Clustering ensembles using genetic algorithm

The Clustering ensembles combine multiple partitions of a given data into a single clustering solution of better quality. Clustering ensembles has emerged as a powerful method for improving both the robustness and the stability of unsupervised classification solutions. One of the major problems in clustering ensembles is the consensus function. Finding final partition from different clustering results needs expertness and robustness. In this paper we proposed the genetic algorithm in combination with co-association function as consensus function. With special mutation and one point crossover; GA tries to obtain the best partition. It refers to co-association function values for fitness function parameters. Fast convergence, simplicity, robustness and high accuracy are the most properties Of the proposed algorithm. Experimental results illustrated the effectiveness of the proposed method on common datasets.