Innovative machine learning model for malicious URL detection

Malicious URLs impose significant risk when access, including scams, attacks, and fraud either intentionally or unintentionally. This paper utilizes machine learning techniques to safeguard these threats. The research reviews current research surrounding machine learing and employs an ensemble approach to improve detection accuracy. Top models such as Decision Tree, Random Forest, AdaBoost, and XGBClassifier will be combined, in addition Voting Classifier with hard voting will also be used. The machine learning model's performance and its effectiveness are evaluated in the aspects of accuracy, precision, recall, and F1-score.