Detecting and Classifying Phishing Websites by Machine Learning

Aim: Phishing websites are a major threat to online security as they often lead to identity theft and financial loss. A machine learning based approach to detect and classify the phishing websites using a logistic regression model. Materials and Methods: The proposed models are effectively distinguished between legal sites and phishing sites by analyzing functions such as URL structure, domain details, and website operation. Training and testing have been executed in the dataset of the URL label, and the selection of the function has improved the predictive capacity of the model. Result: The logistics regression model has demonstrated impressive reviews, accuracy, precision, recall, highlighting its suitability for real-time phishing detection systems. It reaches the accuracy up to 96% comparative to THE existing model. This indicates the compatibility to the actual phishing system. Conclusion: This approach is simple and provides a practical way to support cyber security.