Skill Gaze: A Comprehensive Edge-Computing Framework for Privacy-Preserving AI Proctoring and Predictive Learning Analytics

DOI

No DOI

Published In

Volume 5, Issue 2 (2026)

Published Online

May-August 2026

Pages

1048-1053

Article Type

Original Article

Check for updates

No DOI available for cross-reference

The “SkillGaze: AI-Powered Proctoring and Learning Management System” is a web-based application developed to simplify and secure academic assessments and e-learning within educational institutions. The system provides a centralized platform for managing courses, student enrollments, assignment submissions, and AI-monitored examinations. Its objective is to reduce impersonation and unauthorized-device risks while reducing reliance on human proctors. SkillGaze provides role-based access for administrators, instructors, and students and follows an edge-computing approach in which AI inference runs in the client browser without transmitting raw webcam video. The implementation uses Node.js, Express, MongoDB, React.js, MediaPipe FaceMesh, and TensorFlow.js. A continuous temporal-buffer mechanism reduces false positives from transient events. The platform also provides a predictive-analytics pipeline using XGBoost to explore the relationship between LMS engagement and proctoring telemetry. A proof-of-concept experiment on 1,000 synthetic student records achieved 88.50% accuracy and 93.98% ROC-AUC; these results are treated as feasibility evidence rather than validation on real students. The resulting system combines centralized academic management, browser-based edge AI, privacy-preserving telemetry, and an extensible predictive-learning workflow