Recent Advancements in Machine Learning Models for Malware Detection: A Systematic Literature Review

Malware detection has become a critical area of research due to the increasing sophistication of cyberattacks targeting various platforms, including IoT devices, Android systems, and desktop environments. This study employed the systematic literature review (SLR) method, following PRISMA guidelines, to analyze recent advancements in malware detection using machine learning (ML) models. A total of six studies were selected based on strict inclusion and exclusion criteria, focusing on algorithms, datasets, performance metrics, and targeted platforms. The review reveals that ensemble methods like Gradient Boosting and XGBoost achieve high detection accuracy, with several models exceeding 90% on benchmark datasets such as VirusShare and MSCAD. Additionally, IoT platforms emerged as the most commonly targeted environment in malware detection research, emphasizing their vulnerability. Despite these advancements, the review identifies gaps in dataset diversity and platform-specific optimizations. This study provides insights into the current trends, challenges, and future directions for machine learning-based malware detection.