Information-based clustering.

Publication Year
2005

Type

Journal Article
Abstract

In an age of increasingly large data sets, investigators in many different disciplines have turned to clustering as a tool for data analysis and exploration. Existing clustering methods, however, typically depend on several nontrivial assumptions about the structure of data. Here, we reformulate the clustering problem from an information theoretic perspective that avoids many of these assumptions. In particular, our formulation obviates the need for defining a cluster "prototype," does not require an a priori similarity metric, is invariant to changes in the representation of the data, and naturally captures nonlinear relations. We apply this approach to different domains and find that it consistently produces clusters that are more coherent than those extracted by existing algorithms. Finally, our approach provides a way of clustering based on collective notions of similarity rather than the traditional pairwise measures.

Journal
Proc Natl Acad Sci U S A
Volume
102
Issue
51
Pages
18297-302
Date Published
12/2005
Alternate Journal
Proc. Natl. Acad. Sci. U.S.A.