PhishGuard: Machine Learning-Powered Phishing URL Detection

Phishing is a major threat to internet security, targeting human vulnerabilities instead of software vulnerabilities. It involves directing users to malicious websites where their sensitive information can be stolen. Many researchers have worked on detecting phishing URLs, but their models have limitations such as low accuracy and high false positives. To address these issues, we propose a machine-learning model to detect phishing URLs. To detect these malicious URLs, we use a dataset of over 500K entries collected from the Kaggle website. The dataset is used to train five supervised machine-learning techniques, including K-Nearest Neighbors (KNN), Logistic Regression (LR), Decision Tree (DT), Support Vector Machine (SVM), and Random Forest (RF). The aim is to improve the performance of the classifier by studying the features of phishing websites and selecting a better combination of them. To measure the performance, we considered three parameters: accuracy, precision, and recall. The LR technique yielded the best performance, demonstrating its efficacy in detecting phishing URLs.