Title

AWS re:Invent 2023 - Preventing student debt using predictive data at scale (IDE106)

Summary

• Speakers: Sarah O'Hearn (VP of Product) and Ben May (Product Manager) from MoneyThink.
• Topic: Solutions for college affordability using AWS infrastructure and predictive modeling.
• Background: MoneyThink, founded in 2008 in Chicago, focuses on financial literacy and college access, aiming to address the issue of college students not graduating due to financial barriers.
• Problem: Non-standardized, confusing financial aid letters that often mislead students about the true cost of college and the amount of debt they may incur.
• Initial Solution: A tool called Decided, which standardizes financial aid information to help students and advisors compare costs across different colleges.
• Implementation: Utilizing AWS Textract to extract data from award letters and parsing it to make sense of various formats. This process saves time and resources for organizations and provides a large dataset for further analysis.
• Expansion Challenges: As MoneyThink expanded beyond California, they faced increased complexity in data classification and a significant number of students with no affordable college options.
• Predictive Affordability Solution: Incorporating Amazon Comprehend for better classification of financial aid data and Amazon SageMaker to predict college affordability for students based on academic and demographic data.
• Goal: To use the collected data to inform students earlier in the college application process and ensure they have affordable options among their choices.

Insights

• Data Standardization: MoneyThink's approach to standardizing financial aid letters is crucial for transparency and informed decision-making, highlighting the importance of clear communication in financial matters.
• Automation and Efficiency: The use of AWS Textract demonstrates how automation can significantly reduce manual labor and improve efficiency in data processing tasks.
• Predictive Analytics: The move towards predictive modeling with Amazon Comprehend and SageMaker indicates a trend in leveraging machine learning to provide personalized, actionable insights for students.
• Data-Driven Decision Making: MoneyThink's initiative underscores the power of data-driven decision making in education, particularly in helping students navigate financial barriers to college access.
• Collaboration and Community Input: The presenters' openness to community feedback and collaboration at AWS re:Invent suggests a commitment to continuous improvement and the value of shared expertise in tackling complex issues like student debt.