基于 Jensen-Shannon 散度的严格直觉模糊距离/相似度度量
摘要
作为一对对偶概念,归一化距离与相似度度量是直觉模糊集框架下决策与模式识别的重要工具。为在决策与模式识别应用中更有效,良好的归一化距离度量应确保其对偶相似度度量满足公理化定义。本文中,我们首先构造若干例子说明 [A distance measure for intuitionistic fuzzy sets and its application to pattern classification problems, \emph{IEEE Trans. Syst., Man, Cybern., Syst.}, vol.~51, no.~6, pp. 3980--3992, 2021] 与 [Intuitionistic fuzzy sets: spherical representation and distances, \emph{Int. J. Intell. Syst.}, vol.~24, no.~4, pp. 399--420, 2009] 中引入的两个非线性距离度量的对偶相似度度量不满足直觉模糊相似度度量的公理化定义。我们表明:(1) 它们不能有效区分某些具有明显大小关系的直觉模糊值(IFVs);(2) 除端点外,存在无穷多对 IFV,在这些距离下可达到最大距离 1;导致反直觉结果。为克服这些缺陷,我们引入严格直觉模糊距离度量(SIFDisM)与严格直觉模糊相似度度量(SIFSimM)的概念,并提出一种基于 Jensen-Shannon 散度的改进直觉模糊距离度量。我们证明:(1) 它是 SIFDisM;(2) 其对偶相似度度量是 SIFSimM;(3) 其诱导熵是直觉模糊熵。对比分析与数值例子表明,我们所提距离度量完全优于现有度量。
引用
@article{arxiv.2207.06980,
title = {Strict Intuitionistic Fuzzy Distance/Similarity Measures Based on Jensen-Shannon Divergence},
author = {Xinxing Wu and Zhiyi Zhu and Guanrong Chen and Tao Wang and Peide Liu},
journal= {arXiv preprint arXiv:2207.06980},
year = {2023}
}
备注
Please replace the first version, because the first version lacks copyright description and has some errors