Data Mining-based Intrusion Detection Technology

With the rapid development of Internet, people enjoy the convenience of the network but also by the growing threat from the network, so security technology as network attacks and intrusion prevention, an important means of dealing with cyber threats. Intrusion detection technology as a proactive security measures to effectively make up the traditional network security protection technology deficiencies in the current network security intrusion detection technology in the field of technology is one of the most important. Intrusion detection technology in the introduction of data mining technology, intrusion detection technology has the ability to learn, can automatically identify intrusions and intrusion rules to update the library to improve intrusion detection in network security defenses.This article uses data mining techniques in frequent pattern growth (FP increase) association rules algorithm and the Naive Bayes algorithm for classification rules, and the frequent pattern growth (FP growth) algorithm is optimized association rules, proposed a system model, and it carried out a detailed design and implementation of specific modules. The work done in the following points:WEB server by browsing from logging into ACCESS database, delete the unwanted records in the database.Optimization of frequent pattern growth (FP increase) association rules algorithm by using the database to solve because the database is too large in the memory load on the system performance.A combination of Naive Bayes algorithm for classification rules applied to the classification analysis. Intrusion detection in the misuse detection.Finally, the experimental environment, the use of VC6.0+ACCESS implemented a simple system model.