ACCESS control is the process of mediating requests to data and services maintained by a system, determining which requests should be granted or denied. Significant research has focused on providing formal representation of access control models. Role- Based Access Control (RBAC) has become the norm in most organizations. This success is greatly due to its simplicity: a role identifies a set of permissions; users, in turn, are assigned to roles based on their responsibilities. To implement a RBAC system, it is important to devise a complete set of roles. This design task, known as role engineering, has been recognized as the costliest part of a RBAC-oriented project. We propose a new role engineering approach to Role-Based Access Control (RBAC), referred to as visual role mining. The key idea is to graphically represent user-permission assignments to enable quick analysis and elicitation of meaningful roles. We propose two algorithms: ADVISER and EXTRACT. A heuristic algorithm ADVISER is used to sort the users and Permission matrix. To avoid large gaps between items, compute the similarity between items using Jaccard coefficient. EXTRACT is a probabilistic algorithm and it generates a list of pseudoroles. We offer a graphical way to effectively navigate the result.