Machine Learning for Early Detection of Phishing URLs in Parked Domains

Phishing attacks have become a significant threat to internet users, with phishing websites targeting sensitive data through deceptive URLs. This project focuses on the early detection of phishing URLs in parked domains using machine learning techniques. The system applies four machine learning algorithms—Support Vector Machine (SVM), K-Nearest Neighbors (KNN), AdaBoost, and Decision Tree—to classify URLs into two categories: Legitimate Website (safe) and Phishing Website (dangerous). By leveraging these algorithms, the system analyzes the characteristics of the URLs and identifies patterns that distinguish phishing sites from legitimate ones. The models are trained and evaluated using Python, with libraries such as scikit-learn for SVM, KNN, AdaBoost, and Decision Tree implementations. This system aims to provide an efficient and scalable solution for early phishing detection, ensuring enhanced cybersecurity for users and organizations. The project highlights the importance of utilizing machine learning algorithms in safeguarding internet traffic and protecting users from malicious online activities. By combining various traditional machine learning techniques, this research demonstrates the potential of automated detection systems for identifying phishing threats in parked domains.

Keywords: Phishing Detection, Machine Learning, SVM, KNN, AdaBoost, Decision Tree, URL Classification, Cybersecurity, Parked Domains, URL Analysis, Phishing Websites, Legitimate Websites, Early Detection, Python.