Clustering with utility function

Hi all, I’m new to Knime and I hope for a little help.

I have 10 attributes (colums) both numeric and categorical, and one more numeric attribute describing an utility function. I want to mine the dataset in order to identify clusters that maximize the sum of the utility scores inside the cluster. In other words, I want to find the cluster(s) of combinations of values of the first 10 attributes so that the sum  of the utility function for the observation in that cluster(s) is the highest possible.

Can you help me in suggesting how I can implement this in Knime? 

Thank you in advance for any support!