Fuzzy based trust model for cloud access control classification mechanism using ghost net architecture
D. Kalaiyarasi,R. Priscilla Joy,M. Victor Jose,P. Sridhar
DOI: https://doi.org/10.1007/s11276-024-03843-w
IF: 2.701
2024-11-23
Wireless Networks
Abstract:Cloud computing (CC) is the rapid utilization of network resources, specifically processing power and data storage, without the need for extra or direct customer management. One of the crucial and difficult problems facing the scientific and business user communities is ensuring security and trust between users ( ) and Cloud Service Providers (CSP) from malicious users. As a result, it is important to trust and cloud resources before using them. In this research, a novel two-phase Access Control (AC) mechanism based on Deep learning (DL) has been proposed to confirm the reliability of the cloud and the trust of the in the cloud environment. In the first phase, the Mamdani fuzzy method with Trapezoid fuzzy logic (TFL) is applied to assess the trust value (TV) of based on Bogus Requests (BR), Unauthorized Requests (UR), Forbidden Requests (FR), Not Found Requests (NFR), and User Behavior (UB). In the second phase, the ghost net is used for attribute matching for deciding to access or deny the user request based on TV. The efficacy of the developed methodology is examined by employing several parameters, such as Time, Accuracy (Acc), recall, F-measure, and precision. The overall accuracy of the proposed model is 1.41%, 2.61%, and 5.14% better than the Multidimensional Dynamic Trust Evaluation Scheme (MDTES), Evolutionary Genetic Algorithm with Intelligent Rules (EGAIR), and Task-Role Based Access Control (T-RBAC) respectively. In comparison to existing methodologies, the proposed method showed improved efficiency in terms of Acc, and prediction time.
computer science, information systems,telecommunications,engineering, electrical & electronic