Enhanced Phishing URL Detection Using Convolutional Neural Networks: A Deep Learning Approach

Phishing attacks exploit the trust of victims, posing significant threats to individuals and businesses worldwide. This study aims to address the challenge of detecting and warning against such attacks by applying deep learning models. Specifically, the study focuses on utilizing Convolutional Neural Networks (CNN) to classify URLs as malicious or benign. The study involves the collection and preprocessing of data from various sources, the design and implementation of deep learning models, and the evaluation of these models against traditional machine learning approaches. The proposed model achieved an accuracy of 93.19% on a newly aggregated and preprocessed dataset of 1,454,518 samples. This study contributes to enhancing cybersecurity measures and offers practical insights for further development in the field of malicious URL detection.