Abstract

Phishing websites have become a significant cybersecurity threat, hosting malware and exploiting users by mimicking popular sites. Victims suffer financial loss, compromised privacy, and damaged reputation. Urgent solutions are needed to mitigate these threats promptly. Our goal is to develop a Chrome extension that utilizes machine learning and deep learning techniques to identify phishing websites and alert users when they visit such sites. When a user opens a website in their browser, the extension operates in the background. If the website is recognized as phishing, a pop-up box with an 'OK' button informs the user. If the website is not flagged as phishing, no action is taken. We employed machine learning algorithms (such as Random Forest) and a Single Layer Perceptron for training the model. The browser extension is written in JavaScript, and its JavaScript file extracts URL features like browser popups, use of frames, and shortening services. The algorithm then determines whether the website is legitimate or not.