DEVELOPMENT OF A MULTIMODAL PHISHING DETECTION SYSTEM USING DEEP LEARNING
Abstract
Phishing attacks remain one of the most prevalent forms of cybercrime, exploiting human vulnerabilities to gain unauthorized access to sensitive information. As phishing techniques evolve from deceptive emails to fake websites and convincing visual replicas, traditional detection systems struggle to keep up. This study proposes a multimodal phishing detection system using machine learning and deep learning techniques to identify phishing attempts across emails, websites, and image-based screenshots. The system integrates natural language processing for email analysis, URL and website inspection for phishing detection, and image classification using convolutional neural networks. Performance evaluation using metrics such as accuracy, precision, recall, and F1-score shows strong results across different modalities. The multimodal approach improves detection accuracy and reduces false detections compared to single-mode systems.
Full-Text Access Notice
In accordance with the NERD Policy on promoting peer-reviewed publication, public access to the full text of a project, thesis or dissertation is restricted for three years, allowing the author and supervisors sufficient time to pursue peer-reviewed publication.
During this period, researchers with legitimate academic or research purposes may request authorisation directly from the author to enable NERD to release the indexed full texts of the work using the form below.