Application of Matrix Factorization Role Mining Algorithm in Role-Based Access Control for Edge RMP

The Edge Role Mining Problem (Edge RMP), a subset of the Role Mining Problem (RMP), has been acknowledged as an efficient approach to role allocation. This method curtails the direct relationships between users and permissions, subsequently minimizing the security risks posed to the system. Despite the existence of an algorithm rooted in Boolean matrix factorization for addressing this issue, it falls short in further refining the resulting User-Role Assignment (UA) and Role-Permission Assignment (PA) relationships. Moreover, when dealing with large datasets, it exhibits considerable redundancy in roles, leaving ample room for further refinement of Edge RMP outcomes. To address these shortcomings, this paper introduces the NEP algorithm, which is built upon Boolean matrix factorization. This methodology achieves substantial optimization in Edge RMP results by adeptly handling the intricate relationships between roles that possess diverse permissions. The proposed approach yields optimal numbers of generated roles and minimizes the sum of edges created, effectively addressing the aforementioned challenges. Experimental results unequivocally demonstrate that the NEP algorithm effectively diminishes management overhead while delivering remarkably efficient execution results.