WEAKNESS OF PERMISSION SYSTEM IN CLOUD CONNECTED MOBILE APPLICATION

Abstract

When mobile applications are utilized in cloud-connected environments, they often seek an excessive amount of permissions, which may lead to security and privacy concerns. Malicious applications have the potential to misuse sensitive data that is accessible to over-privileged programs, hence revealing vulnerabilities in the authorization method. According to the findings of this study, a multi-tiered detection framework that employs machine learning techniques in combination with signature verification, user input, and permission-based analysis is recommended for the purpose of determining if a program is dangerous, non-malicious, or suspicious. In terms of detection accuracy, the results of the experiments demonstrate that the proposed system outperforms the approaches that are currently being used, while at the same time reducing the number of false positives. According to the findings of the research, a strategy that is capable of effectively securing mobile applications contained inside ecosystems that are connected to the cloud might be discovered by combining permission analysis with approaches that are behavioral and user-driven.