Background and Objective: Java vulnerabilities correspond to 91% of all exploits observed on the World Wide Web. Then, this present work aims to create an antivirus software with machine learning and artificial intelligence, master in Java malwares detection..
Methods: Within the proposed methodology, the suspect Jar sample is executed in order to infect, intentionally, Windows OS monitored in a controlled environment. In all, our antivirus monitors and ponders, statistically, 6,824 actions that the suspected Jar file can do when executed.
Results: Our antivirus achieves an average performance of 91.58% in the distinction between benign and malwares Jar files. Different initial conditions, learning functions and architectures of our antivirus are investigated in order to maximize their accuracy.
Conclusions: The limitations of commercial antiviruses can be supplied by intelligent antiviruses.Instead of blacklist-based models, our antivirus allows Jar malware detection in a preventive way and not in a reactive manner as Oracle's Java and traditional antivirus modus operandi.