加权条件熵在异常检测中的应用

通过分析信息熵在异常检测中的应用和条件熵的固有特性, 提出了一种基于条件熵和加权熵理论的异常检测方法。分析网络流量特征, 选取具有内在相关性的网络特征属性, 给出了加权系数矩阵的公式和异常检测的具体实现方法。通过实验验证了该方法的有效性和实用性, 根据不同属性特征在加权条件熵理论中的应用效果, 分析了不同特征属性选择造成检测效果差异的原因。

Application of weighted conditional entropy in anomaly detection

This paper proposed an anomaly detection method based on conditional and weighted entropy by analyzing entropy application and conditional entropy inherence property. This paper analysed the character of network traffic, chose the network property with relativity, and proposed the formula of weighted coefficient matrix and method of anomaly detection. The experiment shows the validity and practicability of this method. Meanwhile it analysed the reason of detection difference under different character property chooses base on application effect.