|
Title:
|
GENETIC ALGORITHM TO DETERMINE RELEVANT FEATURES FOR INTRUSION DETECTION |
|
Author(s):
|
Namita Aggarwal , R K Agrawal , H M Jain |
|
ISBN:
|
978-972-8924-88-1 |
|
Editors:
|
Ajith P. Abraham |
|
Year:
|
2009 |
|
Edition:
|
Single |
|
Keywords:
|
Feature selection, Support Vector Machine, Intrusion Detection, Genetic Algorithm |
|
Type:
|
Full Paper |
|
First Page:
|
75 |
|
Last Page:
|
82 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Real time identification of intrusive behavior based on training analysis remains a major issue due to high dimensionality
of the feature set of intrusion data. The original feature set may contain irrelevant or redundant features. There is need to
identify relevant features for better performance of intrusion detection systems in terms of classification accuracy and
computation time required to detect intrusion. In this paper, we have proposed a wrapper method based on Genetic
Algorithm in conjunction with Support Vector Machine to identify relevant features for better performance of intrusion
detection system. To achieve this, a new fitness function for Genetic Algorithm is defined that focuses on selecting the
smallest set of relevant features which provide maximum classification accuracy. The proposed method provides better
result in comparison to the other commonly used feature selection techniques. |
|
|
|
|
|
|
