Semi-Supervised Possibilistic Fuzzy c-Means Clustering Algorithm on Maximized Central Distance

DOI

10.2991/iccsee.2013.342How to use a DOI?

Keywords

PCM, maximized central distance, semi-supervised clustering, robustness

Abstract

Abandoning the constraint conditions of memberships in traditional fuzzy clustering algorithms, such as Fuzzy C-Means (FCM), Possibilistic Fuzzy c-Means (PCM) is more robust in dealing with noise and outliers. A small amount of labeled patterns guiding the clustering process are easy to be obtained in practical applications. In this study, a novel semi-supervised clustering technique titled semi-supervised possibilistic clustering (sPCM) is proposed. Because the PCM algorithm is easy to fall into identical clusters, we introduce the center maximization to overcome this difficulty. The proposed algorithm makes distance between different classes as far as possible, which can avoid identical clusters. The experimental results demonstrate that the accuracy of the proposed sPCM algorithm has been improved, making algorithm more robust by inheriting the characteristics of PCM.

Copyright

© 2013, 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  - Li Liu
AU  - Xiao-Jun Wu
PY  - 2013/03
DA  - 2013/03
TI  - Semi-Supervised Possibilistic Fuzzy c-Means Clustering Algorithm on Maximized Central Distance
BT  - Proceedings of the 2nd International Conference on Computer Science and Electronics Engineering (ICCSEE 2013)
PB  - Atlantis Press
SP  - 1366
EP  - 1370
SN  - 1951-6851
UR  - https://doi.org/10.2991/iccsee.2013.342
DO  - 10.2991/iccsee.2013.342
ID  - Liu2013/03
ER  -