Analysis of Diversity-Accuracy Relations in Cluster Ensemble

Abstract

AbstractEnsemble approaches based on aggregated models have been successfully applied in the context of supervised learning in order to increase the accuracy and stability of classification. Recently, analogous techniques for cluster analysis have been introduced. Research has proved that, by combining a collection of different clusterings, an improved solution can be obtained. Diversity within an ensemble is very important for its success. An ensemble of identical classifiers or clusterers will not be better than the individual ensemble members. However, finding a sensible quantitative measure of diversity in classifier ensembles turned out to be very difficult (Kuncheva 2003; Kuncheva and Whitaker 2003). Diversity in cluster ensembles is considered here. The aim of the research is to look into the relationship between diversity and the accuracy of the cluster ensemble.KeywordsCluster SolutionJaccard IndexRand IndexEnsemble ApproachAdjust Rand IndexThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.