基于逆向云和概念提升的定性评价方法

基于逆向云和概念提升的定性评价方法 Qualitative Evaluation Method Based on Backward Cloud and Concept Generalizing 陈昊 CHEN Hao1 1 first-author 李兵 LI Bing1 1 2 武汉大学软件工程国家重点实验室 武汉大学软件工程国家重点实验室 State Key Laboratory of Software Engineering,Wuhan University State Key Laboratory of Software Engineering,Wuhan University 武汉大学计算机学院 武汉大学计算机学院 School of Computer,Wuhan University School of Computer,Wuhan University 基于自然语言的定性评价在人类活动中有着广泛应用.逆向云能有效实现从定量数值到定性概念的转化,已有的逆向云发生器由于估计误差原因,会出现超熵为虚数问题.本文提出的逆向云算法通过合理改进超熵的估计方法,确保了逆向云发生器的有效性.概念提升操作可以将用数字特征表示的原子概念组合成粗粒度的综合概念.将基于新的逆向云发生器和概念提升的定性评价方法用于词汇量定性评价,并与传统的定性评价方法相比较,说明了云定性评价方法的实用性、全面性和有效性. The qualitative evaluation using natural language has the widespread application in the human activities.Backward cloud can transform quantitative values into qualitative concept effectively and naturally,while because of the existing estimation error backward cloud generator may meet problem that the value of hyper-entropy is an imaginary number.A new backward cloud algorithm improves the method of estimating hyper-entropy reasonably to ensure the effectiveness of backward cloud generator.Concept generalizing operation may climb up from fundamental concepts expressed by digital characteristics into the desired conceptual granularity.A qualitative evaluation method based on new backward cloud generator and concept generalizing operation is used for vocabulary qualitative evaluation.Compared with the traditional qualitative evaluation method,it is verified that the cloud qualitative evaluation method is practical,comprehensive and effective. 云模型 逆向云 概念提升 定性评价 cloud model backward cloud concept generalizing qualitative evaluation 10.14188/j.1671-8836.2010.06.019 TP18 国家重点基础研究发展计划（973）项目（2007CB310801）;国家自然科学基金（60873083）;湖北省青年杰出人才基金（2008CDB351）资助项目 2010-06-01 2021-04-01 6 56