User access group discovery

Abstract

A method, computer program product, and computer system are provided for user access group discovery. The method includes: generating a matrix that depicts the respective resources of a set of resources accessed by each of a plurality of users; obtaining, for each user, a similarity measurement using the matrix, wherein the similarity measurement is based on a comparison of the resources accessed by the user to the resources accessed by the other users; generating, for each user, a feature vector for clustering the users into user access groups, wherein the feature vector represents the resources accessed by the user; applying the feature vectors and the similarity measurements to a centroid-based clustering algorithm to obtain the user access groups; and outputting the user access groups for application in a security environment.

CPC Classifications