Volume 3, Issue 11, November 2012

QoS of High Speed Congestion Control Protocols

M. A. Mani, J. B. Abed, S. Laârif and R. Mbarek

Department of Computer Science, ISITCOM, Hammam Sousse, Sousse University, Tunisia
Polytechnique Sousse University, Tunisia
Department of Telecommunication, ISITCOM Sousse University, Tunisia

Abstract - Since many years, computer networks have became more and more extensive; however, this evolution brings with her many problems that we can find among them the congestion phenomenon, which actuates many researchers to propose several solutions for these difficulties, aiming to find some TCP alternative congestion control protocols in order to enhance the performance as well as the link utilization. In this paper, we will be studying the Quality of Service of some congestion control protocols. We will evaluate and compare some of high speed congestion control protocols using a dynamic tool of choice of congestion control protocols.

Index Terms - High-Speed, QoS, TCP Protocols, Congestion Control, Performance and Multiple Flows

A Multifaceted Approach Towards Friend Recommendation in Social Network

Pages: 7-11 (5) | [Full Text] PDF (286K)

S. Simranjit, M. Nikunj and M. Nishant

Sardar Patel Institute of Technology, Mumbai, India
Nirma Institute of Technology, Ahmedabad, India
Sardar Patel Institute of Technology, Mumbai, India

Abstract - Social networking has become an intertwined part of our daily lives, with these websites having a user base of several hundred million. Friend recommendation system is a crucial aspect of these social networking platforms, but it hasn’t received the importance it deserves. A good recommendation system would not only give the platform a more intuitive look but it will also improve performance of entire architecture. We have proposed a system using neural network and diversified weights based on multilayer text extraction, frequency of communication for friend recommendation from friends of friends. We have taken into account various factors which will assign score to his friends of friends and will recommend them more efficiently.

Index Terms - Data Mining, Graph, Neural Network, Recommendation and Social Network

A Method for Linguistic Reasoning Based on Linguistic Lukaseiwicz Algebra

Pages: 12-17 (6) | [Full Text] PDF (920K)

Le Anh Phuong and Tran Dinh Khang

Department of Computer Science, Hue University of Education, Vietnam
School of Information & Communication Technology, Hanoi University of Science and Technology, Vietnam

Abstract - This paper studies the linguistic truth value domain (AX) based on finite monotonous hedge algebra and then we extend lukaseiwicz algebra on [0,1] to linguistic lukaseiwicz algebra on linguistic truth value domain (AX), in an attempt to propose a derivatives system based on hedge moving rules and linguistic lukaseiwicz algebra for linguistic reasoning.

Index Terms - Hedge Algebra, Linguistic Truth Value Domain, Linguistic Lukaseiwicz Algebra and Derivatives

Fuzzy-Genetic Algorithm Based Association Rules for Wireless Sensor Data

Pages: 18-22 (5) | [Full Text] PDF (300K)

T. Abirami and Dr. P. Thangaraj

Department of Computer Technology, Kongu Engineering College Perundurai Erode Tamilnadu, India
Bannari Amman Institute of Technology, Sathyamangalam Erode Tamilnadu, India

Abstract - Wireless Sensor Networks (WSN) generates a huge amount of data for efficient application of discovering essential knowledge from it is important. Usually, WSN data is generated in streams and forwarded to a sink. Raw data leads to higher communication overhead adversely affecting WSN performance. Association mining processes data to locate frequent patterns. Thus, when association mining is applied to WSN network data only frequent raw data patterns are sent to the sink, thereby reducing communication overhead. This paper proposes mining of WSN data through an association rule to extricate patterns. A Fuzzy based genetic algorithm is used with the rule for efficient extraction of rules from data.

Index Terms - Wireless Sensor Networks (WSN), Association Rules, Genetic Algorithm and Fuzzy Logic

Towards a Perspective of Hybrid Approaches and Methodologies in Recommender Systems

Pages: 23-32 (10) | [Full Text] PDF (261K)

Nana Yaw Asabere

Mobile Computing Laboratory, School of Software, Dalian University of Technology, Dalian, P.R. China

Abstract - Recommender Systems apply machine learning and data mining techniques to filter undetected information and can predict whether a user of a system would like a given resource based on his/her interests and preferences. To date a number of recommendation algorithms have been proposed, where Collaborative Filtering (CF) and Content-Based Filtering (CBF) are the two most famous and adopted recommendation techniques. CF Recommender Systems recommend items by identifying other users with similar taste and use their opinions for recommendation. CF Recommender Systems suffer from problems and challenges such as scalability, first rater (new item), data sparsity and cold-start problems. On the other hand, CBF Recommender Systems recommend items based on the content information of the items and match these items with interest and preferences of a user and therefore suffer from an overspecialization problem. In order to generate accurate and good recommendations, Hybrid Recommender Systems combine CF and CBF Recommender Systems to avoid the above aforementioned problems and challenges. This paper initially discusses Recommender Systems in general, then presents an overview of the state-of-the-art research in the area of Hybrid Recommender Systems, specifically from the perspective of types, applications, architectures and algorithms and finally discusses relevant open issues of Hybrid Recommender Systems.

Index Terms - Collaborative Filtering (CF), Content Based Filtering (CBF), Hybrid Recommender Systems and Hybridization Methods/Techniques

An Overview of DNS Poisoning and a Possible Solution

Pages: 33-34 (2) | [Full Text] PDF (142K)

Manan Lalaji, Mangesh Bhangare and Jaison Manitharavely

St.Francis Intitute of Technology, Mumbai University, India

Abstract - It is the DNS which makes away remembering strings of illogical numbers to use the internet. Converting readable names into their allotted IP is the basic function of a DNS server. Hence, it provides a very useful and a critical function to the users all over the world. Therefore, any unethical changes in a DNS server can make users not able to reach their desired sites. Hackers can then redirect the users to their malicious sites. This is an example of DNS poisoning. In this paper, we propose a possible solution to DNS poisoning by using multiple DNS servers.

Index Terms - DNS, DNS Poisoning, Internet and IP Address

Chaotic Based Key Management and Public-key Cryptosystem

Pages: 35-42 (8) | [Full Text] PDF (523K)

Mazen Tawfik Mohammed, Alaa Eldin Rohiem, Ali El-moghazy and A. Z. Ghalwash

Military technical College, Cairo, Egypt
Helwan University, Egypt

Abstract - Public key cryptography is an emerging field whose cryptosystem is based on number theory. However, these cryptosystems suffer from problems such as dealing with large numbers and large prime numbers as well. A recent trend for public key cryptosystems is based on chaotic systems. In this paper a new system for public key cryptosystem and chaotic key management system is introduced. The cryptosystem has been used to provide public-key cryptosystem features such as key-exchange, chaotic key management system and encryption/decryption of intended text. In addition, the proposed cryptosystem protocol solves the man in the middle attack problem since it is based on chaotic management systems. The provided chaotic key management system is based on beta-transformation mapping. The new chaotic key exchange protocol is evaluated against the Diffie-Hellman elliptic curve cryptosystem (DHECC). The proposed system is developed using open source bccrypto-net-1.7 project and C#.net programming language. The test results show that the proposed cryptosystem fills the lack of security gap found in the traditional public-key cryptosystems related to proper key size generation.

Index Terms - Public-Key, Diffie-Hellman Elliptic Curve Cryptosystem (DHECC), Chaotic System and Key-Exchange

Recurrent Neural Network Method in Arabic Words Recognition System

Pages: 43-48 (6) | [Full Text] PDF (327K)

Yusuf Perwej

Department of Computer Science & Engineering, CMJ University, Shilong, Meghalaya , India

Abstract - The recognition of unconstrained handwriting continues to be a difficult task for computers despite active research for several decades. This is because handwritten text offers great challenges such as character and word segmentation, character recognition, variation between handwriting styles, different character size and no font constraints as well as the background clarity. In this paper primarily discussed Online Handwriting Recognition methods for Arabic words which being often used among then across the Middle East and North Africa people. Because of the characteristic of the whole body of the Arabic words, namely connectivity between the characters, thereby the segmentation of An Arabic word is very difficult. We introduced a recurrent neural network to online handwriting Arabic word recognition. The key innovation is a recently produce recurrent neural networks objective function known as connectionist temporal classification. The system consists of an advanced recurrent neural network with an output layer designed for sequence labeling, partially combined with a probabilistic language model. Experimental results show that unconstrained Arabic words achieve recognition rates about 79%, which is significantly higher than the about 70% using a previously developed hidden markov model based recognition system.

Index Terms - Recurrent Neural Networks (RNN), Arabic Word, Recognition, Feature Extraction and Language Model

Design and Implementation of Research Module in Educational KM System

Pages: 49-55 (7) | [Full Text] PDF (381K)

Anubhav Kumar and P. C. Gupta

JNU Jaipur, India
Kota University, Kota, India

Abstract - Enterprise Resource Planning (ERP) technology has great impact on our business world. Many multinational companies are using ERP technology for the improvements in their productivity. Currently, this technology is being used in Higher Educational Institutes (HEI) as a replacement of their legacy system to achieve their mission and vision. In this paper a research module is designed and implemented as a knowledge management tool which can be helpful to the faculties, management and researchers of the higher educational institutions.

Index Terms - ERP, KM, EKMS, Tacit Knowledge and Explicit Knowledge

Modeling the Human Face and its Application for Detection of Driver Drowsiness

Pages: 56-59 (4) | [Full Text] PDF (565K)

Lam Thanh Hien, Tran Van Lang, Ha Manh Toan and Do Nang Toan

Faculty of Information Technology, Lac Hong University, Vietnam
Institute of Applied Mechanics and Informatics, Vietnam
Institute of Information Technology, Vietnam

Abstract - The detection problem of human face in the image has been studied since the 70s. However, this is a complex problem so there are many research groups on the world interested. In the detection of human face, the pattern of the human in the image, which is a concept showing the relationship between the image information with a human face that definition. The one of paper results is to give a human face model in the space approach, which can detect some states face. From that the paper proposed an algorithm to detect driver drowsiness. The algorithm has been implemented and tested, and it proved effective in cases where the image was taken a direct and bright enough.

Index Terms - Image Processing, Human Face Recognition and Driver Drowsiness

Volume 3, Issue 11, November 2012

IJCST Managemet

Current Issue

Recent Issues

Author Guidelines

IJCST RSS Feeds

News & Updates