Adaptive Detection and Response System for Post-Installation Cyber Attacks on Smartphones

Smartphone cyber-attacks are one of the major security threats of the present-day mobile computing era. Such attacks are used on the weaknesses that arise upon initial device configuration and they tend to extract sensitive data of the users, the system integrity and the functionality of a given device. This paper presents the post-installation attack vectors and detection together with a new Dynamic Security Assessment Framework (DSAF) that can be used to detect and mitigate the attacks in real time. The proposed method integrated behavior analysis, machine learning, and anomaly detection methods to detect post-installed suspicious activity. The proposed method uses two prominent algorithms, i.e., the Adaptive Threat Detection Algorithm (ATDA) and the Risk-Based Response Algorithm (RBRA). We have carried out an experimental study that shows that our strategy reaches a 94.7 percent accuracy in detection with a false positive of 2.3 percent that is much higher than what the current security solutions are capable of. The effectiveness of the framework is proved through thorough testing on 1,500 smartphones of various platforms whereby a 78 percent post-installation invasion is curbed using the framework as compared to conventional security systems.