基于条件信息熵的决策表约简

Decision Table Reduction based on Conditional Information Entropy

Rough集理论是近年来发展起来的一种有效地处理不精确、不确定、含糊信息的数学理论方法 ,在机器学习、数据挖掘、智能数据分析、控制算法获取等领域取得了很大的成功 .研究者从不同的角度对这个理论进行研究 .本文将从信息论观点出发对 Rough集理论的基本概念和主要运算进行分析讨论 ,通过与 Rough集理论的代数观点进行比较分析 ,得到这两种观点下的一些等价性质和不同的特性 ,并基于条件信息熵提出决策表的约简算法 .

This paper analyzes the information view of rough set theory and compares it with the algebra view of rough set theory. Some equivalence relations and other kind of relations like inclusion relation between the information view and the algebra view of rough set theory are resulted through comparing each other. Two novel heuristic knowledge reduction algorithms are developed based on conditional information entropy, that is, conditional entropy based algorithm for reduction of knowledge with computing core (CEBARKCC) and conditional entropy based algorithm for reduction of knowledge without computing core (CEBARKNC). These two algorithms are compared with a mutual information based algorithm for reduction of knowledge (MIBARK) of Duoqian Miao through theoretical analysis and experimental simulation. CEBARKCC algorithm and CEBARKNC algorithm have good performance in simulation.