A Decision Model for Fuzzy Clustering Ensemble
Yanqiu Fu, Yan Yang · 2007
Recent researches and experiments have showed that clustering ensemble approaches can enhance the robustness and stabilities of unsupervised learning greatly.Most of them focused on crisp clustering combination.However, in this paper, we offer a decision model based on fuzzy set theory for fuzzy clustering ensemble.Firstly, o btain the optimal partition called " expert" from H individual fuzzy partitions generated by fuzzy c-means algorithm.Then, use fuzzy voting scheme to generate the majority judger.Finally, the two matrixes are combined by Decision Model.Experimental results show the effectiveness of the proposed method comparing to the results based on crisp clustering.