Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

Low-Connectivity Learning Analytics: Lightweight Predictive Models for School Dropout Prevention in Mozambique.

Domain:

education

Record type:

paper
Creator:
Mpf
Publisher:
Zenodo
Host:avatar
Abstract School dropout remains one of the greatest challenges to human development in Sub-Saharan Africa, particularly in Mozambique. Traditional Learning Analytics approaches require high connectivity and real-time interaction, which are infeasible in rural and low-resource environments. This article presents the design, validation, and evaluation of a Machine Learning (ML) model optimized for predicting dropout risk using exclusively asynchronous, low-volume data. By leveraging only three input features—frequency of task submission (FET), periodic test performance (DAP), and absenteeism tendency (TF)—we achieved high predictive accuracy with minimal computational cost. The Random Forest Classifier achieved an AUC of 0.95 and a critical Recall (Sensitivity) of 92.1% ± 1.8%, significantly outperforming the Logistic Regression baseline (AUC 0.88). The model’s runtime was under 150 milliseconds, and its memory consumption was below 50MB, demonstrating its viability for offline-first and SMS-based architectures. Partial Dependence Analysis ensured interpretability, revealing Performance (DAP) and Task Submission (FET) as the strongest predictors. This study contributes to Educational Data Mining (EDM) by demonstrating a lightweight, equitable AI solution that is viable under extreme resource constraints, directly addressing digital inequities and providing actionable recommendations for the Mozambican Ministry of Education (MINEDH).

Visit

doi.orgzenodo.org

Licenses

Creative Commons Attribution 4.0 Internationalhttps://creativecommons.org/licenses/by/4.0/legalcode

Similar

Contribution of deep learning to predictive models for early dropout detection: the case of high school students in the Rahmna region.Lightweight Deep Learning Models for Brain Tumor Classificationrafaelbmpfumo-crypto/Low-Bandwidth-Dropout-Prediction-MozambiquePredictive Analytics Models for Health Management in Kenyan Livestock Herdsmen: A Comparative StudyLarge-scale School Mapping using Weakly Supervised Deep Learning for Universal School ConnectivityHassan-2468/Predictive-Models-for-Student-Dropout-Risk-A-Review-and-Empirical-Evaluation-Using-Real-World-Data

Contribution of deep learning to predictive models for early dropout detection: the case of high school students in the Rahmna region.

Despite improved investment in the education sector, a paradox persists which can be summed

Lightweight Deep Learning Models for Brain Tumor Classification

Differentiating brain tumours by MRI using computer algorithms remains a huge challenge in clinical

rafaelbmpfumo-crypto/Low-Bandwidth-Dropout-Prediction-Mozambique

Código e dados sintéticos para o modelo ML-Evasion-LB, focado em Equidade Digital em contextos de ba

Predictive Analytics Models for Health Management in Kenyan Livestock Herdsmen: A Comparative Study

Predictive analytics models are increasingly being used to improve health management in liv

Large-scale School Mapping using Weakly Supervised Deep Learning for Universal School Connectivity

Improving global school connectivity is critical for ensuring inclusive and equitable quality educat

Hassan-2468/Predictive-Models-for-Student-Dropout-Risk-A-Review-and-Empirical-Evaluation-Using-Real-World-Data

This project evaluates machine learning models (XGBoost, Random Forest, and LR) to predict student d