International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 31 - Number 11 |
Year of Publication: 2011 |
Authors: Dr.S.Siva Sathya, Dr. R.Geetha Ramani, K.Sivaselvi |
10.5120/3938-5527 |
Dr.S.Siva Sathya, Dr. R.Geetha Ramani, K.Sivaselvi . Article:Discriminant Analysis based Feature Selection in KDD Intrusion Dataset. International Journal of Computer Applications. 31, 11 ( October 2011), 1-7. DOI=10.5120/3938-5527
Intrusion detection system (IDS) plays a major role in providing network security by analyzing the network traffic log and classifying the records as attack or normal behavior. Generally, as each log record is characterized by a large set of features, an Intrusion Detection System consumes large computational power and time for the classification process. Hence, feature reduction becomes mandatory before attack classification for any IDS. Discriminant analysis is a technique which can be used for selecting important features in large set of features. In this paper, important features of KDD Cup ‘99 attack dataset are obtained using discriminant analysis method and used for classification of attacks. The results of discriminant analysis show that classification is done with minimum error rate with the reduced feature set.