English

Abstract

Cluster ensemble has emerged as a powerful technique for improving robustness, stability, and accuracy of clustering solutions. In this paper we present a novel use of cluster ensemble to handle another most difficult problem in data clustering - a model order selection. Each ensemble component is viewed as an expert domain for building the case-based reasoning. Our proposed method is simple and fast, but effective. Three simulations with different state-of-the-art segmentation algorithms are presented to illustrate the efficacy of the proposed approach. We exten-sively evaluate our approach on a large dataset in comparison with recent approaches for determining the number of regions in segmentation combination framework. Experiments demonstrate that our approach can significantly reduce computational time required by the existing methods, without the loss of segmentation combination accuracy. This contribution would make the segmentation ensemble approach more feasible in real-world applications.