Within the last years, IT security has become a major economic issue, Intrusion detection systems are an effective way to detect malicious connections in a computer Network. The network traffic data employed to build an intrusion detection system (IDS) is always large with ineffective information, that what decrease it efficiency.to address this problem, we must to keep just the worthless information from the original high dimensional data by using a feature extraction technique. One of the most used in this field is Linear Discriminant Analysis unfortunately this method uses between-class scatter against within-class scatter which are almost always singular. In this paper, we suggest to use the optimal mean PCA as prestep before LDA. Many experiments on KDDcup99 and NSL-KDD indicate the superiority of the proposed technique.
[1]
Leonid Portnoy,et al.
Intrusion detection with unlabeled data using clustering
,
2000
.
[2]
Ali A. Ghorbani,et al.
A detailed analysis of the KDD CUP 99 data set
,
2009,
2009 IEEE Symposium on Computational Intelligence for Security and Defense Applications.
[3]
Rupali Datti,et al.
Feature Reduction for Intrusion Detection Using Linear Discriminant Analysis
,
2010
.
[4]
Feiping Nie,et al.
Optimal Mean Robust Principal Component Analysis
,
2014,
ICML.
[5]
Siguang Chen,et al.
LKM: A LDA-Based K-Means Clustering Algorithm for Data Analysis of Intrusion Detection in Mobile Sensor Networks
,
2015,
Int. J. Distributed Sens. Networks.