Phishing Detection and Mitigation: A Cybersecurity and Machine Learning Approach

Ramesh, Krithika (2024) Phishing Detection and Mitigation: A Cybersecurity and Machine Learning Approach. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (947kB) | Preview
Preview	PDF (Configuration Manual) Download (3MB) | Preview

Abstract

Phishing emails, one of the fastest-growing cybercrimes, make use of human vulnerabilities to leak sensitive data, including financial and login password information. Due to the continuously evolving nature of phishing attacks, traditional methods often fail to detect them and require intelligent solutions. This research aims to perform a comprehensive analysis of cybersecurity frameworks and explore machine learning models to reduce phishing risks. The focus is also majorly on the Naive Bayes method since it is non-iterative; thus, it can manage categorical data and is computationally efficient. The work implements a customized Naive Bayes model which I developed using Google Collab, featuring selection approaches, and data preprocessing techniques to classify the emails into phishing and non-phishing classes. For this, I used Django to create a web interface in order to classify spam and non-spam emails. Accuracy, precision, recall, and F1 score are some of the metrics used to analyze the robustness of the system. Cybersecurity frameworks are recommended as additional steps to prevent phishing scams. Naive Bayes had a better performance compared to other detection techniques and was found to be a reliable tool in email security, which is evident from the accuracy of classification-98%. Its strong sensitivity will guarantee the detection of most phishing emails, and the reasonable specificity reduces false alarms. This paper shows that, due to its simplicity, speed, and accuracy, Naive Bayes is a potential algorithm for phishing email detection. Comparisons with related methods in the literature further support the findings. The practical usefulness of this solution is further enhanced by the integration of cybersecurity and machine learning frameworks. Despite the model's outstanding accuracy, issues like ever-changing phishing strategies and ensuring wider dataset generalization do call for further efforts. Further research will focus on the enhancement of cybersecurity frameworks to address complex threats with the integration of adaptive learning strategies.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Hafeez, Khadija UNSPECIFIED
Uncontrolled Keywords:	Phishing Detection; NIST; ISO 27001; DORA; Naïve Bayes; Decision Trees
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science 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 10:11
Last Modified:	28 Jul 2025 10:11
URI:	https://norma.ncirl.ie/id/eprint/8249

Actions (login required)

View Item