山西大学学报（自然科学版）2025，Vol.48Issue(1)：77-88,12.DOI:10.13451/j.sxu.ns.2024150
基于自信息和模糊邻域条件熵的特征选择方法
Feature Selection Method Based on Self-information and Fuzzy Neighborhood Conditional Entropy
摘要
Abstract
The feature selection method for fuzzy neighborhood rough sets usually only considers the classification information in the approximation,but cannot evaluate the classification information in the approximation and boundary domains.In this paper,we pro-pose a feature selection algorithm based on self-information measure and fuzzy neighborhood conditional entropy.Firstly,three mea-sures of self-information uncertainty are proposed by combining the lower approximation,the upper approximation and the bound-ary domain,and the similarity self-information is proposed by combining the three types of self-information.Secondly,from the per-spective of information theory,the uncertainty measure of fuzzy neighborhood conditional entropy is given,and combined with simi-lar self-information,a more comprehensive feature evaluation function is proposed to measure the uncertainty of feature subset clas-sification information,and based on this,the feature selection algorithm is designed by using the maximum correlation and mini-mum redundancy technology.Finally,through comparative experiments on the dataset,the results show that the proposed algorithm can effectively process the classification information in the approximation and boundary domains;and under the two classifiers of the proposed algorithm,its average classification accuracy is improved by 2.55％and 4.15％,respectively,in the low-dimensional da-ta set compared with the existing algorithms,and is improved by 0.83％and 2.54％,respectively,in the high-dimensional data set.关键词
模糊邻域粗糙集/自信息/不确定性度量/模糊邻域熵/模糊邻域条件熵Key words
fuzzy neighborhood rough set/self-information/uncertainty measure/fuzzy neighborhood entropy/fuzzy neighborhood conditional entropy分类
信息技术与安全科学引用本文复制引用
徐久成,段江豪,牛武林,张杉,白晴..基于自信息和模糊邻域条件熵的特征选择方法[J].山西大学学报（自然科学版）,2025,48(1):77-88,12.基金项目
国家自然科学基金(61976082 （）
62076089 （）
62002103) （）
