Role mining - revealing business roles for security administration using data mining technology

In this paper we describe the work devising a new technique for role-finding to implement Role-Based Security Administration. Our results stem from industrial projects, where large-scale customers wanted to migrate to Role-Based Access Control (RBAC) based on already existing access rights patterns in their production IT-systems. The core of this paper creates a link between the use of well established data mining technology and RBAC. We present a process for detecting patterns in a data base of access rights and for deriving enterprise roles from these patterns. Moreover, a tool (the SAM Role Miner) is described. The result allows an organized migration process to RBAC with the goal of building a single point of administration and control, using a cross-platform administration tool. Categories and Subject Descriptor