粗糙集理论决策规则确定性因子在医疗诊断中的改进方法

An improved diagnosing method backed by deterministic factors of decision rules in rough set theory

通过医生诊断病人是否患流感的决策表为实例,对以元素个数及元素支持数2种方法确定的决策规则确定性因子进行对比,系统地阐明了以元素个数确定的决策规则中,不确定规则的确定性因子对医生诊断病症没有任何参考作用,从而引入了由大量历史数据确定的附带支持数的决策表,通过支持数确定的决策规则的规则确定性因子对医生诊断病症具有很好的参考价值.

The paper uses the decision tables adopted by doctors for diagnosing flu, and compares the deterministic factors of decision rules in the two methods: the number of elements and the support number of elements. It gives a systematic analysis and reaches the following conclusion: in the decision rules determined by the number of elements, the deterministic factors without decision rules are meaningless to the doctor’s diagnosis. It proposes a decision table dependent on a large number of historical data plus a support number, and the deterministic factors of decision rules backed by the support number have much reference value to the doctor’s diagnosing diseases.