Data Sensitivity – Similarity Based Access Control Mechanism for Mobile Cloud Computing

Abstract

Cloud computing is probably the most paradigms that dominate the information and knowledge Technology (IT) industry these days. It gives new services that are cost-effective such as Platform as a Service (PaaS), Infrastructure as a Service (IaaS), and Software as a Service (SaaS). However, along with these services promising benefits and facilities, there are some challenges connected with utilizing cloud computing such as malicious, abuse of cloud services, cyber-attacks, and data security. Among all security specifications of cloud computing, access control is just one of the fundamental requirements to prevent access that is unauthorized systems and protect the organization's assets. These models may not fulfill cloud's access control requirements although, various access control models and policies have been developed for different environments. Simply because cloud computing has a set that is diverse with various units of security conditions. Also, it has security that is unique such as heterogeneity of security domains, rules, and policies, and multi-tenant hosting. The three-models are proposed to enhance the access control for the sensitive data in this paper. An information gain is used to calculate the data sensitivity. The second data similarity computation, Siamese Neural Network is utilized for taking into consideration the semantic similarity. The model is proposed as SNN with MLP classifier given that the Classification process to categorize the info (or allowing the dataset) predicated on data sensitivity and similarity.