Copyright © 2026 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{209700,
        author = {Yajuvendra Singh and Anamika Agarwal and Rajveer Shukla},
        title = {AI-Based Early Detection and Mitigation of Ransomware Attacks Using Behavioural Analysis},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {5},
        pages = {2978-2983},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=209700},
        abstract = {Ransomware is a malicious software type which prevents users from accessing files or systems and demands payment from them. While signature-based defense mechanisms can be effective against known threats, they become less effective against new or modified variants since the signature for such variants is unknown. This paper outlines the design of an AI-based system for detecting ransomware early on and safely mitigating its effects using behavioural analysis. The proposed system collects behavioural data (i.e., file operations, process activity, system calls, network behavior, registry modifications, CPU and memory utilization, and file entropy), processes it, selects the most informative features, and develops machine-learning models (Decision Tree, Random Forest, SVM, Logistic Regression, Gradient Boosting, XGBoost and Neural Networks) to categorize activity as Normal, Suspicious or Potential Ransomware. When highly suspicious activity is identified, the system provides early warnings and safe, controlled responses within a laboratory setting. This paper describes the problem formulation, methodology, evaluation plan and anticipated results of the proposed study. It is a research proposal; therefore, no experimental results are presented.},
        keywords = {ransomware detection, behavioural analysis, machine learning, early warning, cybersecurity, malware, anomaly detection, file entropy, controlled mitigation, false-positive reduction},
        month = {October},
        }