Toward edge minability for role mining in bipartite networks

摘要: Abstract Bipartite network models have been extensively used in information security to automatically generate role-based access control (RBAC) from dataset. This process is called role mining . However, not all the topologies of bipartite networks are suitable for mining; some edges may even reduce quality mining. causes unnecessary time consumption as NP-hard. Therefore, promote results, capability that an edge composes roles with other edges, minability , needs be identified. We tackle problem angle importance complex networks; easily covered by considered more important. Based on this idea, k -shell decomposition extended reveal different edges. By way, a can quickly purified excluding low-minability mining, and thus effectively improved. Extensive experiments via real-world datasets conducted confirm above claims.