Soft clustering #4
Description
We should investigate the possibility of adding additional column to cluster arrays containing some confidence value. We could potentially use method similar to metric used for calculating cluster quality for eta selection (meaning calculate some ratio between similarity of given sample with samples from its cluster and samples outside of it).