In the digital age, the proliferation of malicious phishing URLs poses a significant threat to online security. While conventional machine learning algorithms have been employed to combat this menace, our research pioneers the use of ensemble methods, including XGBoost and Random Forest, for phishing URL detection. Our methodology involves collection of the data, preprocessing it then feature extraction followed by model training, evaluation and comparison. Notably, our results reveal the superior accuracy of ensemble methods in distinguishing phishing URLs from legitimate ones. These findings underscore the potential of ensemble methods as a game-changing asset in the battle against cyber threats, promising enhanced online security and the protection of sensitive user information.

Keywords : Social Engineering, Phishing URLs, Cyber Security, Machine Learning.