Fig. 2From: Identifying gene-specific subgroups: an alternative to biclusteringCritical difference of ranks. Comparison between the average rank of each algorithm over N=35 datasets, with a 5% level of significance and Hochberg’s correction for multiple testingBack to article page