Cluster Ensemble Selection

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 but 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 performances in comparison with both full ensembles and a random selection strategy. Our results indicated that by explicitly considering both quality and diversity in ensemble selection, we can achieve statistically significant performance improvement over full ensembles.

On Ambiguity Detection and Postprocessing Schemes Using Cluster Ensembles

“…Future work is to increase the number of clusterers in the ensemble and investigate ensemble selection approaches (Fern and Lin, 2008) in order to avoid the potential degradation of the ensemble performance if a significant number of "bad" clusterers inappropriate for a dataset are present among the ensemble components.…”

Generalized hard cluster analysis

“…One practical advantage is that if two different hard clustering algorithms are applied to the same data set, results are mostly different. It is considered to be very hard to find an optimal way to combine these different clusterings [3,5]. The basic reason for the difficulty of combining different clustering results is the inconsistency between the clusterings, more precisely the fact that while a data element belongs to one cluster according to the first algorithm, it belongs to another cluster according to the second algorithm.…”

Privacy-Preserving Clustering on Distributed Databases: A Review and Some Contributions

“…Oza & Tumer (2008) do the same in a more recent work, in which they present real applications, where using classifier ensembles has been obtaining a greater success in comparison to using individual classifiers, including remote sensoring, medicine and pattern recognition. Fern (2008) analyses how to combine several available solutions to create a more effective cluster ensemble, based on two critical factors in the performance of a cluster ensemble: quality and diversity of solutions. Leisch (1998), one of the pioneers in the branch of cluster ensembles, introduced an algorithm named bagged clustering, which performs several instances of K-means algorithm, in the attempt of obtaining a certain stability in the results and combines partial results through a hierarchical partitioning method.…”