Semi-supervised Gaussian Mixture Models Clustering Algorithm Based on Immune Clonal Selection

DOI

10.2991/nceece-15.2016.214How to use a DOI?

Keywords

Semi-supervised clustering; Gaussian Mixture Models; immune clonal selection.

Abstract

Semi-supervised Clustering with constraints is an active area of machine learning and data mining research.. Shental used the Expectation Maximization (EM) procedure to handle semi-supervised Gaussian Mixture Models (GMM) estimation, in which positive and negative constraints are incorporated with to improve clustering results. However the conventional EM algorithm only produces solutions that are locally optimal, and thus may not achieve the globally optimal solution, and it is sensitive to initialization, moreover, the number of components of mixture model must be known in advance. This paper introduces the artificial immune clonal selection algorithm into semi-supervised GMM-based clustering techniques, where the EM algorithm is incorporated with the ideas of a clonal selection algorithm. The new algorithm overcomes the various problems associated with the traditional EM algorithm. It can improve the effectiveness in estimating the parameters and determining simultaneously the optimal number of clusters automatically. The experimental results illustrate the proposed clustering algorithm provides significantly better clustering results.

Copyright

© 2016, 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  - Wenlong Huang
AU  - Xiaodan Wang
PY  - 2015/12
DA  - 2015/12
TI  - Semi-supervised Gaussian Mixture Models Clustering Algorithm Based on Immune Clonal Selection
BT  - Proceedings of the 2015 4th National Conference on Electrical, Electronics and Computer Engineering
PB  - Atlantis Press
SP  - 1204
EP  - 1210
SN  - 2352-5401
UR  - https://doi.org/10.2991/nceece-15.2016.214
DO  - 10.2991/nceece-15.2016.214
ID  - Huang2015/12
ER  -