Student performance early-warning dashboard

The brief

A college wants to support struggling students before the final exams, not after. Using student records and online-learning activity, predict who is at risk and show teachers what is driving the risk. Good looks like: risk flags raised early in the term, fair across groups of students, and a dashboard teachers find useful.

What you will build

1. Explore what relates to final grades
2. Build early-term features (attendance, first marks, activity)
3. Train a risk model and check it for fairness
4. Explain the risk for each student
5. Build a teacher dashboard and write the report

Free, open data to build it with

• Student Performance ↗ UCI Machine Learning Repository
• Open University Learning Analytics Dataset (OULAD) ↗ The Open University

Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.