The continued prevalence of phishing attacks highlights the need of research into the creation of reliable detection models for this pervasive online danger to e-business. Using a dataset procured from Kaggle, this study proposes a Convolutional Neural Network (CNN)-based method for identifying phishing scams. Using two Conv1D layers, our model successfully distinguishes between safe and harmful websites. Training results were quite encouraging, with a loss of just 0.077525 and an accuracy of 0.972125 throughout the process. These findings validate our CNN-based phishing attack detection model's sturdiness and adaptability. Our results not only provide a useful tool for spotting phishing attacks but also shed light on the possibilities of CNNs, and in particular Conv1D layers, in the realm of cybersecurity. This study is an important contribution to the ongoing effort to counter the rising danger of phishing attempts and improve the safety of e-business users worldwide.
Gupta, Brij B.; Gaurav, Akshat; and Chui, Kwok Tai, "Convolution neural network (CNN) based phishing attack detection model for e-business in enterprise information systems" (2023). ICEB 2023 Proceedings (Chiayi, Taiwan). 71.