Information-based clustering Journal Article

Author(s): Slonim, N.; Atwal, G.; Tkačik, Gašper; Bialek, William S
Article Title: Information-based clustering
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.
Keywords: Environment; Algorithms; Gene Expression Regulation; Fungal; Gene Expression Profiling; Saccharomyces cerevisiae/genetics; Cluster Analysis
Journal Title: PNAS
Volume: 102
Issue 51
ISSN: 1091-6490
Publisher: National Academy of Sciences  
Date Published: 2005-12-20
Start Page: 18297
End Page: 18302
DOI: 10.1073/pnas.0507432102
Open access: yes (repository)
IST Austria Authors
  1. Gasper Tkacik
    67 Tkacik
Related IST Austria Work