An Artificial Intelligent Enabled Framework for Malware Detection

Abstract

Malware (Malicious Software) has become a severe threat to society, growing in numbers and sophistication daily. Malware writers increasingly use advanced techniques like server-side polymorphism, code obfuscation, and encryption to evade the detection by traditional signature-based malware detection approaches. Several Machine Learning (ML) and Artificial Intelligence (AI) driven approaches have been proposed in the last few years to replace conventional signature-based methods. This chapter presents an intelligent malware detection framework based on static analysis of Windows API calls and PE header files. It uses an ensemble approach and the Chi-square-based feature selection method. The framework also uses locality-sensitive hashing (LSH) to store all previously seen malware and detect known variants to increase computational efficiency. Experimental results demonstrate the effectiveness of the proposed framework. © 2024 selection and editorial matter, Mayank Swarnkar and Shyam Singh Rajput; individual chapters, the contributors.