Phishing attacks are categorized as one of the greatest threats to cybersecurity. They are a form of misinfor-mation designed to make users provide important and personal information via fake websites or emails. This paper realizes the notion of a machine learning-based phishing detection tool aimed at classifying URLs as ”phishing,” ”suspicious,” or ”safe.” Utilizing a Random Forest classifier, the system examines URL-based characteristics, including URL length, special symbols, and the usage of HTTPS to distinguish between real URLs and fake ones (phishing URLs) with high accuracy. The model was trained and validated on a dataset of labeled URLs, achieving 95.2% classification accuracy, which is higher compared to other results. For the sake of usability, the detection tool is implemented as a web application for real-time classification with a user-friendly interface. The performance is evaluated using metrics such as accuracy, precision, recall, and F1-score to assess effectiveness. This paper helps to improve the level of online security by providing an automated approach that reduces dependence on human judgment and can efficiently detect phishing threats.
Kumar et al. (Sun,) studied this question.