Scholarlink

Clustering aims at grouping data into homogeneous clusters. However, setting parameters such as a cluster count or a dissimilarity bound can be challenging. This paper introduces Curgraph, a novel iterative approach for radius-constrained clustering. Curgraph identifies optimal partitions with respect to maximum cluster radius by computing minimum dominating sets across partial graphs. Experimental results demonstrate Curgraph’s effectiveness compared with state-of-the-art algorithms.

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

Iterative Graph-Based Radius-Constrained Clustering

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Quentin Haenn,
  • Brice Chardin,
  • Mickael Baron,
  • Allel Hadjali

摘要

Clustering aims at grouping data into homogeneous clusters. However, setting parameters such as a cluster count or a dissimilarity bound can be challenging. This paper introduces Curgraph, a novel iterative approach for radius-constrained clustering. Curgraph identifies optimal partitions with respect to maximum cluster radius by computing minimum dominating sets across partial graphs. Experimental results demonstrate Curgraph’s effectiveness compared with state-of-the-art algorithms.