Projects / AI Innovation Challenge

Understanding the Causes and Effects of Student Debt through Machine Learning

Project completed!

Omdena Featured image

In this project, ShapingEDU partnered with Omdena to leverage AI applications to better understand — and potentially recommend solutions to — the student debt crisis.

The problem

Student debt has reached crisis proportions. In the United States, student loan borrowers owed a collective $1.6 trillion in federal and private student loan debt as of March 2019, according to the Board of Governors of the Federal Reserve System.

Sixty-five percent of the class of 2018 graduated with student debt, according to the data available from The Institute for College Access & Success, a nonprofit organization that works to improve higher education access and affordability. Among these graduates, the average student loan debt was $29,200.

Around 43 million people in the U.S. are in debt for their education. Each year, 70% of college graduates start off their lives in the red. And their average balance is around $30,000, up from $10,000 in the early 1990s.

As we enter a new decade, outstanding student debt trails only mortgages and is expected to top $2 trillion in the next couple of years. Recent stimulus bills addressing the coronavirus crisis have paused payments on loans for many borrowers. Could additional stimulus bills provide better solutions to the student debt crisis?

A graph from the New York Federal Reserve showing increases in student debt from 2004 to 2017.

A graph from the New York Federal Reserve showing increases in student debt from 2004 to 2017.

The approach

The ShapingEDU community worked together with selected Omdena collaborators in this eight-week project to refine the problem statement, collect the data, and build ethical AI solutions.

Approximately 50 AI engineers and academics applied processes such as sentiment analysis to analyze social media and machine learning to review publicly available student loan datasets. The diversity in the project team prevents biased solutions and enables breakthroughs by using fast iteration cycles and constant perspective sharing.

This project touches upon many of the 10 Actions to Shape the Future of Learning, the guiding principles of ShapingEDU’s community.

  • Innovative Artificial Intelligence Applications
  • Embed Data-Driven Approaches for Student Success
  • Bolster Intergenerational Leadership for Learning Futures
  • Build Constellations of Innovation

Your benefits

Working with world-class mentors and domain experts to acquire real-world experience

Making international friends in a fast-growing supportive community of collaborators

Boosting your technical skills, problem-solving capabilities, and collaboration skills

Building your personal brand and publishing your own articles on our website and blog

Receiving certificates of participation and references to accelerate your career


Good English

A good/very good grasp in computer science and/or mathematics

Student, (aspiring) data scientist, AI engineer, data engineer

Programming experience with C/C++, C#, Java, Python, Javascript or similar

Understanding of ML and Deep learning algorithms

This project has been hosted with our friends at

Application Form
media card
Optimizing & Deploying Climate and Credit Risk Scoring for African SMEs With AI
media card
Building AI-powered Early Warning System for Extreme Weather Conditions in Tanzania
media card
Optimizing the Accuracy & Explainability of Medical Insurance Claim (Fraud, Waste and Abuse) FWA Detection by Leveraging AI & Anomaly Detection

Become an Omdena Collaborator

media card
Visit the Omdena Collaborator Dashboard Learn More