Intrusion Detection for Database Security using a Hidden Naïve Bayes Binary Classifier

Abstract

The Hidden Naive Bayes Binary Classifier is used for Database Security to detect Intruders. Data mining is used a lot in intrusion detection systems to classify normal or anomaly events. This method is a transparent, effective, and widely used mining method based on the idea of conditional attribute independence. HNB classifier is a more advanced version of Naive Bayes classifier algorithm and is efficiently used for intrusion attacks. It keeps the simplicity and efficiency of Naive Bayes, but loosens the independence condition. In the tests, it is proved that this binary classifier model can be used to solve the intrusion detection problem.