Data Mining-based Intrusion Detection System

With the rapid development of computer network technology, people pay more attention to the network security. Attack on the network gradually taken on some new features, just rely on anti-virus software to protect security of the system is not enough.Intrusion detection is a goal of real-time monitoring system to detect intrusion attacks on security technology, has become a research hotspot in the field. However, the traditional intrusion detection system effectiveness, flexibility and scalability are shortcomings. For these deficiencies, the data processing from the point of this article, using data mining method described under the massive intrusion of audit data to establish the model.This article on intrusion detection technology and data mining in-depth study based on the traditional intrusion detection system is not strong adaptability, scalability is poor, can not detect unknown defects in the form of invasion proposed intrusion detection based on data mining model, mainly the following:(1) Intrusion detection technique is described, including intrusion detection features, classification and methods used.(2) Introduced data mining techniques, with emphasis on the association rules algorithm is described.(3) Researched the traditional association rules algorithm efficiency of the existing reference to research results of many articles, presented based on Hash table ordered improved algorithm of association rules.(4) Proposed based on data mining intrusion detection system model, and experiments proved that the system has a higher detection rate and low false alarm rate.