Clustering Ensemble based on the Fuzzy KNN Algorithm
Fangfei Weng, Qingshan Jiang, Lifei Chen, Zhiling Hong · 2007
Compared with the single clustering algorithm, Clustering Ensembles are deemed to be more robust and accurate, with combining multiple partitions of the given data into a single clustering solution of better quality. In this paper, we proposed a new Clustering Ensemble algorithm based on Fuzzy K Nearest Neighbor (FKNNCE) to generate the similarity matrix of data to summarize the ensemble and then use hierarchical clustering algorithm to get the final partition, without specified number of clusters in advance. After discussing some related topics, the paper adopts real data and conducts an Intrusion Detection Model to evaluate the performance of the Clustering Ensemble algorithm, furthermore compare it with other algorithms. Experimental results demonstrate the effectiveness of the proposed algorithm.