Contributed Discussion: Centered Partition Processes: Informative Priors for Clustering

We would like to congratulate the authors on a well written article containing innovative ideas and a compelling application. The authors propose to incorporate subjective prior information on the clustering structure by defining a centred partition process. The proposed family of processes combines well known priors for partitions (specifically exchangeable partition probability functions, EPPFs) with a measure of discrepancy from an initial partition c_0, summarizing prior belief.