A Risk-Aware Privacy-Preserving Online Examination System Using Behavioural Anomaly Detection Using Blockchain Technology

Abstract

The fast implementation of online examination systems has yielded a dire requirement of proctoring solutions that safeguard and maintain privacy. Standard webcam-based surveillance systems improve detection rates and add serious privacy problems, excessive computing needs and reliance on a well-functioning network presence. In order to overcome these shortcomings, this paper suggests a risk-conscious and privacy-conscious online examination framework that gets rid of constant video surveillance by using the behavioural analysis of browsers, identifying anomalies, and recording the logs using blockchain technology. One of the systems identifies suspicious behaviour patterns like tab switching, inactivity, and patterns of abnormal interactions by a weighted risk-scoring model. Moreover, a lightweight anomaly detection system is incorporated to detect the violations of the regularity of behaviours. Critical events and examination results are safely stored in blockchain logging based on SHA-256 in order to guarantee data integrity and non-repudiation. Experimental outcomes show that the proposed system is highly accurate in detecting and has low latency without compromising user privacy. The structure offers an efficient, scalable and technically sound system to the current web-based test setting.