Patent Application: Techniques for Improving Automated Prediction Accuracy

ChangeBridge: Patent Apps - Business Methods (G06Q)

Detected March 27th, 2026

Summary

The USPTO has published a patent application (US20260087548A1) detailing techniques for improving automated prediction accuracy, particularly for credit risk assessment. The application, assigned to Lendingclub Bank, National Association, focuses on clustering underperforming and performant entities to enhance credit models and identify potential future risks.

View original document View source feed page

What changed

This document is a published patent application (US20260087548A1) from the USPTO, detailing "Techniques for Improving the Accuracy of Automated Predictions." The application, assigned to Lendingclub Bank, National Association, describes methods for forming clusters of individual prediction targets (IPTs), such as borrowers, by grouping "core" underperforming entities (e.g., defaulting borrowers with similar features) and "boundary" performant entities (e.g., non-defaulters with similar credit qualifications). These clusters are intended to improve credit model accuracy, identify future risks, and generate visualizations.

As this is a patent application, it does not impose direct regulatory obligations or compliance deadlines on entities. However, the techniques described could inform the development of future financial technology and risk management systems. Companies involved in credit scoring, lending, and predictive analytics may find the methodologies relevant for enhancing their operational efficiency and risk assessment capabilities. Compliance officers in the financial sector should be aware of such technological advancements as they may influence industry best practices and future regulatory considerations.

Source document (simplified)

← USPTO Patent Applications

TECHNIQUES FOR IMPROVING THE ACCURACY OF AUTOMATED PREDICTIONS

Application US20260087548A1 Kind: A1 Mar 26, 2026

Assignee

Lendingclub Bank, National Association

Inventors

Jianju Liu, Dhwani Umeshbhai Bosamiya, Jianglan Han, Phani Pradeep Benarji Kommana, Shi Tang

Abstract

Techniques are provided for forming clusters of individual prediction targets (IPTs). An initial prediction target is a target for which an automated prediction has been generated. IPTs may be, for example, borrowers to which a lending entity has extended loans based on predictions generated by a credit policy. Each cluster includes (a) a “core” of underperforming entities (UEs), and (b) a set of boundary performant entities (PEs). The UEs that belong to the UE core of a cluster are “similarly situated” relative to the values of their features. For example, in the context where the IPTs are borrowers, the UEs at the core of a cluster may correspond to defaulting borrowers that had similar bureau data, lending entity data, and borrower data. The boundary performant entities of the cluster may be borrowers that have not defaulted, but had similar credit qualifications as the UEs of the cluster. Having formed these clusters, the clusters may be used in a variety of ways, including but not limited to improving the accuracy of the credit model, identifying potentially problematic future borrowers, generating visualizations that illustrate the relative importance of clusters of defaulting borrowers, etc.