【摘要】 为了提高网络流量的分类精度,提出一种面向关注应用的多分类器网络流量分类方法.该方法中基分类器是为每个关注的应用而构建的独立分类器,利用行为知识空间对基分类器的输出进行组合并导出最终判定.实验结果表明,该方法能降低多分类器系统构建的复杂性,且分类精度优于单分类器方法.更多还原

【Abstract】 In order to improve the classification precision,this paper puts forward a kind of applicationoriented multiple classifier method to classify network traffic.The basic classifier by this method is dedicated to a kind of application which uses the behavior knowledge space on the output of the basic classifier combination and exports final judgement.Compared with the existing method of multiple classifiers,this method reduces the complexity of constructing multiple classifier systems.The experimental results show that compared with single classifier methods,this method improves the classification precision.更多还原