Research And Application On Data Mining In Intrusion Detection System

With the continuous development of the information society , the human society has become increasingly dependent on computer network. At the same time, the computer network security issues are increasingly prominent, which is caused by the open and shared characteristics of network itself.The issue how we can ensure the huge network operates in normal, safe, efficient and smooth way has become a top priority. Intrusion Detection Technology comes into being, which is a new network security technology after the traditional security technology of the firewall and data encryption technology. Intrusion Detection Technology is a pro-active detection of security technology and different to the past passive safety technology, which is a necessary complement to the traditional technology of firewall. Intrusion detection technology applications will greatly enhance the security of the network.This paper aims at the Intrusion Detection Technology based on data mining. The paper describes the basic principles of the intrusion detection and data mining techniques and analysis the main data mining algorithms which is used in the intrusion detection system.The data mining algorithm of Apriori is the main part we discussed in the paper. We improve the traditional Apriori algorithm. We focus on the detection efficiency and two indicators of misrepresentation and omission factor to improve the effectiveness of the invasion identify.Excluding some frivolous rules that have little impact on the outcome to raise the efficiency of the system and that makes it more applicable to intrusion detection systems. The effectiveness is proved by the experiment in the paper. At the end of the paper, we proposed an intrusion detection model based on the improved algorithm .illustrate and analyze the principle of work.