Dropout Prediction by Interpretable Machine Learning Model Towards Preventing Student Dropout

In the education industry, needs of online learning are significantly increasing. However, web-based courses demonstrate higher dropout rates than traditional courses. As a result, engaging students with data analysis is getting more crucial especially for distance learning. this study, we analyze on daily status in order to predict student’s schools. Specifically, trained prediction machine leaning model 1) Basic attributes students, 2) Progress materials, and 3) Slack conversation between teachers. The experimental results show that accuracy rate has reached 96.4%. was able 78% who actually dropped out school. We also looked into feature importance by SHAP value gain ML interpretability.