Patent recognizes proprietary machine learning credit scoring methodology
Equifax Inc. (NYSE: EFX) has received a utility patent (Patent Number 10133980) for its proprietary and innovative NeuroDecision® Technology from the U.S. Patent & Trademark Office.
NeuroDecision is the first machine learning credit scoring methodology reviewed by regulators and credit scoring experts. NeuroDecision develops an explainable neural network model that improves performance and accuracy, which gives customers the ability to make more informed business decisions when assessing risk. The innovative technology was first introduced to Equifax customers in March and is in market today.
“Every day, we are using artificial intelligence and machine learning to help our customers make smarter, more informed decisions to grow their businesses and better serve consumers,” said Prasanna Dhore, Chief Data & Analytics Officer for Equifax. “The response from our customers to NeuroDecision in the short time it has been available is outstanding. We’re proud to be the only company that can deliver real value for customers through this unique, explainable credit scoring methodology, and we will continue to invest in proprietary research and development to evolve the technology and lead the industry.”
The NeuroDecision Technology Difference
In the past, the fundamental challenge in using neural networks for credit scoring has been the ability to explain the basis for an outcome, which is a necessary regulatory requirement. Through NeuroDecision, the models developed for credit scoring yield increased performance and flexibility, while also being explainable, predictive and stable over time.
This unique process is interpretable to customers and regulators, enables a deeper understanding of consumer behavior and is applicable and deployable wherever traditional scorecards are appropriate.
“As the technology has been developed and refined, we have met frequently with regulators to ensure compliance and to illustrate how NeuroDecision helps ensure fairness in scoring while also providing consumers with logical, actionable explanations for the key factors that impacted their credit scores,” said Kim Oliver, Chief Compliance Officer for Equifax. “The NeuroDecision solution shatters the ‘black box’ challenge for customers who wish to use neural networks to better predict the future behavior of consumers; reducing risk and empowering them to create more customized and appropriate offers, products and services.”
Equifax is currently using NeuroDecision in configurable solutions for customers using the Equifax Ignite® environment. Equifax Ignite is a portfolio of premier data and advanced analytics solutions that supports the full analytical life cycle – from data access and transparency to visualization and deployment – using a single, connected suite of advanced analytical processes, technology and tools. NeuroDecision is also currently deployed in commercial and consumer risk models. The company’s data scientists have deployed NeuroDecision around the globe in auto, communications, utilities and mortgage risk models.
Consistent Innovation in Data & Analytics
Scientists in the global Equifax Data & Analytics Lab are leading the development of new innovations in predictive analytics, machine learning and explainable artificial intelligence.
The company recently launched Equifax Ignite™ + InterConnect®, an enhanced solution that creates an integrated experience helping Equifax customers make smarter, faster decisions with unique data. Other recent innovations include Peak Attributes and Advanced Model Engine. From machine learning to natural language processing, Equifax is exploring how to harness these techniques in order to improve lending decisions, bring new consumers into credit markets, and help lenders identify fraudulent activity.
Source: Equifax Press Release