Figure 2

Clustering results using k-means and FCM. The cluster results from different cluster methods are compared using z-score, a measurement based on the mutual information between cluster membership and known gene attributes. Three clustering results are plotted: k-means clustering and FCM clustering with two m values (m is the fuzziness parameter): default value (m = 2) and optimal value (m = 1.1573). K-means outperforms FCM with default m value, whereas FCM with the optimal m value yields the highest z-score for cluster numbers ranging from 2 to 100. This demonstrates that FCM clustering with optimal m value has the potential to detect the underlying data structure with biological significance.