Abstract

Phishing attacks remain an evolving challenge to web user’s world all over. These attacks which emanates from phishing URLs are many and very problematic to internet users globally. The curiosity by well-meaning security professionals has led to further research, and thus propels the development of newer models to solve the lingering challenge in the Cyber domain. The updated classification models are been developed in other to curb the gaps of experiencing phishing attacks. This study aims at strategically using text to image conversion for phishing attack classification by using the URLs key features which are address and extension. CNN was also used as deep learning algorithm in other to increase models detection accuracy, reduce time complexity and also address misclassification issues as well as poor prediction accuracy. This was done in other to increase the resilience of the suggested model as well as enhancing classification prediction. The performance evaluation metrics employed for the proposed model are accuracy, precision, recall, F1 score, confusion matrix and AUC-ROC. This study outlined a novel method capable of identifying phishing attacks using features primarily obtained from the phishing and real URL addresses. A temporal tokenizer was also generated and used for URL text processing which scanned, recognized characters, symbols and redundant tokens. This made it easier to separate specific features from the URL This work is licensed under the Creative Commons Attribution International License (CC BY 4.0). http://creativecommons.org/licenses/by/4.0/ Vol. 6. Issue.1. 2026 (January) Journal of Artificial Intelligence and Modern Technology Page 21 address and return as a list while also identifying directories, keyword arguments, and extensions