International Journal of Advanced Research in Computer and Communication Engineering

Abstract: The rapid growth of cyber threats has left traditional signature-based malware detection methods less effective against new and complex attacks. To address this issue, the AI-Powered Malware Detection System identifies malicious software using machine learning, focusing on behavior instead of static signatures. Developed in Python, the system detects and classifies malware and non-malicious software using algorithms like Random Forest and XG Boost. By training on datasets such as the Microsoft Malware dataset and CIC-MalMem, the model identifies complex patterns in system behavior, including file operations, network activity, and process interactions associated with malware. The features extracted are then processed to create a high-performance model that detects malware with low false positives. This system is also resilient to future variant developments, making it more effective than traditional methods. With applications in cybersecurity defense systems, enterprise IT infrastructure, and cloud security, this paper enhances proactive malware detection and improves system resilience against cyberattacks.

Keywords: Malware, cloud security, cyberattacks, Ai, machine learning, cybersecurity.

Downloads: | DOI: 10.17148/IJARCCE.2025.141126