Implementing a Hybrid System for Accurately Detecting Phishing URLs with Machine Learning and Deep Learning Techniques

Madeti, Preetham (0022) Implementing a Hybrid System for Accurately Detecting Phishing URLs with Machine Learning and Deep Learning Techniques. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (765kB) | Preview

Abstract

Phishing attacks, especially through deceptive URLs, present a critical challenge in cybersecurity. This study embarks on addressing the increasing complexity of phishing attacks through the lens of advanced computational strategies. At its core, it examines the efficacy of combining machine learning (ML) and deep learning (DL) techniques to enhance the detection of phishing URLs. The primary aim is to surpass the capabilities of traditional detection methods, which are often outpaced by the evolving sophistication of phishing tactics. To this end, the research develops and assesses an array of hybrid models. These models integrate various algorithms, including AdaBoost, Random Forest, Gaussian Naïve Bayes, Decision Trees, and Multi-layer Perceptron (MLP), each chosen for their distinct strengths in data analysis and pattern recognition. A notable outcome of this research is the superior performance of the Random Forest-MLP hybrid model. It exhibits an impressive accuracy of 88%, striking an optimal balance between swift training (768.395 seconds) and quick response in testing phases (0.575 seconds), marking it as a robust solution for real-time threat detection. Other models like AdaBoost-MLP and Gaussian Naïve Bayes-MLP also show commendable accuracy, around 86%, albeit with variations in training and testing durations. The implications of these findings are substantial for the field of cybersecurity. They highlight the versatility and heightened effectiveness of hybrid approaches in countering phishing URLs. This research not only contributes valuable insights to academic discourse but also paves the way for practical advancements in cybersecurity measures, advocating for innovative strategies in the ongoing battle against digital phishing threats.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Qayum, Abdul UNSPECIFIED
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 Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	15 May 2025 16:19
Last Modified:	15 May 2025 16:19
URI:	https://norma.ncirl.ie/id/eprint/7559

Actions (login required)

View Item