RanSAM: Randomized Search for ABAC Policy Mining.

This paper presents a novel approach for generating Attribute-based Access Control policies from a given Access Control Matrix (ACM). In contrast to the existing techniques for policy mining, which group the desired accesses in the ACM using certain heuristics, we pose it as a search problem in the policy space. A randomized algorithm is then used to identify the policy that best represents the given ACM. Our initial experiments show promising results.