Abstract: In order to improve the accuracy of intrusion detection, this paper puts forward a classifier integration method which uses tabu genetic search. It uses Bagging algorithm to found classifiers. It adopts tabu genetic search to select classifier, and uses the selected classifiers to establish an intrusion detection system. Experimental results show that compared with Bagging algorithm, this method can improve detection accuracy and reduce false alarm rate.

Key words: classifier, genetic tabu search, Bagging algorithm, genetic algorithm

摘要： 为提高入侵检测的精度，提出一种使用遗传禁忌搜索的分类器选择集成方法。该方法采用Bagging算法构建初始分类器集合，根据遗传禁忌搜索算法选择分类器子集，以该子集建立多分类器系统进行入侵检测。实验结果表明，与Bagging算法相比，该方法能有效提高检测精度、降低误报率。

关键词: 分类器, 遗传禁忌搜索, Bagging算法, 遗传算法

CLC Number: 

• TP391