Reduct Driven Pattern Extraction from Clusters

DOI

10.2991/jnmp.2009.2.1.2How to use a DOI?

Abstract

Clustering algorithms give general description of clusters, listing number of clusters and member entities in those clusters. However, these algorithms lack in generating cluster description in the form of pattern. From data mining perspective, pattern learning from clusters is as important as cluster finding. In the proposed approach, reduct derived from rough set theory is employed for pattern formulation. Further, reduct are the set of attributes which distinguishes the entities in a homogenous cluster, hence these can be clear cut removed from the same. Remaining attributes are then ranked for their contribution in the cluster. Pattern is formulated with the conjunction of most contributing attributes such that pattern distinctively describes the cluster with minimum error.

Copyright

© 2009, 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  - JOUR
AU  - Shuchita Upadhyaya
AU  - Alka Arora
AU  - Rajni Jain
PY  - 2009
DA  - 2009/03/01
TI  - Reduct Driven Pattern Extraction from Clusters
JO  - International Journal of Computational Intelligence Systems
SP  - 10
EP  - 16
VL  - 2
IS  - 1
SN  - 1875-6883
UR  - https://doi.org/10.2991/jnmp.2009.2.1.2
DO  - 10.2991/jnmp.2009.2.1.2
ID  - Upadhyaya2009
ER  -