Open Access

ARTICLE

Enrichment Procedures for Soft Clusters: A Statistical Test and its Applications

R.D. Phillips¹, M.S. Hossain¹, L.T. Watson^1,2, R.H. Wynne³, Naren Ramakrishnan¹
1 Department of Computer Science, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0106, USA.
2 Department of Mathematics, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0123, USA.
3 Department of Forest Resources and Environmental Conservation, Virginia Polytechnic Instituteand State University, Blacksburg, VA 24061-0324, USA.

Computer Modeling in Engineering & Sciences 2014, 97(2), 175-197. https://doi.org/10.3970/cmes.2014.097.175

Abstract

Clusters, typically mined by modeling locality of attribute spaces, are often evaluated for their ability to demonstrate ‘enrichment’ of categorical features. A cluster enrichment procedure evaluates the membership of a cluster for significant representation in predefined categories of interest. While classical enrichment procedures assume a hard clustering definition, this paper introduces a new statistical test that computes enrichments for soft clusters. Application of the new test to several scientific datasets is given.

---

Keywords

Cluster enrichment, fuzzy clustering, statistical significance test.

---