IJSR - International Journal of Scientific Research

Brajesh Kumar; Prof. Aishwarya Mishra

Volume : VI, Issue : VII, July - 2017

IMPROVED THE INTRUSION DETECTION CLASSIFICATION RATE USING FEATURE REDUCTION TECHNIQUE BASED ON DCA AND PCNN NETWORK

Brajesh Kumar, Prof. Aishwarya Mishra

Abstract :

Reduction and selection of intruder attribute in intrusion detection system play an important role in process of detection. The huge number of attribute in intruder induces a problem in detection process and increase more time in decision making process. In current research trend some authors used some standard technique for feature reduction such as PCA, PCNN and neural network, but these methods not consider all features for processing fixed some number of feature. In this dissertation proposed a feature selection and feature reduction method based on improved GSA algorithm. The proposed algorithm select multiple feature for reduction and the reduce feature set participant the process of detection. The reduce feature of network file classified by GSA classification algorithm. The DCA algorithm in the case of small data size, if sizes of data are increase the selection of attribute process raised some problem related to feature selection. For the improvement of this problem used PCNN function for increasing the biased value of feature and feature subset selection. < clear="all" style="page–eak–before:always; mso–eak–type:section–eak" />