Enhanced Ransomware Detection on Android Devices Using Traffic Analysis and Advanced Feature Selection Techniques

Android phones, in particular, are highly vulnerable to attacks launched by ransomware, which is widely known to be one of the most common and devastating forms of cyber-attacks that we encounter in the current digital environment. In spite of the application of various techniques for the purpose of identifying such malicious attacks, most of these techniques are suffering from serious limitations, such as extremely low detection rates as well as the limitation of increased computational cost. With the purpose of addressing this hot topic, this paper proposes a new technique for the identification of ransomware attacks launched on Android phones with the assistance of network traffic analysis as a main approach. The proposed technique is developed with the purpose of enhancing the performance and accuracy of machine learning (ML) models, with the assistance of sophisticated techniques such as feature reduction and feature selection in order to achieve its goals.