Enhancing organizational cybersecurity: A framework for mitigating email phishing attacks

Ayankoya, Folasade.Yetunde and Olakunle, Olasubomi Priscilla and Umezurike, Daniel Uchechukwu and Ekpetere, Chioma Favour (2025) Enhancing organizational cybersecurity: A framework for mitigating email phishing attacks. Global Journal of Engineering and Technology Advances, 23 (3). 038-047. ISSN 2582-5003

Phishing attacks via email continue to pose significant cybersecurity threats by exploiting human vulnerabilities and deceiving users into disclosing sensitive information. Despite measures put in place by organizations to avert this attack, phishing still proved a hard nut to crack. Hence, this study presents the design and implementation of a deep learning-based phishing detection system using Convolutional Neural Networks (CNNs). Unlike traditional rule-based or machine learning approaches, the proposed model leverages CNN's automatic feature extraction capability to analyze email content, including subject lines, body text, and embedded links. The system was evaluated using a real-world dataset, achieving high accuracy, precision, recall, and F1-score, thereby demonstrating its effectiveness in detecting both conventional and sophisticated phishing attempts. By integrating advanced regularization techniques and a user-facing web application, the model ensures adaptability and practical deployment. The results affirm CNN's potential to enhance cybersecurity defenses and reduce exposure to phishing risks in both individual and enterprise environments.