On the Evaluation of Restricted Boltzmann Machines for Malware Identification

In the last years, tablets and smartphones have been widely used with the very same purpose as desktop computers: web browsing, social networking, banking and others, just to name a few. However, we are often facing the problem of keeping our information protected and trustworthy. As a result of their popularity and functionality, mobile devices are a growing target for malicious activities. In such context, mobile malwares have gained significant ground since the emergence and growth of smartphones and handheld devices, thus becoming a real threat. The main contribution of this paper is to evaluate Restricted Boltzmann Machines (RBMs) for unsupervised feature learning in the context of malware identification. In order to evaluate the results, we employed two supervised pattern recognition techniques, say that Optimum-Path Forest and Support Vector Machines, as well as a classification approach based on RBMs. Keywords—Optimum-Path Forest, Restricted Boltzmann Machines, Malware Detection.