Fuzzy Bayesian validation for cluster analysis of yeast cell-cycle data

Clustering for the analysis of the genes organizes the patterns into groups by the similarity of the dataset and has been used for identifying the functions of the genes in the cluster and analyzing the functions of unknown genes. Since the genes usually belong to multiple functional families, fuzzy clustering methods are more appropriate than the conventional hard clustering methods which assign a sample to only one group. In this paper, a Bayesian-like validation method selecting a fuzzy partition is proposed to evaluate the fuzzy partitions effectively. The theoretical interpretation of the obtained memberships is beyond the scope of this paper, and an empirical evaluation of the proposed method is conducted by comparing to the four representative conventional fuzzy cluster validity measures in four well-known datasets. Analysis of yeast cell-cycle data follows to evaluate the proposed method.