Eye-gaze-tracking-based proctoring systems report cheating-detection accuracies around 95–98% in their original studies¶

Claim · All claims
Evidence · 1 study · review r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? The review reports that eye gaze tracking systems for cheating detection achieved accuracies of about 95% (Li et al.; Dilini et al.) and more than 98% (Jagendra Singh et al.). → Malhotra 2026

Evidence¶

Malhotra 2026¶

Malhotra, M., & Chhabra, I. (2026). A comprehensive review of the changing landscape of academic dishonesty in automated proctoring in the era of artificial intelligence. Discover Education. https://doi.org/10.1007/s44217-026-01275-6

q2 · i? · review · r2

The review reports, citing Jagendra Singh et al., an eye-gaze classification system using Haar Cascade and CNN that "achieved an accuracy of more than 98%". The review also reports 95% accuracy for Gaze View Tracking (Li et al.) and for an OCSVM-based browser-plugin system (Dilini et al.).

"The eyeball localization is then classified using the Haar Cascade classifier, which was trained and tested using CNN. The model achieved an accuracy of more than 98%."

Discussion¶

Related Claims¶

• Head-pose-estimation-based proctoring systems report widely varying accuracies, from 75.6% for cheating detection to 100% for yaw-angle estimation — related
• Substantial proportions of students self-report cheating and misconduct during exams, with higher rates in high schools than colleges — related
• Facial recognition models for proctoring report accuracies up to 99.21%, with limitations in dim-light and blurring conditions — related
• Deep learning techniques such as CNNs and RNNs better detect cheating from visual cues than traditional techniques, but each carries stated trade-offs — related
• Time delay and head pose variations relative to the screen were significant predictors of cheating, yielding 75.6% average detection accuracy — related
• Model accuracy of vision-based proctoring systems can vary across demographics, lighting conditions, and hardware setups, potentially disadvantaging some students — a broader claim this one bears on
• Remote proctoring is perceived as an effective deterrent against cheating in nursing assessments, with 98–100% of graduate NP students agreeing webcam proctoring and lockdown browsers deterred cheating — related
• A webcam-only automatic cheating detector achieved recall of 78.6%, precision of 84.6%, and accuracy of 83.3% in a MOOP experiment — related