Association rule based classifier with novel interest measure for intrusion detection

Abstract

newlineIn the third research contribution, a novel interest measure named Rule Precision index (RPI) is introduced which helps us to prune Association rules efficiently and the impact is observed in the classification of attack and non-attack data. Here association rule classification is done by using the relevance vector machine. The performance of the proposed classifier is compared with conventional classifiers against three intrusion detection datasets KDD Cup 99, NSL KDD and CICIDS 2017 Dataset. The overall implementation of the research work is done in the WEKA tool from which it is proved that the proposed methodology can attain an accurate classification outcome which is better than the existing research methodologies.