Research On The Bayes Classification In IDS

The security of information is becoming more and more critical. However, network is vulnerable by outside attacks and destruction because of its openness. Thus it is a big challenge to maintain the secrecy and security of information. Intrusion Detection is a very important field in maintaining computer and network security and its technique is an important supplement to other techniques of information security.Firstly, IDS (Intrusion Detection System) based on Bayesian Classification is introduced. Secondly, three Bayesian Classifications--Naive Bayesian Classification, Bayesian Network Classification and Incremental Bayesian Classification are introduced and analysed. Lots of algorithms used in ID were limited in practicing for consuming time. It can be more efficient and rapidly by using rough set in attribute reduction. Finally, this paper presentes a new algorithm for redundant attribute reduction based on the rough set, and then classified by Bayesian Classifications. Experiment results show that Naive Bayesian Classification based on this method is more efficient and has better performance and less time in IDS.Experiments show that this new algorithm for redundant attribute reduction based on the rough set is effectively and better than classical redundant attribute reduction software Rosetta in the privacy, accuracy and complexity.