MERN Driven: AI proctor Examination System (Empowering Exam Security).

This research presents an innovative approach to online examination security through the implementation of an AI-driven proctoring system. The study explores the integration of cutting-edge technologies such as TensorFlow models, object detection, face recognition, and activity monitoring to ensure the integrity of online assessments. The proposed system leverages machine learning algorithms, specifically COCO-SSD, to identify suspicious activities, such as multiple face detection, object manipulation, tab switching, and unauthorized screenshot capture, in real-time. Additionally, it offers a seamless user experience through a robust frontend interface developed using React.js, complemented by a secure and scalable backend powered by Node.js and MongoDB. The core of the system revolves around the AI Proctored System, which facilitates a transparent examination environment by continuously monitoring the candidate's behavior during the test. The system ensures that students are authenticated through secure JWT-based login mechanisms and undergo a comprehensive verification process before the commencement of the test. Furthermore, the system accommodates multiple test formats, such as Multiple-Choice Questions (MCQs) and coding challenges, with support for various programming languages. By requiring full-screen mode and webcam and microphone access, the system minimizes the chances of cheating, while real-time activity logging enables the teacher to monitor student behavior and activity throughout the examination. The comprehensive monitoring and analysis provided by the AI-driven proctoring system not only enhances the reliability of online testing but also lays the groundwork for more secure and efficient digital assessment environments in the future. This research contributes significantly to the field of educational technology by offering a scalable, AI-powered solution to the prevalent issues of academic dishonesty and online exam integrity.