Understanding the Causes and Effects of Student Debt through Machine Learning

Background

Student debt has reached critical levels in the United States, with 43 million borrowers collectively owing $1.6 trillion in federal and private student loans as of 2019. Nearly 65% of the class of 2018 graduated with student debt, averaging $29,200 per borrower. Over the last few decades, the average balance has tripled, with many borrowers beginning their careers in financial distress. With student debt trailing only mortgages in total outstanding balances, this crisis demands innovative solutions.

Objective

The project aimed to leverage Artificial Intelligence to analyze and better understand the scope of the student debt crisis. Key goals included refining problem statements, analyzing data for actionable insights, and building ethical AI-based solutions that could recommend improvements to current policies or financial practices.

Approach

Over an eight-week collaboration between ShapingEDU and Omdena, approximately 50 AI engineers and academics worked on the project. The team:

1. Defined the Problem Statement: Clarified key challenges through discussions and research.
2. Collected and Processed Data: Analyzed publicly available student loan datasets.
3. Implemented AI Techniques: Conducted sentiment analysis of social media conversations and applied machine learning models for data exploration.
4. Ensured Ethical Standards: Maintained diversity within the team to prevent biases and promoted fast iterative cycles for solution development.

The project aligned with ShapingEDU’s “10 Actions to Shape the Future of Learning”, promoting data-driven approaches and innovation in tackling education challenges.

Results and Impact

The project delivered several tangible outcomes:

• Generated detailed insights into public sentiment and financial trends related to student debt.
• Built predictive models to identify risk factors and potential solutions for loan management.
• Enabled policymakers, educational institutions, and financial stakeholders to understand the crisis from a data-driven perspective.

These solutions have the potential to influence how student debt is managed, paving the way for better financial outcomes for borrowers and addressing systemic inequities in higher education financing.

Future Implications

The findings from this project could shape future research and policy initiatives by:

• Influencing Policy Decisions: Offering data-backed recommendations to improve loan repayment structures.
• Advancing AI in Education Financing: Demonstrating how AI can be applied to large-scale financial challenges.
• Driving Innovation: Encouraging institutions to adopt ethical and scalable AI-driven approaches to mitigate student debt and support long-term educational affordability.