【摘要】 针对目前基于K-Means算法的计算机取证技术存在对符号类型数据处理能力欠缺、误报率较高的问题,提出了一种处理混合型属性的聚类算法的计算机取证技术.该方法将对符号类型特征进行编码映射,并使用主成分分析对编码后增加的维数进行降维,从而解决了在计算机取证中使用聚类分析无法对符号型数据进行处理的问题.文中详细的阐述了改进的具体实现方案,并通过实验验证了该方法的可行性.更多还原

【Abstract】 This paper analyses the exists problems of the current computer dynamic forensic techniques base on K-Means algorithm: can not analysis the feature composed by character,higher false-detection rate,etc,brings forward some improvement.We transform the feature of character to numerical value by mapping,then use the technique of Principal Components Analysis to reduce increased dimensionality after mapping.In this paper,we introduce the improved method concretely,and show the feasibility and effect through an experiment.更多还原