evidoria

← Back to browse

Good practice Imported

IoT + AI Attendance Monitoring to Predict Dropout Risk — Wisdom School Musanze, Rwanda

Rwanda · Musanze · See the Rwanda profile

Evidence: Observational / pre–post Top 89% 27/100 · Ask Evidence Copilot about this practice

Researchers from Rwanda Polytechnic and the University of Rwanda built a fingerprint- and infrared-sensor attendance system at a Musanze secondary school and trained a decision-tree model on 2020–2021 records, reaching 91.4% accuracy linking low attendance to dropout.

91.4 %
Model accuracy linking attendance to dropout (2020-2021 records)
IoT + AI Attendance Monitoring to Predict Dropout Risk — Wisdom School Musanze, Rwanda

Details

Maturity
Pilot
Promoter
Rwanda Polytechnic (IPRC-Kigali) and University of Rwanda, with Wisdom School Musanze
Period
2020–2023
Keywords
secondary education, attendance monitoring, dropout prediction, IoT, machine learning

Context

Seven researchers from Rwanda Polytechnic (IPRC-Kigali) and the University of Rwanda built an IoT and AI-based attendance monitoring system at Wisdom School Musanze, a non-boarding secondary school in Musanze District, to address dropout linked to unmonitored absenteeism.

Activities

The system combines a fingerprint sensor for attendance registration and verification, a passive infrared (PIR) sensor to detect a student's physical presence, and a real-time clock to timestamp records, feeding a web application that gives parents and administrators real-time visibility and generates daily, monthly and annual reports. Using secondary data from the school's 2020-2021 academic-year records -- including sex, orphan status and school-fees payment history -- the team trained a decision-tree machine-learning classifier.

Results

The decision-tree model reached 91.4% accuracy in demonstrating that higher attendance rates are strongly associated with lower dropout in this school.

Conclusions

This is a single-school case study with an undisclosed sample size, so the 91.4% figure has not been shown to generalise beyond Wisdom School Musanze. The paper does not address the data-protection and consent implications of fingerprinting minors, nor whether the system has been adopted at any other school since publication.

Implementation
Indicative cost
Low (< €50k)
Time to results
Short (< 1 year)
Staffing & skills
Rwanda Polytechnic (IPRC-Kigali) research team, University of Rwanda research team, Wisdom School Musanze (implementing school)

Conditions for success

  • Multi-sensor design (fingerprint, PIR and real-time clock) for reliable attendance capture
  • Web application giving parents and administrators real-time visibility

Common failure modes

  • Undisclosed sample size limits generalisability beyond this single school
  • Data-protection and consent implications of fingerprinting minors not addressed
  • No evidence of adoption at other schools since publication

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Replication kit

Reusable artefacts from this practice — as published by their sources.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful