AI-Powered Student Success & Early Intervention Platform

Helping universities identify students who may be at risk of dropping out through machine learning and continuous student monitoring.

I built this web application for a client who needed a better way to monitor university students, identify potential dropout risks, and help lecturers intervene early.

Rather than building a system that only makes predictions, I developed a complete platform with separate dashboards for students and lecturers, powered by two integrated machine learning models.

What I built

1. Dual Machine Learning Models

Entry Prediction Model: Evaluates incoming students using their initial enrolment information to identify potential dropout risks.

Academic Progress Model: Uses students' academic results after their first two semesters to reassess their predicted risk as they progress.

2. Student Dashboard

Students can access their dashboard and submit weekly self-reports, giving lecturers additional insight into their academic progress and experiences.

3. Lecturer Dashboard

Lecturers can:

View and filter students by predicted risk level.

Access AI-generated predictions, including categories such as "Graduate" and "Needs Support".

Monitor weekly student self-reports.

Identify students who may require additional attention.

The Technology & Implementation

Python

Machine Learning

Data Processing

Model Training and Integration

Web Application Development

Database Integration

The Outcome

The client received an integrated platform that combines machine learning predictions with regular student feedback, helping lecturers monitor student progress and identify potential challenges earlier.

My contribution: Machine learning model development, integration of both models into the application, and development of the student and lecturer dashboards.

Interested in building an AI-powered application, integrating machine learning into your existing platform, or turning an idea into a functional product?

Let's work together.