入侵检测系统中贝叶斯分类器的改进

Improvement of Bayesian Classifier for Intrusion Detection

针对入侵检测系统在实时检测能力和自适应能力方面的不足,提出了一个改进贝叶斯分类器,通过引入滑动窗口技术改善入侵检测的实时性。通过该性能调节器对贝叶斯分类器中参数的动态设置,实现了入侵检测系统的自适应性,有效地实现了入侵检测的实时性、主动性和自适应性。

Aimed at the deficiency between real - time defect and auto adjust of intrusion detection system, an improved Bayesian classifier is implemented. The skipping window technology is used to reform the reaction time of the IDS, and at the same time it can auto adjust the system parameters to reform the running performance of the IDS by using the designed performance adjuster. Improved Bayesian classifier can effectively reform the ability of the real - time, active and auto adjust of IDS.