BaitNet: A Deep Learning Approach for Phishing Detection

Recent years have witnessed a surge in phishing attacks, propelled by advancements in technology. This paper addresses the evolving threat landscape, particularly the rise in deception scams and highly sophisticated online attacks. This work presents a novel approach to detecting phishing sites using deep learning. Here, A URL-centric solution called BaitNet, employs a Convolutional Neural Network (CNN) model to significantly enhance accuracy in phishing website detection. BaitNet achieves outstanding performance metrics, including a precision rate of 98.99%, an F1 score of 99.00, and an accuracy of 98.44%. These results underscore its exceptional efficiency in safeguarding Internet users from phishing threats. This research underscores both technological innovation and the urgency of protecting individuals and organizations in an era of evolving cyber threats. As online threats grow more sophisticated, proactive defenses like BaitNet play a crucial role in upholding security and integrity of online activities, making it a pivotal advancement in cybersecurity.