HYBRID FEATRUE SELECTION FOR NETWORK INTRUSION

Journal Title: International Journal on Computer Science and Engineering - Year 2011, Vol 3, Issue 5

Abstract

In Computer Communications, collecting and storing characteristics about connections into a data set is needed to analyze its behaviour. Generally this data set is multidimensional and larger in size. When this data set is used for classification it may end with wrong results and it may also occupy more resources especially in terms of time. Most of the features present are redundant and inconsistent and affect the classification. In order to improve the efficiency of classification these redundancy and inconsistency features must be eliminated. In this paper, we have proposed a new algorithm based on hybrid method to identify the significance of features. The Proposed hybrid method combines Information Gain and Genetic Algorithm to select features. Clustering is carried out on selected features for classification. The experiment is conducted with NLS-KDD network intrusion data set. It classifies the data set with good accuracy.

Authors and Affiliations

S. Sethuramalingam , Dr. E. R. Naganathan

Keywords

Related Articles

Design and Implementation of Neural Processor for Parsing Manufacturing Query Language 

Practically, all the approaches employed for parsing with natural languages use some or other type of neural network architecture and some typical statistical function for obtaining a parsing decision. In parsing with ne...

Inspiring Particle Swarm Optimization on Multi-Robot Search System

Multi-robot Search system is one area that attracts many researchers. In the field of multi-robot system one of the problem is to design a system that allow the robot to work within a team to find a target. There are man...

Throughput Analysis of Many to One Multihop Wireless Mesh Ad hoc Network

Mobile ad hoc networks are formed by co operative association of wireless nodes communicating with each other without the use of infrastructure. Every node acts as a router in the network and enables the communication be...

SArEM: A SPEM extension for software architecture extraction process

In order to maintain a system, it’s critical to understand its architecture. However even though every system has an architecture, not every system has a reliable representation of its architecture. To deal with this pro...

Download PDF file
  • EP ID EP119097
  • DOI -
  • Views 115
  • Downloads 0

How To Cite

S. Sethuramalingam, Dr. E. R. Naganathan (2011). HYBRID FEATRUE SELECTION FOR NETWORK INTRUSION. International Journal on Computer Science and Engineering, 3(5), 1773-1780. https://europub.co.uk/articles/-A-119097