A Novel Clustering-Ensemble Approach

Clustering Ensemble is more and more becoming a new hotspot to improve clustering performance since it was firstly presented in 2002. Through using the mechanism of the integrate mechanism of testing and information rolling method to decrease the error probability of matching cluster members, and using a method of category weight to ensemble clustering members, this paper proposes a novel Clustering Ensemble algorithm. Experiments show that this algorithm could effectively improve the clustering results.