Answers are grounded in this paper and cite their sources.
Accurate and tamper-resistant attendance recording is a persistent operational challenge in educational institutions. Manual roll-calls, paper registers, and single-factor digital check-ins remain vulnerable to proxy attendance, transcription errors, and inadequate identity or location verification. This paper presents a design and implementation study of a web-based multi-factor attendance management system — AttendAI — that combines three complementary verification mechanisms: browser-side facial recognition, JSON Web Token (JWT) session validation, and Haversine geofencing. Facial descriptors of 128 dimensions are extracted using face-api.js and TensorFlow.js and matched against registered identities via Euclidean distance. JWTs authorize attendance operations within active teacher-initiated sessions. Haversine distance calculations enforce a configurable geographic boundary around the classroom. Attendance is accepted only when identity verification, token validity, and location eligibility are simultaneously satisfied. The system supports four user roles — administrator, class teacher, subject teacher, and student — with role-differentiated dashboards, email OTP registration, colour-coded attendance analytics, and CSV export. The backend is implemented in Python and Flask with SQLite managed through Flask-SQLAlchemy. As experimental results are not yet available, this paper presents a design and implementation study with quantitative evaluation proposed as future work.
No takes yet. Share an insight, caveat, or question.
Patil et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: