Abstract –The growing prevalence of phishing assaults, especially in online banking and e-commerce, calls for the creation of reliable detection systems. A thorough analysis of the use of machine learning methods for phishing website identification is presented in this research. By leveraging supervised classification approaches, we analyze various algorithms, including ensemble methods and deep learning models, to enhance detection accuracy. Our research highlights the importance of feature extraction from URLs and webpage content, which significantly contributes to the performance of predictive models[1]. We also discuss the challenges posed by sophisticated phishing tactics that exploit human vulnerabilities and technical weaknesses. Through extensive experimentation with labelled datasets, our findings demonstrate that machine learning can effectively identify phishing attempts, thereby providing a critical layer of security in the ever-evolving landscape of cyber threats. By providing insights into practical methods for phishing detection utilizing cutting-edge machine learning techniques, this work seeks to support continuing efforts in cyber security.
