Enhancing Cybersecurity through AI-Driven Threat Detection Systems

Srinivasulu, Harini (2024) Enhancing Cybersecurity through AI-Driven Threat Detection Systems. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (2MB) | Preview

Abstract

As organizations increasingly depending on the digital infrastructures, the complexity and frequency of cybersecurity threats have grown significantly. These new threats are hard to tackle by traditional security mechanisms, thereby jeopardizing the security of important systems. In response to this challenge, AI and ML have become innovative solutions in cybersecurity, which provide timely solutions for threat identification. This research study aims to evaluate the performance of the five different algorithms of machine learning, namely KNearest Neighbors (KNN), Decision Trees, Logistic Regression, Random Forest, and Deep Neural Networks (DNN) in identifying DDoS attacks from the network traffic data. Through the performance analysis of these models, it is also possible to compare the effectiveness of the different approaches to traffic classification with the goal of identifying the best approach for the identification of traffic type as benign or DDos malicious. The findings indicates that the proposed Deep Neural Network model provides the highest accuracy of 99.92% & is 100% precise, recall, and F1score for both classes. The Random Forest model also had high accuracy of 98.26% while KNN, Decision Tree, and Logistic Regression models had accuracy of 95.67% – 96.94%. These results emphasize the possibility of the application of AI-based systems for the faster and more accurate identification of cyber threats compared to conventional techniques. The study shows how which machine learning model to use for threat detection and how AI can be used for the prevention of threats in digital environments. As for the limitations of this study, future work will be devoted to improving the real-time detection rates and investigating the possibility of detecting multiple classes of attacks, which will improve the existing state of affairs in the sphere of cybersecurity.

Item Type:	Thesis (Masters)
Supervisors:	Name Email McLaughlin, Eugene UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Cyber Security
Depositing User:	Ciara O'Brien
Date Deposited:	28 Jul 2025 11:35
Last Modified:	28 Jul 2025 11:35
URI:	https://norma.ncirl.ie/id/eprint/8263

Actions (login required)

View Item