Date of Publication :5th April 2025

Abstract: Phishing attacks continue to be a concern, in the realm of security as they target sensitive information by using deceptive websites. The usual methods of detecting phishing websites heavily rely on heuristics and manual inspection. These approaches have limitations when it comes to adapting to phishing tactics. In our study we put forward an approach for identifying phishing websites using learning techniques. We utilize a neural network architecture that analyzes both the content and structure of websites. This architecture allows us to automatically detect phishing websites by learning patterns that indicate behavior. Additionally we explore the application of networks to capture time based changes in website content. This further enhances our models ability to identify phishing attacks that evolve over time. To assess the effectiveness of our approach we conduct experiments on datasets and compare the performance of our deep learning model with traditional machine learning methods. Our results clearly demonstrate that our proposed deep learning approach surpasses existing techniques in terms of accuracy and reliability when it comes to detecting phishing websites across scenarios. The findings from this study carry implications for enhancing security as they offer a more efficient and scalable solution, for detecting phishing websites. By utilizing deep learning methods we can improve the effectiveness of security systems, in detecting and minimizing the threats presented by phishing attacks. This in turn ensures the protection of users sensitive data, within our growing environment.

Reference :