Automatic segmentation of a collection of user profiles

A computer program product comprising a non-transitory machine-readable storage medium having machine-readable program code embodied therewith, said machine-readable program code comprising machine-readable program code configured:

• to obtain a plurality of user profiles derived from unstructured data associated with a plurality of users, wherein a given one of the user profiles corresponds to a respective one of the users;
• to repetitively execute at least one machine learning technique on the plurality of user profiles, each execution producing a respective set of one or more segments from the plurality of user profiles, wherein each segment is a collection of users having common attributes;
• to generate a graph by performing pairwise comparisons between collections of users having common attributes from respective executions;
• to apply at least one persistency graph algorithm to the graph to find one or more coherent clusters comprising one or more collections of users having common attributes that are persistent across the repetitive executions of the machine learning technique;
• and
• to produce an output at least in part by selecting at least one of the collections of users having common attributes from at least one of the coherent clusters.