PHISHING WEBSITE DETECTION USING MACHINE LEARNING

Computer Engineering

With the increasing integration of the Internet into our social and professional lives, we are exposed to a growing number of serious security attacks. One of the primary concerns that needs immediate attention is the ability to detect various network threats, particularly those that involve attacks not seen before. Phishing, in particular, poses a significant risk as attackers aim to collect private information such as user identities, passwords, and financial transactions by creating websites that closely resemble legitimate ones. Government and financial organizations, being popular online destinations for users, have experienced a significant increase in phishing threats and attacks over the years.To combat these evolving phishing methods, it is crucial to employ anti-phishing techniques that can effectively detect and prevent such attacks. Machine learning presents a powerful tool for countering phishing assaults. Various machine learning approaches have been proposed and implemented to identify phishing websites. In this project, three supervised classification models are utilized: Support Vector Machine, Random Forest Classifier, and Logistic Regression.Among these models, the Random Forest Classifier has been found to provide the highest accuracy for the selected dataset, achieving an accuracy score of 97%. This classifier leverages the power of ensemble learning by combining multiple decision trees to make predictions. By using a combination of features extracted from URLs and employing the trained model, it becomes possible to accurately classify websites as phishing or non-phishing.By implementing these machine learning techniques, the project aims to enhance the detection and prevention of phishing attacks, thereby mitigating the risks associated with online security threats.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.