COMBATING SOCIAL ENGINEERING THREATS IN MOBILE DEVICES: COMPARATIVE ANALYSIS OF EMERGING THREATS, MACHINE LEARNING, AND USER AWARENESS IN AFRICA

Abstract

Mobile devices have become an integral part of our daily lives, and contain vast amounts of information. However, the ubiquity of mobile technology has made the device a prime target for social engineering attacks by malicious actors, thereby posing significant threats to user privacy and security. To shield mobile users from social engineering threats, this study provides a comparative analysis of the current vulnerabilities, attack vectors, and cutting-edge solutions for safeguarding mobile device users against these nefarious acts. This study adopted a mixed-methods approach to analyze region-specific mobile threats and evaluate tailored defenses in Nigeria. Primary data were collected from 12,000 attack samples during Nigeria’s 2023 mobile banking Trojan outbreak. The data was supported by other security reports e.g, INTERPOL. Benchmarks were established to enable direct comparison of detection rates for African threat variants. Machine learning models were trained on localized features and evaluated using detection accuracy, adoption feasibility and scalability. The study revealed that ML-SMEPF has the high percentage in fraud detection globally with 40% better detection in Africa. Based on this, it is recommended that the use of AI-driven social engineering defense systems that are capable of predicting and neutralizing attacks in real-time be encourage.