The Repository @ St. Cloud State

Open Access Knowledge and Scholarship

Date of Award

5-2026

Culminating Project Type

Starred Paper

Styleguide

apa

Degree Name

Information Assurance: M.S.

Department

Information Assurance and Information Systems

College

Herberger School of Business

First Advisor

Abdullah Abu-Hussein

Second Advisor

Lynn A. Collen

Third Advisor

Balasubramanian Kasi

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Keywords and Subject Headings

ransomware detection, artificial intelligence in cybersecurity, machine learning malware detection, ai-driven threat detection, deep learning ransomware detection, explainable artificial intelligence, adversarial machine learning, dataset bias in machine learning, behavioral malware analysis, real-time ransomware detection, supervised learning cybersecurity models, unsupervised anomaly detection, scalable ai security systems, cyber threat intelligence, limitations of ai in cyber defense

Abstract

Ransomware has emerged as one of the most disruptive cybersecurity threats, causing significant financial and operational damage to individuals and organizations worldwide. This study examined existing research on Artificial Intelligence (AI) and Machine Learning (ML) approaches for ransomware detection, evaluating their effectiveness, limitations, and areas for improvement. The objective was to assess the current state of AI-driven ransomware detection methods and identify key gaps that hinder their real-world applicability.

A systematic literature review was conducted, analyzing peer-reviewed studies and industry reports that focused on AI-based detection techniques. Findings revealed that AI models, particularly deep learning and ensemble learning methods, demonstrated high accuracy in identifying ransomware threats. However, several challenges were identified, including dataset limitations, susceptibility to adversarial attacks, lack of explainability, and scalability concerns. Many models were trained on constrained datasets, limiting their ability to detect novel ransomware variants effectively. Additionally, issues related to computational efficiency and real-time detection capabilities were noted as critical barriers to widespread adoption.

The study concluded that while AI holds significant potential in ransomware detection, further research is needed to enhance dataset diversity, improve model transparency, and develop lightweight, scalable detection frameworks. These findings contribute to the ongoing discourse on AI-driven cybersecurity, providing insights for researchers and practitioners seeking to strengthen defense mechanisms against evolving ransomware threats.

Comments/Acknowledgements

Acknowledgements

I would like to express my deepest gratitude to my wife, Mamuna, for her unwavering support, patience, and encouragement throughout this academic journey. Her belief in me has been a source of strength and motivation. To my wonderful children, Ellisa, Theo, Giovanni and Ellora, thank you for your love and the joy you bring into my life—I hope this work serves as an inspiration for your own future pursuits.

I extend my sincere appreciation to my professor, Abu-Hussein, whose persistent guidance and relentless encouragement pushed me to complete this paper. His expertise and dedication have been invaluable in shaping my research and refining my academic approach.

A heartfelt thank you to my friends and family, who have provided unwavering support, insightful discussions, and the much-needed motivation to see this project through. Your encouragement has made this journey both fulfilling and meaningful.

This achievement would not have been possible without each of you, and I am profoundly grateful.

Download

Share

COinS