Tools
Tools
Bhanarkar, Tejas (2025) Multi-Agent Based Distributed Malware Detection Using Static Analysis and Machine Learning. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (2MB) | Preview |
Abstract
As malware continues to evolve in complexity whereas traditional centralized detection systems struggle to deliver timely and scalable responses. This project proposes a decentralized Multi-Agent System (MAS) for malware detection and mitigation which is based solely on static analysis of executable files. The architecture integrates autonomous agents that collaboratively detect, classify, mitigate, and log threats using machine learning model and secure communication protocols. The Detection Agent extracts predefined static features such as Portable Executable (PE) headers and opcode frequencies, while a trained Random Forest classifier determines infection likelihood. The Decision Agent interprets results, triggering appropriate action via the Mitigation Agent and ensuring auditability through a Blockchain Logger Agent. Results demonstrate accurate classification performance, low-latency mitigation, and modular design for scalable deployment on edge, cloud, and hybrid infrastructures. This research validates MAS as a viable solution for distributed malware detection, combining autonomy, security, and adaptability.
Actions (login required)
 |
View Item |