抽象的な

Intrusion Detection System Based on Fuzzy Association Rule with Genetic Network Programming

Harinee.k , Veeramuthu.A

Intrusion detection which classifies the attacks on the Internet from usual behaviour of usage on the Internet. Here intrusion detection systems are vital tool in the cluster environment fight to keep its computing resources secure .It is an unavoidable portion of the information security system. Emerging variety of network behaviours and the rapid development of attack scenarios, it is vital to develop fast machine-learning-based intrusion detection algorithms with high detection rates and low false positive and false negative -alarm rates with the help of association rule mining. In this course of work a fuzzy class-association rule mining method based on genetic network programming (GNP) for intrusion detection. GNP is an evolutionary optimization technique, which uses directed graph structures leads for enhancing the representation ability .In combination with fuzzy set theory and GNP, the proposed work can deal with mixed database that contains both discrete and continuous attributes and also extract many important class association rule .Therefore, the proposed method can be flexibly applied to both misuse and anomaly detection in network-intrusion-detection .It can extract important rules using these tuples and this mechanisms can calculate measurements of association rules directly using GNP which provides detection rate for prediction based approach.