Phishing Detection Techniques: Exploring Machine Learning and Deep Learning Models

The number of Internet users has increased significantly in recent years, driven by the growing popularity of online education, e-commerce, and other digital services. E-commerce, in particular, has seen significant growth due to increasing consumer demand for a convenient and secure online shopping experience. The COVID-19 pandemic has significantly sped up the uptake of e-commerce, changing consumer habits and propelling online transactions at an extraordinary rate. Nevertheless, this swift digital shift has heightened vulnerability to cyber threats, resulting in a significant rise in phishing attacks targeting the theft of confidential user data. In this article, we explore the use of Machine Learning (ML) and Deep Learning (DL) methods for detecting phishing, emphasizing conventional models, ensemble techniques, and hybrid systems. We analyze the key obstacles in this domain, such as data imbalance, significant computational expenses, and the challenges of real-time applications. Regarding research, this research emphasizes the potential of hybrid models and advanced methods to enhance the accuracy, efficiency, and scalability of Phishing systems. This result emphasizes the urgent need for a reliable, adaptive, and flexible detection system in order to express the growing risk of personal and organizational security in the digital development environment and to fight the increase in phishing attack.