An In-Depth Methodology to Predict At-Risk Learners

Nowadays, the concept of education for all is gaining momentum thanks to widespread use e-learning systems around world. The consists in providing learning content via Internet physically dispersed learners. main challenge this regard high fail rate particularly among k-12 learners who are our case study. Therefore, we established an in-depth methodology based on machine models whose objectives early prediction at-risk and diagnosis problems. Going through was a great importance thus it starts by identifying most relevant indicators performance, engagement, regularity reactivity. Then, these indicators, extract select adequate features. For modeling part methodology, apply k-nearest neighbors (K-nn), Support Vector Machine (SVM), Random Forest Decision tree real data sample 1361 evaluation step comparing ability each model correctly identify class failure using both accuracy False Positive Rate (FPR) measures.