IJCST Logo

International Journal of Computer Science and Technology
Vol 6.3 ver – 2 (July to September 2015)

S.No. Research Topic Paper ID Download
28

A Comparative Study on the Technologies of Project Management and their Compatibility with Software Development Methodologies (Case Study: Software Companies in Iran)

Ersin Karaman, Hasan Asil

Abstract

In developing countries, one of the problems in the successful implementation of projects is that project management techniques are not applied. This problem results in a huge loss of financial and human resources . Nowadays project management has become a knowledge for which some standards have been defined to guarantee the existence of desired executive methods and identify the relevant elements. In this paper, the standards of project management were explained comparatively first. Then various software development methodologies were studied. Each methodology had a dedicated process and attributes. After that, the consistence of this project with software development standards were investigated .
In this study, a questionnaire was used to investigate the compatibility of project management methods with software development methodologies. The information of statistical
samples were provided by different software roles in various software companies with statistical methods. Then the information was analyzed in SPSS. After that, it was clarified that knowing PMBOK had a positive impact on projects so that it would become more compatible with RUP.

Full Paper

IJCST/63/1/A-0543
29

Detection of Malware in Mobile Networks Using Heterogeneous Devices

Badiga Neeraja

Abstract

Malware attacks are more often made in cellular networks such as viruses, worms and other malicious software. Spyware which disrupts the functionality of the computer to the network, hypersensitive piracy information and access to their private systems. It’s just a program that is specifically designed to be able to corrupt the computer can be either virus or worm can be.
The mobile malware can spread through two different approaches these are usually powerful MMS and Bluetooth. To prevent the spread of malware and help infected nodes to extract better or dispersed optimal method signature. But this kind of ignores the hybrid viruses that spread through both BT and MMS routes.
Therefore, to increase the efficiency of spreading malware restrict your cell phone, we have proposed a new approach called any model hybrid virus detection. The hybrid malware could spread therefore end to end messaging services through personal and social SRDs communication services Wi-Fi. In this method, a new process -based differential equation is proposed to examine the behavior of delocalized mixed infection and the spread of ripple for hybrid malware based on widespread Internet sites including individual sociable and spatial relationships. An experimental result means that the proposed technique is computationallyefficient to differentiate the hybrid malware.

Full Paper

IJCST/63/1/A-0544
30

Impact of Information Technology on Management in Small and Medium Industrie

Abdullah NARALAN, Hasan Asil

Abstract

The impact of information technology on better management in small and medium industries was studied in the present paper. This research was a causal study which aimed to find answer to the question that whether the use of information technology can affect the management of small and medium industries or not or, in other words, whether the use of information technology leads to a better management in small and medium industries or not. Statistical population included small and medium manufacturing industries
and the sample members were selected by random sampling. The required data and information were collected by handing out a questionnaire among the participants and statistical analyses were done in SPSS17. The results showed that information technology affects the accessibility of new tools of marketing for small and medium industries.

Full Paper

IJCST/63/1/A-0545
31

A New Shortest First Path Finding Algorithm for a Maze Solving Robot

Amanpreet Singh

Abstract

The Proposed Algorithm is an improvement to the author’s previous algorithm ‘A New Shortest path Finding Algorithm for A Maze Solving Robot’. It is an addition of two new rules
that increases overall performance of the algorithm. It works on the policy of shortest first path and rejects the paths those will be definitely equal or larger as compared to previously found destination path.

Full Paper

IJCST/63/1/A-0546
32

Software Effort Estimation Use of Bayesian Network for web effort estimation

Ekta Rani, Parbhat Verma

Abstract

Approaches electric power and judgment that looks effort is an pressing set-up in other words software systems event functions possibly will trick. The values cultivate with beautifying the particular issue of one’s a pc software function and application.
Indicating reliability and security through matter towards the beginning of cycles of top envision enormously easy. By all accounts, the fee that is associated finishing may be valid of things exists among resolute reasons good luck of the whole constitution provider. Several conditions that could possibly be proper as employees and experience that is most certainly experience overly helpful within the money for the solar panel systems. The prospective undoubtedly number one on industry needs to be can help turn products that are good could be ages this is certainly superior within tight budget. The problems and dimensions of packages happen to be strengthening rapidly truth it costs often difficult because progress of devices modern technology consuming the exact property and that is exactly quick of industry. But bit of sorts or analysts can estimated an method
which happens to be make it possible for which will be manager that is cost-efficient job that is realistic, network number and renovation carryout.

Full Paper

IJCST/63/1/A-0547
33

Detection and Avoidance of Black hole Attack in MANETs using Diffie-Hellman Algorithm

Jeveen Kumar Yumnam, Maninder Kaur

Abstract

A mobile ad-hoc network (MANET) is a collection of wireless mobile nodes that can dynamically be setup anywhere and anytime forming a temporary network without the assistance
of any stand-alone infrastructure or centralized administration.
Due to openness, dynamic topology, infrastructure-less nature, MANETs are vulnerable to various kinds of attacks. Ad-hoc Ondemand Distance Vector (AODV) is one of the popular protocols used for routing in MANET. But the security of AODV protocol can be wrecked by Black Hole attack. A Black Hole attack can ruin routing in mobile ad-hoc network. In a black hole attack, a malicious node impersonates a destination node by sending a spoofed route reply packet to a source node that initiates a route discovery. By doing this, the malicious node can deprive the traffic from the source node. In this research paper, we have presented a new solution to detect and avoid the Black hole by using Diffie-Hellman Algorithm which does increases the performance metrics like throughput, packetloss by a large extent. Also the false detection ratio is negligible in this approach.

Full Paper

IJCST/63/1/A-0548
34

Research on Traceback with Detecting Distributed DDos Attacks in MANETS

Mitali Sagar, Aman Arora

Abstract

Recommended denial-of-service (DDoS) is truly an uncomfortable crisis might growing rapidly. Content writing provide you with the improvements you like campaigns to see the DDoS issues in final and overview some basic safety which is a lot which could be the typical capable. Your machine for providing a DDoS stop are actually acquirable but there is genuinely too little of adequate issues that prevent this blasts in a amount that happens to be short of. DDoS opponents process by infiltrating variety that is vast off
by making use of deficiencies, for DDoS addresses. Using solar energy focus and wind energy stay possibly which will do a attack that may be complete one customer service which targeted. The DDoS attempts happen to be developed tactics for occupying a residential area construction to turn out to be purposeless to network that is electronic really trustworthy. These events is extremely detrimental to practically any la red. Every traceback get to use is required to be well calculate the types of tipping boutique but
file re-construction formulas in transmission authentically display determining most important medialink handheld n broadband router that comes with the track. Nowadays there are picked out significantly more hold that might be preferred real-time gain access to blend could’ve the uniqueness of early modem with no the data of all the dubious modems within gateway. Because invader could possibly get any market over the web process tackle whack, cheney cannot obtain.

Full Paper

IJCST/63/1/A-0548
35

Simulation & Analysis of Energy Detection of Spectrum in Cognitive Radio System

Swagata B Sarkar

Abstract

Cognitive Radio is an emerging and demanding technology of wireless technology of wireless communication system. With the advancement of wireless communication, a lot of constraints on the usage of available radio spectrum take place. Manny survey of spectrum utilization shows that entire spectrum is not used at all the times. Some of the frequency bands are occupied, some of the bands are less occupied and some frequency bands are not at all occupied. Cognitive radio system is the technique to overcome the problem of under utilization of bandwidth. A function of Cognitive Radio is to identify the spectrum hole and allot secondary users to utilize the band with high spectral resolution capability. In this paper, effective utilization of spectrum is shown with single primary user multichannel scenario. Four users and four channels are used with different threshold to utilize
the channels effectively.

Full Paper

IJCST/63/1/A-0548
36

An Effective Authentication Technique to Prevent Keylogging

Badiga Neeraja

Abstract

Keylogging is an activity to capture user keystrokes and records the activity of a computer user in a covert manner using keyboard logger hardware and software. Secretly monitors and keyloggers record all keystrokes. Unlike other malware, keyloggers cause any threat to the system. But it can be used to intercept passwords and other confidential information entered via the keyboard, residents are considering various rootkits on computers (PCs) that violates safety. Cybercriminals can get user names, email passwords, PIN codes, account numbers, email addresses, account passwords, online games, electronic payment systems, etc. As a result, it is seen as a user authentication for financial transactions. To prevent
keylogging, strict authentication is required. The QR code can be used to design the visual authentication protocols to achieve high ease of use and safety. The two authentication protocols are One-Time-Password Authentication Protocol and time-based protocol based passwords. Through accurate analysis, protocols have proven robust authentication several attacks. And also for the deployment of these two protocols in real-world applications, especially in online transactions, the strict safety requirements can be met.

Full Paper

IJCST/63/1/A-0549
37

Quality Service Activity Applications Combination in Cloud Computing Model

Chaitanya.P, Dr. Guru Kesava Das.G

Abstract

The cloud quality service model is taken to create multiple teams most carefully not only in public clouds also in private clouds and larger organizations. Present cloud service users isolate different user models that are proposed and data within a single tenant limits with max or minimum cross tenant interaction. Latest generations have seen the negative migration of model applications to the cloud .One of the methods based on cloud applications is Quality-of-Service (QoS) system. This paper take EXACT and number of Polynomial Time Approximation Scheme (FAPTS) algorithms for QoS method service comparisons to change user experiences for Cloud service access methods. It introduces the concept of cloud computing and explains the QoS Aware Services Mash up (QASM). Many number of resource proved models are used and must take Quality of Service (QoS) functions like availability and
security response time, security reliability and thereby avoiding Service Level Agreement (SLA) observations. Static and Dynamic models location provisioning becomes insufficient to allocate resources number of times to the user demands in order to satisfy their requests and take care of the Service Level Agreements (SLA) provided by the service providers. This paper discusses various Resource Allocation models that are used to allocate resources efficiently

Full Paper

IJCST/63/1/A-0550
38

Support and Confidence based Methods for Mining and Hiding Sensitive Rules

Surbhi Bhasin, Gagan Kumar

Abstract

Traditional pattern mining algorithms may not discover a number of most beneficial, high priced patterns, due to their lower support. These algorithms reflect only statistical correlation, but it does not reflect semantic significance of the pattern. Although a number of relevant algorithms have been proposed in recent years, they incur the problem of producing a large number of candidate item-sets for High Utility item-sets. Such a large number of candidate item-sets degrade the mining performance in terms of execution time and
space requirement. Also the previous algorithms do not consider the impact of item sets with high tolls. The Proposed strategies in this work can not only decrease the overestimated utilities of potential high utility item sets but greatly reduce the number of candidates. Different types of both real and synthetic data sets are used in a series of experiments to the performance of the proposed algorithm with state-of-the-art utility mining algorithms.
Experimental results show that these algorithms outperform other algorithms substantially in term of execution time, especially when databases contain lots of long transactions or low minimum utility thresholds are set

Full Paper

IJCST/63/1/A-0551
39

Cocktail Aspects Proceed For Travel Package Recommendation Using Clustering

Sruthi.M, Dr. Balakrishna Prasad.P, Dr.Guru Kesava Das.G

Abstract

The present generations in research show that some package recommendation models are easier to use instead of single data. The problem is challenging due to different ways of object direction models. Here we first analyze the characteristics of the prevailing travel packages and develop a tourist-area-season topic (TAST) model .Real time results show the TAST model effectively captures the unique characteristics of the travel data and the cocktail model is much more effective than traditional techniques for travel
packages .We provide the new models and we update the current news status regarding that place. With the help of this information the tourist find the shortest path for their destination region. We also extend the TAST model to the tourist-relation-area-season topic (TRAST) model for capturing the latent relationships among the tourists in each travel group. A new approach called Clustering and typicality based Collaborative Filtering has been detailed out herewith this paper, which includes preprocessing methods, clustering of items and measuring user typicality degree in user groups. After pre-processing the remaining recommendation process is done based on user typicality degree instead of coated
items as in present Collaborative Filtering.The developed model is test on aTrip Advisor information and compared with a proposed method for package recommendation. Finally we evaluate TAST model, TRAST model, and Cocktail Approach and Clustering techniques on real world travel package data

Full Paper

IJCST/63/1/A-0552
40

Methods for Estimating Software Reuse in Component based Software Systems

Yudhveer Singh, Gagan Kumar

Abstract

A outstanding deal of research above the past countless years has been devoted to the progress of methodologies to craft reusable multimedia constituents and constituent libraries, whereas there is an supplementary price encompassed to craft a reusable constituent from scratch. That supplementary price might be evaded by recognizing and removing reusable constituents from the by now industrialized colossal catalog of continuing systems.
But the subject of how to recognize good reusable constituents from continuing arrangements has stayed moderately unexplored. Our way, for identification and evaluation of reusable multimedia, is established on multimedia models and metrics. As the precise connection amid the qualities of the reusability is tough to institute so a Neural Web way might assist as an frugal, automatic instrument to produce reusability ranking of multimedia by devising the connection established on its training. With the goal of seizing supremacy of the features of the both, in the research work Neuro-Fuzzy way is utilized to frugally ascertaining reusability of multimedia constituents in continuing arrangements as well as
the reusable constituents that are in the design period

Full Paper

IJCST/63/1/A-0553
41

Topical Context Aware Community Detection in Social Media Discussion

Amit Dhumal, Pravin Kamde

Abstract

Community detection algorithms are used to uncover hidden properties of network. Existing methods of discovering community in online social network largely relies upon existing network topology for identifying community of users. With the immense of large number of users and multi type relationship between them, networks become larger and complex. Identifying user interest and based upon that identifying community of like minded users from such online social networks (e.g. Twitter, Facebook) becomes important for targeted advertising and friend recommendation. We propose a context aware community detection approach to identify users with similar topical interest. Initially topic modeling method
is used to extract topic from textual content (e.g. messages) shared by user. Further users interested in similar topic are identified, relationship links between them is used to model network and finally community detection method is used to identify cluster of strongly connected users from the modeled network.

Full Paper

IJCST/63/1/A-0554
42

Classification Criteria for Pedagogical Agents

Dr. Mohamedade Farouk NANNE

Abstract

We studied a set of pedagogical agents referenced in the scientific literature. From this study we classified them according to several criteria. The categories obtained are not exclusive, rather they are most often complementary, that is to say that an agent can belong to multiple categories. We present in this paper a summary of the main criteria for categorization of pedagogical agents that we extracted from our study. These criteria are: the character of the agent, the type of environment in which the agent is integrated, the educational role the agent plays in this environment, nonverbal components whose agent has (look, gesture, facial expressions, emotions), the multiplicity of agents in an application, the management of collaborative learning, application areas and conditions in the entry that takes into account pedagogical agent.

Full Paper

IJCST/63/1/A-0555
43

A Framework to Deduce Distinct User Search Objectives for an Enquiry by Using Feedback Sessions

K. Simha Madhuri, S. Rama Sree

Abstract

A search engine is a program that is used to search information on the internet. When an enquiry is issued in search interface, distinct users may have distinct search objectives in mind for the same enquiry. The study of user search objectives is helpful in refining search engine significance and user practice. In the first phase, a model is suggested to find out distinct user search objectives for an enquiry by grouping Feedback sessions. Feedback sessions are shaped from user click logs which can provide information to
the user. In the second phase, a novel approach is recommended to manufacture Pseudo-documents to better denote the Feedback sessions for grouping. In the third phase, a new principle “Classified Average Precision” is proposed to estimate the deduction of user search objectives. By using these user search objectives, results are reorganized. In the last phase, PageRank is applied to obtain ranked results in which most visited URLs are displayed on the top. Results from search engine show the efficiency of
the proposed method. By using above methodology, users can retrieve their information more efficiently. To improve grouping performance and user search relevance, Fuzzy and bisecting clustering algorithms can be enhanced in future.

Full Paper

IJCST/63/1/A-0556
44

Dimensionality Reduction Using CLIQUE and Genetic Algorithm

Behera Gayathri, A. Mary Sowjanya

Abstract

Clustering high dimensional data is an emerging research field as; it is becoming a major challenge to cluster high dimensional data due to the high scattering of data points in the full dimensional space. Most of the conventional clustering algorithms consider all the dimensions of the dataset for discovering clusters whereas only some of the dimensions are relevant. In this paper, we propose a “Dimensionality Reduction using Clique and Genetic Algorithm” (DRCGA) for the problem of high-dimensional clustering.
The DRCGA algorithm consists of two phases. The first phase is the preprocessing phase where CLIQUE is used for subspace relevance analysis to find the dense subspaces. In the second phase a genetic algorithm is used for clustering the subspaces detected in the first phase. This algorithm is advantageous when high-dimensional data is considered. The problem of losing some of the regions that are actually densely populated due to the high scattering of data points in high-dimensional space can be overcome. The experiments performed on large and highdimensional synthetic and real world data sets show that DRCGA performs with a higher efficiency and better resulting cluster accuracy. Moreover, the algorithm not only yields accurate results when the number of dimensions increases but also outperforms the individual algorithms when the size of the dataset increases.

Full Paper

IJCST/63/1/A-0557
45

An Enhancement in Energy Optimization of Nodes Using Bacteria Foraging Optimization

Shweta Nikhanj, Dinesh Kumar

Abstract

Energy consumption is one of the constraints in Wireless Sensor Networks. The key issue in WSN is that these networks suffer from packet overhead which is the root cause of more
energy consumption in sensor networks. As a result of which the transmission of the packets from source to destination is not appropriate. Researchers in the area have proposed several different approaches to optimizations of a wireless sensor network design. However, most of the optimization procedures do not take into account the principles, characteristics and requirements of an application-specific WSN at the system level. So in proposed work optimization of nodes will be done so that various computational
parameters can be optimized. In this work, we have used Bacterial foraging optimization algorithm i.e. also known as BFO algorithm for minimization of total power consumption and increasing network lifetime which compares of others in wireless sensor networks. And our results are evaluated on parameters like throughput, routing overhead, error rate, energy, packet delivery ratio, packet loss ratio, & end delay.

Full Paper

IJCST/63/1/A-0558
46

A Dynamic Variant of AP Clustering to Achieve Efficient Clustering Performance

R.Veera Meenakshi, N.Naga Subrahmanyeswari

Abstract

Affinity propagation clustering be an exemplar-based scheme that comprehends via the handover of each data point to its nearest exemplar, where exemplars be acknowledged via
passing messages on bipartite graph. There are two kinds of messages passing on bipartite graph. They are responsibility and availability, collectively called ’affinity’. The goal of this paper is to propose a dynamic variant of AP clustering, which can accomplish equivalent clustering performance with traditional AP clustering by just middle with the current clustering results according to new arriving objects, slightly than re-implemented AP clustering on the whole dataset. Therefore, a great deal of time can be saved, which makes AP clustering well-organized, sufficient to be used in dynamic environment.

Full Paper

IJCST/63/1/A-0559
47

Performance Evaluation of Scheduling Algorithms with Different MIMO Techniques in LTE Systems

Krishna Teja Yadav CH. T, C.Y. Gopinath, Mohankumar N. M., Devaraju J.T

Abstract

MIMO techniques are used in Wireless Broadband Access (BWA) networks to maximize spectrum efficiency and minimize the bit error rate. LTE is one such BWA network which has adopted MIMO techniques in both the uplink and downlink along with Radio Resource Management (RRM) aspects like scheduling to improve the data rate. Scheduling is mainly concerned with allocating the available radio resources among the users depending upon the metrics such as Quality of Service (QoS) requirements of users, channel conditions etc. Hence in this paper, an attempt is made to study and compare the performance of scheduling algorithms (RR, PF, MT and BET) with MIMO techniques such as SISO, SIMO, SFBC and OLSM for Constant Bit Rate (CBR) traffic scenario. The performance metrics used are average throughput and average delay.

Full Paper

IJCST/63/1/A-0560
48

An Implementation for Combining Neural Networks and Genetic Algorithms

Md. Mijanur Rahman, Tania Akter Setu

Abstract

Neural Network (NN) and Genetic Algorithm (GA) are two very known methodology for optimizing and learning. Each having its own strengths and weakness. These two have generally evolved along separate paths. Recently there have been attempts to combine the two technologies. This research is devoted to implement a method for combining genetic algorithm with neural network (GANN). A collaborative approach has been used in this research. To integrated GA and NN into a single system, a population of neural networks is evolved, i.e., the goal of the proposed system
is to find the optimal neural network solution. In collaborative system GA and NN work parallel; for optimization it is very necessary to work with parallel.Several MATLAB functions and tools have been used to implement the proposed GANN method. The development and experiments demonstration of GANN is done on MATLAB 7.0.12. The proposed method consists of neural learning by the backpropagation algorithm and applies evolutionary, genetic operators. The learning behavior of the algorithm was tested on a simple pattern recognition example and it was able to prove its performance. The new ideas, concepts and processes of GANN bring new life in the field of Artificial Intelligence research.

Full Paper

IJCST/63/1/A-0561
49

Elliptic Curve Cryptography based Security Framework for Internet of Things and Cloud Computing

T Daisy Premila Bai, S Albert Rabara, A Vimal Jerald

Abstract

Internet of Things (IoT) and Cloud Computing paradigm is a next wave in the era of computing and it has been identified as one of the emerging technologies in the field of Computer Science and Information Technology. It has been understood from the review reports that integration of IoT and Cloud Computing is in its infantile phase and it has not been extended to all application domains due to its inadequate security architecture. Hence, in this paper a novel Elliptic Curve Cryptography based security framework for Internet of Things and Cloud Computing is
proposed. This is a secured and adoptable one for the public to access diversified smart applications and services distributed in the cloud, anywhere, anytime, any device and any network irrespective of the underlying technologies in a smart environment. The cloud services are integrated and connected through a novel IP/MPLS (Internet Protocol/ Multiprotocol Label Switching) core. Elliptic Curve Cryptography (ECC) is used to ensure complete protection
against the security risks such as confidentiality, integrity, privacy and authentication. The security strengths of various public key cryptosystems are analysed and ensured ECC is one of the best cryptosystems for internet of things and cloud computing. Full Paper

IJCST/63/1/A-0562
50

Channel Estimation in Long Term Evolution

Nidhi Nayal, Kanchan Sharma

Abstract

Long Term Evolution is based on two novel technologies i.e. MIMO (Multiple input and Multiple Output) and OFDM (Orthogonal Frequency Division Multiplexing) and is evolved due to increasing demand of current wireless communications. Channel Estimation plays crucial role in determining Channel State information at receiver side. In this paper pilot based channel estimation has been studied. This paper also proposes Channel Estimation Using ANFIS (Adaptive Neuro Fuzzy Inference System) which utilizes best features of both Artificial Neural Network and Fuzzy Logic. The results are compared with MATLAB simulation. Simulation Results verify that ANFIS with less computational complexity performs better in wireless environment.Full Paper

IJCST/63/1/A-0563