Zest AI, a leader in software for AI-driven lending, announced the launch of Zest Race Predictor (ZRP). This open-source machine-learning algorithm estimates the race/ethnicity of an individual using only their full name and home address as inputs.
ZRP can be used to analyze racial equity and outcomes in critical spheres such as health care, financial services, criminal justice, or anywhere there’s a need to attribute the race or ethnicity of a population dataset when race/ethnicity data is missing. The financial services industry, for example, has struggled for years to achieve more equitable outcomes amid charges of discrimination in lending practices. A better yardstick can help reverse this legacy of bias.
Recommended AI News: Airtel Acquires Strategic Stake in Blockchain Technology Startup Aqilliz
ZRP improves upon the most widely used racial and ethnic proxying method, Bayesian Improved Surname Geocoding (BISG), developed by RAND Corporation in 2009. In multiple tests against BISG, ZRP was able to identify African-Americans correctly 25% more often, identify 35% fewer African-Americans as non-African American, and 60% fewer Whites as non-White.
“Zest AI began developing ZRP in 2020 to improve the accuracy of our clients’ fair lending analyses by using more data and better math,” says Mike de Vere, CEO of Zest AI. “We believe ZRP can significantly improve our understanding of the disparate impact and disparate treatment of protected-status borrowers.”
Recommended AI News: Contentstack Now Runs on Microsoft Azure, Becoming the Only Multi-Cloud SaaS Content Management System on the Market
“I’ve employed the ZRP output myself and found that it provided results consistent with our predictions, in the context of predicting the race of PPP borrower firm owners,” says Sabrina Howell, Assistant Professor of Finance at NYU Stern. “Getting race estimates right is key to facilitating fair lending practices in America, and by making their tool open-source and freely available, Zest’s application is an important step towards that goal.”
“We have known since our 2014 study that BISG leaves much room for improvement,” says Dr. Marsha J. Courchane, Vice President and Financial Economics Practice Leader, Charles River Associates. “We are thrilled Zest took the initiative to apply modern data science methods to develop a better race estimator, and we are looking forward to further validating this work.”
More accurate race prediction will help the entire lending ecosystem:
- Lenders will be better able to identify unfair outcomes to improve models.
- Regulators will have a better tool to enforce fair lending rules that drive equity in access to credit products that could help people of color earn better credit scores.
- Borrowers will benefit by knowing their race and ethnicity are more accurately reflected alongside their credit history.
Recommended AI News: Marketplaces Face Growing Competition as 4 in 5 Shoppers Report ‘Improved Ease’ of Shopping Elsewhere
[To share your insights with us, please write to sghosh@martechseries.com]
Comments are closed.