exploits features extracted from social network data to accurately identify potential cases of SNMDs. In this proposal a new SNMD-based Tensor Model (STM)  is used to improve the accuracy. To increase the scalability of STM the efficiency with performance assurance has to be improved. Full Paper Download Please Click Here


V.kowsalya1  M.Lakshmi2  P.Premalatha3


Deep learning (DL) plays a vital role in atmost all applications. Several companies like Microsoft,     google, sales force and facebook have embedded DL into the products. But, cyber security is challenging task. Malware detection and network intrusion detection are the two areas where DL shows significant improvement over rule based and classic machine learning (ML) based solutions. DL based neural nets are used in user and entity behaviour analytics (UEBA). UEBA enhanced the detection of insider threats. DL solves problem of traffic detection and also detect many types of anamolies. TOR (The onion Router) is free software that enable anonymous communication over the internet through onion route protocol.TOR safeguard privacy of users.keras with tensorflow is used in backend to train DL module.DL based system helps to detect TOR traffic with high precision and recall. DL based classifier outperforms all other classifiers. Full Paper Download Please Click Here









International Scientific Global Journal in Engineering Science and Applied Research (ISGJESAR)
ISSN : 2456-1894