Creating Prognostic Systems by the Mann-Whitney Parameter

We proposed two approaches to compute Mann-Whitney parameter based initial dissimilarities for the Ensemble Algorithm for Clustering Cancer Data (EACCD). These two approaches are non-parametric and produce robust prognostic systems. The breast cancer data from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute were used to demonstrate these two approaches. Results showed that our proposed methods generated prognostic systems with a comparable performance to the AJCC's cancer staging system.