EDU Proctorx: AI Powered Remote Assessment Monitor

The rapid proliferation of online exams has led to a significant demand for reliable systems that uphold academic integrity, fairness, and security in remote assessment settings, and since monitoring students without physical supervision is challenging and conventional methods are often inadequate, this paper proposes the AI Powered Remote Assessment Monitor, a multimodal proctoring framework aimed at ensuring a secure and unbiased examination environment by integrating computer vision, audio processing, and behavioral modeling techniques to continuously monitor candidate activities during an examination, with identity checks based on facial recognition and liveness detection, supplemented by gaze tracking, head pose estimation, object detection, and sound classification methods to detect suspicious behaviors, thereby leveraging a multimodal analysis of signals to yield more accurate insights into candidate activities. First, unlike traditional rule-based or entirely manual methods, the framework emphasizes automated monitoring through modular AI elements and a weighted decision fusion approach that integrates data from various sources to reach balanced and contextually aware decisions instead of relying on isolated alerts; also, to ensure responsible utilization, the platform incorporates secure data handling protocols, including encrypted logging and privacy-conscious monitoring; the solution is built to accommodate both live and recorded examinations, and it can be implemented by educational institutions, certification bodies, and training organizations (ibid). Keywords: Artificial Intelligence, Computer Vision, Machine Learning, Facial Recognition, Behaviour Analysis, Ethical AI, Online Proctoring, Secure Assessments.