Scholarlink

In this chapter, we describe the clustering algorithms ComSep and ClusCo, for finding compact and well-separated clusters. Both the algorithms described in this chapter are based on nonsmooth optimization approaches and use incremental approaches to find good-quality starting points. The ComSep algorithm is based on separation error, while ClusCo uses the silhouette coefficient as a constraint to the clustering problem. Detailed descriptions and step-by-step formulation of these algorithms are given.

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

Finding Compact and Well-Separated Clusters

  • 章节

  • 发表日期：

    2025年

查看全文
  • Adil Bagirov,
  • Napsu Karmitsa,
  • Sona Taheri

摘要

In this chapter, we describe the clustering algorithms ComSep and ClusCo, for finding compact and well-separated clusters. Both the algorithms described in this chapter are based on nonsmooth optimization approaches and use incremental approaches to find good-quality starting points. The ComSep algorithm is based on separation error, while ClusCo uses the silhouette coefficient as a constraint to the clustering problem. Detailed descriptions and step-by-step formulation of these algorithms are given.