Two types of Tolerant Hard c-Means Clustering

• HAMASUNA Yukihiro

  University of Tsukuba

• ENDO Yasunori

  University of Tsukuba

Abstract

In this paper, we will propose two types of tolerant hard c-means clustering (THCM). One is an alternating optimization form and the other is a sequential algorithm. Introducing a concept of clusterwise tolerance, we have proposed tolerant fuzzy c-means clustering from the viewpoint of handling data more flexibly. In the concept of clusterwise tolerance, a constraint for tolerance vector which restricts the upper bound of tolerance vector is used. First, the concept of clusterwise tolerance is introduced into hard c-means clustering. Second, optimization problem for tolerant hard c-means clustering is formulated. Third, new clustering algorithms are constructed based on the explicit optimal solutions. Finally, effectiveness of proposed algorithms is verified through numerical examples.

Journal

• IEICE Proceeding Series

  IEICE Proceeding Series 43 C1L-B4-, 2009-10-18

  The Institute of Electronics, Information and Communication Engineers

Details

• CRID: 1390001277356845568
• NII Article ID: 230000008322
• DOI: 10.34385/proc.43.c1l-b4
• ISSN: 21885079
• Text Lang: en
• Data Source:
  • JaLC
  • CiNii Articles
• Abstract License Flag: Disallowed