| Abstract: |
This review focuses on the application of AI tools to detect phishing sites, concentrating specifically in calculations such as Arbitrary Backwoods (LR), XGBoost, Simplistic Bayes (NB), Support Vector Machine(Bolstered vector machine). The study reminiscences the importance of comprehensive data preprocessing which consists, different precursory procedures like cleaning, extraction and standardization to improve the predictive accuracy and robustness of models. Among the tried methodologies, Angle Helping accomplished 97.6% precision which shows its strength in phishing identification and Guileless Bayes recorded the least exactness at 60.5%. Results show the fundamental role of selecting appropriate AI models and preprocessing methods to enhance phishing detection systems. This represents a significant advancement over traditional methods, especially in dealing with happy hour phishing attacks and managing restrictive data sets, providing an effective solution to enhance cyber security. |
