Student Dropout Risk Prediction

In this project, a data-driven approach using machine learning is being applied on students data to predict dropout risk of students across different academic years. The aim is to prevent students from dropping out of their study program by providing early intervention through counselling. The data used involved anonymized data (prior education and academic year specific data) of the students. The end results includes a dashboard consisting of year-specific interactive predictions, risk scores (how severe is the risk of dropping out) and which variables should be the focus of discussion for each student during counselling. The main highlights of the results will be available upon the completion of the project.

---