Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{174212,
        author = {P. Charishma and P. Hima Varsha and M. Rohit Tatayya and P. Viswa Sai Charan},
        title = {Threat-Eye: Early Malware Detection via Spatiotemporal Pattern Analysis},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {10},
        pages = {3101-3111},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=174212},
        abstract = {Traditional cybersecurity techniques including behavior-based approaches, heuristic analysis, and signature-based detection are seriously challenged by the growing sophistication of malware. These traditional systems have trouble identifying new threats, including advanced persistent threats (APTs), polymorphic malware, and zero-day exploits. In today's constantly-changing cyber threat scenario, proactive, flexible, and scalable malware detection solutions are more important than ever. This study presents THREAT-EYE, a novel early detection approach that uses aberrant spatiotemporal patterns in system logs, network traffic, and user activity to identify malware activities. Through the use of anomaly detection algorithms and machine learning techniques, THREAT-EYE detects minute departures from typical behaviour in both temporal and geographical dimensions, allowing it to identify complex threats that conventional methods frequently overlook. Fundamentally, THREAT-EYE uses a mix of deep learning models and ensemble approaches that continuously pick up patterns of typical behaviour. It can identify anomalies that point to hostile activity, like data exfiltration, lateral network movement, and command-and-control interactions, thanks to this learning process. Because THREAT-EYE relies on anomaly detection rather than signature-based techniques, it can detect new malware variants without being aware of particular signatures beforehand. Because it adjusts its models to take into consideration modifications in user behaviour, network traffic, and virus tactics, the framework's adaptability guarantees its efficacy in dynamic contexts. It is an effective tool for early malware detection because of its scalability in a variety of scenarios and capacity to identify anomalies across several domains.},
        keywords = {Malware detection Anomaly detection, Machine learning, Cybersecurity threats, Spatiotemporal analysis},
        month = {March},
        }