为了解决从指标隶属度到目标隶属度转换过程中存在的冗余数值问题,通过挖掘隐藏在指标隶属度中关于目标分类的知识信息与空间信息,定义指标区分权重,清除指标隶属度中对目标分类不起作用的冗余数值。建立基于"一滤、二比、三合成"的隶属度转换算法,并用于防空通信系统效能评估。研究结果表明:该算法是多属性、多层次综合评价的一种有效方法。
In order to solve the problems of redundant values in converting the membership degrees from indexes to objective,this study has developed a new membership degree conversion algorithm of "filtration,comparsion and synthesis" to evaluate the efficiency on air defense communication system. In the study,the knowledge and space information enbedded in index membership degrees are mined out. Sebsequently,the index distinguish weights are defined and the redundant values,which are useless for classifing index membership degree,are eliminated. The study result concludes that the method developed is an efficient multi-attribute and multi-level comprehensive evaluation method.