Tools
Tools
Muralidharan, Akash (2024) Detection of Phishing Websites Using Machine Learning. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (948kB) | Preview |
Preview |
PDF (Configuration Manual)
Download (474kB) | Preview |
Abstract
Phishing attacks pose a significant threat to both corporate and personal internet users, compromising personal data and financial security. As attackers evolve their tactics, there is an urgent need for updated detection strategies. Machine learning models offer distinct advantages in detecting phishing attacks by learning patterns from large datasets and therefore providing more accurate and efficient predictions of evolving threats than traditional rule-based methods. This research examines how machine learning and AI can help prevent phishing attacks and by analysing website features, these models can effectively identify malicious sites, offering a more accurate and adaptive approach to detecting phishing threats. The performance of these models was evaluated on metrics like accuracy, precision, recall and F1 score, and this was done to obtain a better analysis of how these models perform. The results suggest that ensemble methods such as Random Forest are robust which gives high precision and recall for the prediction of phishing efforts. The system demonstrates the possibility to adapt to a variety of phishing patterns, and scales efficiently thanks to preprocessing. This solution represents an attainable and scalable phishing detection framework that promises practical action for organizations toward mitigating the defense against ever changing threats. Though the results are promising, more work is needed to integrate real time detection capabilities and extend the evaluation across larger and dynamic datasets. A huge gain in detection accuracy is also achieved by exploring deep learning models in this case.
Actions (login required)
 |
View Item |