A Systematic Review Comparative Analysis of Machine Learning Algorithms for Malware Classification

Abstract

The increasing prevalence of malware threats has necessitated the development of effective classification techniques to protect systems and networks. This systematic review presents a comparative analysis of various machine learning algorithms employed for malware classification, aiming to identify their strengths, weaknesses, and practical applications. We explore a range of algorithms, including decision trees, support vector machines, neural networks, random forests, and ensemble methods, assessing their performance based on metrics such as accuracy, precision, recall, and computational efficiency. The review encompasses an analysis of feature extraction techniques, dataset characteristics, and evaluation methodologies utilized in the studies, highlighting the impact of these factors on classification outcomes. Additionally, we discuss the challenges faced in malware classification, including data imbalance, evolving malware techniques, and the need for interpretability in machine learning models. By synthesizing findings from the existing literature, this paper aims to provide insights into the current state of machine learning in malware detection and classification, guiding researchers and practitioners in selecting appropriate algorithms for their specific use cases. Ultimately, the review underscores the importance of continuous research and innovation in this field to keep pace with the rapidly evolving malware landscape and enhance cybersecurity defenses.