Enhancing malware detection through ensemble learning techniques

Enhancing malware detection through ensemble learning techniques

연구 분야: Safety

학회: Cluster Computing

The pervasive threat of malware presents a significant challenge to cybersecurity efforts worldwide, necessitating advanced and adaptive detection mechanisms. Traditional malware detection systems, relying on signature-based or heuristic analyses, struggle to keep pace with the rapidly evolving landscape of malware tactics, techniques, and procedures (TTPs). In response, this paper explores the efficacy of ensemble learning techniques in the context of malware detection, offering three scenarios of ensemble models. The evaluation is conducted using a diverse dataset of malware samples, encompassing a wide range of malware types and behaviors. Through rigorous testing and validation, we demonstrate that ensemble learning models significantly outperform single-model approaches in detecting novel and sophisticated malware. The results highlight the potential of ensemble learning to improve detection accuracy and reduce false positive rates.