基于决策熵的不完备信息系统的知识约简方法

Knowledge reduction approaches of incomplete information systems based on decisions information entropy

为有效地解决不完备信息系统的知识约简,得到更优的决策规则集,研究了基于容差关系的决策熵在不完备信息系统中能客观反映决策规则的决策能力,提出了一种基于决策熵的不完备知识约简方法。该方法基于决策熵的思想,考虑了决策规则可信度和对象覆盖度,同时引入了容差关系,以决策熵的属性重要性度量为启发信息进行知识约简,最终得到确定的规则集。仿真实验结果表明了该算法的可行性。

In order to solve the knowledge reduction of incomplete information systems effectively, and get a better set of decision rules, by studying decision entropy based on tolerance relation can reflect the ability of decisions of decision rule objectively in incomplete information systems, knowledge Reduction approaches of incomplete information systems based on decision entropy is presented. The method is based on the idea of the decision entropy, and consider the decision rule credibility and object coverage, and introduce a tolerance relation simultaneously. The importance of the attributes properties of decision entropy as is used heu- ristic information for knowledge reduction, and a set of decision rules are gotten ultimately. The simulation results show that the algorithm prove the feasibility of the proposed algorithm.