Cluster Ensemble Selection
Authors:
- 3 November 2008
- journal article
- research article
- Published by Wiley in Statistical Analysis and Data Mining: An ASA Data Science Journal
- Vol. 1 (3) , 128-141
Abstract
This paper studies the ensemble selection problem for unsupervised learning. Given a large library of different clustering solutions, our goal is to select a subset of solutions to form a smaller yet better‐performing cluster ensemble than using all available solutions. We design our ensemble selection methods based on quality and diversity, the two factors that have been shown to influence cluster ensemble performance. Our investigation revealed that using quality or diversity alone may not consistently achieve improved performance. Based on our observations, we designed three different selection approaches that jointly consider these two factors. We empirically evaluated their performance in comparison with both full ensembles and a random selection strategy. Our results indicate that by explicitly considering both quality and diversity in ensemble selection, we can achieve statistically significant performance improvement over full ensembles. Copyright © 2008 Wiley Periodicals, Inc. Statistical Analysis and Data Mining 1: 000‐000, 2008Keywords
This publication has 16 references indexed in Scilit:
- Moderate diversity for better cluster ensemblesInformation Fusion, 2006
- Combining multiple clusterings using evidence accumulationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2005
- Ensemble clustering in medical diagnostics2004
- A Mixture Model for Clustering Ensembles2004
- Ensemble selection from libraries of models2004
- Solving cluster ensemble problems by bipartite graph partitioning2004
- On choosing a dense subgraph2002
- Global land cover mapping from MODIS: algorithms and early resultsRemote Sensing of Environment, 2002
- 10.1162/153244303321897735Applied Physics Letters, 2000
- Clustering and isolation in the consensus problem for partitionsJournal of Classification, 1986