Scholarlink

<p>How to construct binary relations from interval-valued information tables is one of the important issues in this field. Existing literature has proposed various construction methods, but the binary relations generated by different methods have significant differences and each has its own clustering advantages. Therefore, how to integrate the advantages of these binary relations to construct a comprehensive aggregated binary relation is a problem worth exploring. This paper proposes an optimized method to aggregate multiple binary relations in interval-valued information tables into an optimal binary relation. Based on distribution difference measurement and combined with KL divergence, this method fuses six existing binary relations into a new aggregated relation, referred to as the collaborative aggregated binary relation and denoted as <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\varvec{R}^{\text {I-VI}}\)</EquationSource> </InlineEquation>. On this basis, this paper further defines two uncertainty measurement indicators: roughness and rough entropy. Experimental results on multiple datasets demonstrate that the binary relation derived from the proposed method outperforms those reported in existing literature across multiple quantitative indicators and exhibits good robustness. Therefore, the binary relation constructed by this method has better ability to deal with interval-valued information tables.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Constructing binary relations in interval-valued information tables using optimized methods

  • 期刊论文

  • 发表日期：

    2026年4月10日

查看全文
  • Zhuyun Dong,
  • Zhaohao Wang

摘要

How to construct binary relations from interval-valued information tables is one of the important issues in this field. Existing literature has proposed various construction methods, but the binary relations generated by different methods have significant differences and each has its own clustering advantages. Therefore, how to integrate the advantages of these binary relations to construct a comprehensive aggregated binary relation is a problem worth exploring. This paper proposes an optimized method to aggregate multiple binary relations in interval-valued information tables into an optimal binary relation. Based on distribution difference measurement and combined with KL divergence, this method fuses six existing binary relations into a new aggregated relation, referred to as the collaborative aggregated binary relation and denoted as \(\varvec{R}^{\text {I-VI}}\) . On this basis, this paper further defines two uncertainty measurement indicators: roughness and rough entropy. Experimental results on multiple datasets demonstrate that the binary relation derived from the proposed method outperforms those reported in existing literature across multiple quantitative indicators and exhibits good robustness. Therefore, the binary relation constructed by this method has better ability to deal with interval-valued information tables.