An Approach for Privacy-preserving Mobile Malware Detection through Federated Machine Learning
This year’s ITasec24 was the event where CNR researchers presented their new technique for mobile malware detection using Federated Machine Learning. The process of malware detection has traditionally involved transmitting sensitive information to centralized servers leading to serious privacy and data insecurity concerns. In our approach, however, features are extracted directly on the mobile devices and later securely shipped to a server for observation without compromising any user’s raw data. This approach makes it possible to detect and prevent mobile malware efficiently while ensuring that sensitive user data is protected. The utilization of federated learning in mobile security could transform our approach to cyber security significantly through improved detection and prevention capabilities by researchers and professionals while keeping trust and confidence of our users intact. The latter can be applied beyond academia into industries such as Cybersecurity or Mobile App Development among others as more people continue turning towards relying on portable communication devices each day.