基于Manhattan距离的正态云模型相似性度量方法研究  

Manhattan Distance Based Measurement of Similarity for Normal Cloud Model

作　　者：詹媛媛 周金明[1] 陶胜男 李龙莲 ZHAN Yuanyuan;ZHOU Jinming;TAO Shengnan;LI Longlian(School of Mathematics‑Physics and Finance,Anhui Polytechnic University,Wuhu 241000,Anhui,China)

机构地区：[1]安徽工程大学数理与金融学院,安徽芜湖241000

出　　处：《合肥大学学报》2024年第5期9-15,共7页Journal of Hefei University

基　　金：国家社会科学基金项目“基于云理论的非确定性数据综合评价集成技术研究”(20BTJ048);安徽省高校优秀青年人才支持计划项目(GXYQ2020032);安徽工程大学中青年拔尖人才培养计划(S022021054)。

摘　　要：综合考虑正态云的期望曲线与含熵期望曲线,提出了一种基于Manhattan距离的正态云相似性度量方法。首先,将不确定性语言信息转化为定量数值,分别计算群体云评价与最优云、最劣云之间的相似度,采用加权平均集成给出综合相似度,从而得到各方案的排序。最后,通过实例说明该方法的有效性。This article considers the expectation curve and expectation curve with the entropy of normal clouds,and proposes a similarity measurement method for normal clouds based on Manhattan distance.Firstly,the uncertain linguistic information is transformed into quantitative numerical values.Then,the similarity between the group cloud evaluation and the best and worst clouds is calculated separately.The weighted average ensemble is used to obtain the comprehensive similarity,thereby obtaining the ranking of each scheme.Finally,the effectiveness this method are demonstrated through actual examples.

关 键 词：正态云 期望曲线 Manhattan距离 相似度 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]