一种基于条件熵的增量式属性约简算法

Incremental Algorithm for Attribute Reduction Based on Conditional Entropy

粗糙集是一种处理不确定、不完全知识的数学工具,属性约简是粗糙集理论的重要研究内容之一。提出了一种基于条件熵的快速增量约简方法,主要分析了在对象动态增加情况下信息熵的变化机制。该算法通过判断更新前决策表的约简属性对新增对象的区分情况来计算新的条件熵值,就可以快速求解出更新后的决策表的属性约简结果。实验结果也进一步验证了该方法的有效性。

Rough set theory is a mathematic tool to deal with incomplete and uncertain information,in which attribute reduction is one of important issues.The changing mechanism of condition entropy was analyzed when a new object was added to the original decision table.Based on this mechanism,a new incremental algorithm for attribute reduction was proposed.In this algorithm we divided the added objects into three cases.Furthermore,by these different cases incremental attribute reduces could be calculated quickly.At last,the validity of the proposed algorithm was depicted by an experiment.