An Improved K-modes Clustering Algorithm Based on Intra-cluster and Inter-cluster Dissimilarity Measure

DOI

10.2991/iccia-17.2017.67How to use a DOI?

Keywords

Clustering, categorical data, dissimilarity measure, k-modes algorithm.

Abstract

Categorical data clustering has attracted much attentions recently because most practical data contains categorical attributes. The k-modes algorithm, as the extension of the k-means algorithm, is one of the most widely used clustering algorithms for categorical data. In this paper, we firstly analyzed the limitations of two existing dissimilarity measures. Based on this, we proposed a novel dissimilarity measure--IID. IID considers the relationship between the object and all clusters as well as that within clusters. Finally the experiments are made on six benchmark data sets from UCI. And the corresponding results show that IID achieves better performance than two existing ones used in k-modes and KBGRD algorithms.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Hongfang Zhou
AU  - Yihui Zhang
AU  - Yibin Liu
PY  - 2016/07
DA  - 2016/07
TI  - An Improved K-modes Clustering Algorithm Based on Intra-cluster and Inter-cluster Dissimilarity Measure
BT  - Proceedings of the 2nd International Conference on Computer Engineering, Information Science & Application Technology (ICCIA 2017)
PB  - Atlantis Press
SP  - 398
EP  - 406
SN  - 2352-538X
UR  - https://doi.org/10.2991/iccia-17.2017.67
DO  - 10.2991/iccia-17.2017.67
ID  - Zhou2016/07
ER  -