Inspiration
The inspiration for the Equitable Loan Predictor came from the realization that there is a lot of bias in financial systems including loan approval processes. These biases have mainly affected underrepresented communities and have prevented fair access to financial opportunities. Due to this, I wanted to create a tool that can help lenders make more equitable decisions and reduce human bias.
What it does
This Predictor helps financial institutions make fairer decisions about loans by mitigating biases that greatly affect underrepresented communities.
How we built it
This project was built using Python, using libraries like Panda for data preprocessing, scikit-learn for training the model, and Fairlearn for bias detection.
Challenges we ran into
A main challenge was handling the data as it was inconsistent and impacted how the model ran.
Accomplishments that we're proud of
I am proud of building a model that can predict loan approvals and detect and mitigate biases.
What we learned
I learned the importance of AI. I think it is highly underestimated and had a stigma around it, but when AI is used for things like fairer loan predictions, I think it would be important to implement AI into more things if it can help underrepresented communities.
What's next for Equitable Loan Predictor
I plan to integrate more fairness metrics while also exploring more features that can further reduce bias.
Log in or sign up for Devpost to join the conversation.