Volume : II, Issue : II, February - 2013

Intrusion Detection by Implementing CRF's Using the Technique of Layered Approach

Baji Yadala, Shaik Salma Begum

Abstract :

To function effectively with high amount of network traffic an intrusion system must be have capability of detecting malevolent activities in a network. Intrusion detection system faces a number of challenges in the real world. Using the techniques of both Conditional Random Fields and Layered Approach we have addressed the Accuracy and Efficiency. Considering examples we exhibited that high attack detection accuracy can be achieved by using Conditional Random Fields and high efficiency by implementing the Layered Approach. Our experimental results on benchmark KDD ’99 intrusion data set show that our proposed system based on Layered Conditional Random Fields has better performance as compared to decision trees and the naive Bayes methods. For our methods mathematical testing also provided higher confidence in detection accuracy. Thus we proved that our system is strong and can handle noisy free data providing with high performance.

Keywords :

Article: Download PDF   DOI : 10.36106/ijsr  

Cite This Article:

Baji Yadala,Shaik Salma Begum Intrusion Detection by Implementing CRF's Using the Technique of Layered Approach International Journal of Scientific Research, Vol.II, Issue.II February 2013

Number of Downloads : 274

References :