A data partitioning approach for hierarchical clustering

Seok Ho Yoon*, Suk Soon Song, Sang Chul Lee, Kyo Sung Jeong, Sang Wook Kim, Sooyong Kang, Yong Suk Choi, Jaehyuk Cha, Minsoo Ryu, Byung Soo Jeong

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review

3 Citations (Scopus)

Abstract

In this paper, we propose a parameter-insensitive data partitioning approach for Chameleon, a hierarchical clustering algorithm. The proposed method splits a given dataset into every possible number of clusters by using existing algorithms that do allow arbitrary-sized sub-clusters in partitioning. After that, it evaluates the quality of every set of initial sub-clusters by using our measurement function, and decides the optimal set of initial sub-clusters such that they show the highest value of measurement. Finally, it merges these optimal initial sub-clusters repeatedly and produces the final clustering result. We perform extensive experiments, and the results show that the proposed approach is insensitive to parameters and also produces a set of final clusters whose quality is better than the previous one.

Original languageEnglish
Title of host publicationICUIMC 2013
Subtitle of host publicationProceedings of the 7th International Conference on Ubiquitous Information Management and Communication
Place of PublicationNew York
Pages1-4
Number of pages4
DOIs
Publication statusPublished - 2013
Externally publishedYes
Event7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013 - Kota Kinabalu, Malaysia
Duration: 17 Jan 201319 Jan 2013

Other

Other7th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2013
CountryMalaysia
CityKota Kinabalu
Period17/01/1319/01/13

Keywords

  • Data partitioning
  • Hierarchical clustering
  • Parameter-insensitive

Fingerprint Dive into the research topics of 'A data partitioning approach for hierarchical clustering'. Together they form a unique fingerprint.

Cite this