科研之友
微信
新浪微博
Facebook
分享链接
摘要
This paper proposes a method to network intrusion detection based on fast inductive reasoning-FNIDRLA (Fast Network Intrusion Detection Rule Learning Algorithm). Inductive reasoning method can build detecting model with double profiles and accordingly overcome the disadvantage that confidence in the detecting results from anomaly detection model with single profile is not comparatively high. In FNIDRLA, by modifying the rule growing and rule pruning procedures of the traditional rule learning algorithm, the detecting efficiency of the model is greatly improved in the case that the generalization performance of the model doesn't decline at all. The experiment results validate the proposed intrusion detection method.
- 出版日期2005
- 单位哈尔滨工程大学
