Design of Intrusion Detection System Based on Logical Analysis of Data (LAD) Using Information Gain Ratio

Sneha Chauhan, Sugata Gangopadhyay

Abstract: An intrusion detection system is proposed which is capable of detecting penetration, break-ins and other security breaches in near real time. The system has been developed using Logical Analysis of Data (LAD) where the attack is detected by monitoring the network traffic. LAD generates positive and negative patterns from historical observations to classify the unknown observations. It uses the concepts of partially defined Boolean functions and its extensions to extract patterns for classification. The Information Gain ratio technique is used to produce the support set of features. The performance of the proposed technique has an advantage over other techniques as it can detect anomalous behavior in near real time. WEKA tool has been used to build the classifiers and their performance is compared with the proposed model. Detection of abnormal behaviour is significantly achieved by the proposed implementation than the LAD-WEKA.