Deep Learning Models for Malware Family Detection
Jenna Snead, Krishnendu Ghosh · 2025
Malware attacks have been rising, and the response for detection and mitigation have fallen short to counter the threats. The advances in artificial intelligence have not yielded a comprehensive solution in early detection of mal ware and its variants, making this a critical gap. Our work leverages on the biology-inspired analysis of malware trace sequences and clustering based on trace similarity computation. The malware traces extracted from the community detection algorithms are assumed to represent malware families precisely. The deep learning framework uses the most prototypical features from the extracted traces for prediction of mal ware families. In this work, deep learning models in the form of artificial neural networks (ANNs) are evaluated and experimental results are presented. The results elucidate the deep learning framework for malware family detection across different variations of detection and training splits.