Machine Learning Trust Assessment for Retail Customer Accounts

ChangeBridge: Patent Grants - Business Methods (G06Q)

Published March 31st, 2026

Detected March 31st, 2026

Summary

The USPTO granted Patent US12591887B2 to Walmart Apollo, LLC covering systems and methods for assessing customer account trustworthiness in retail and e-commerce transactions using machine learning. The patent describes computing trust scores based on historical transaction data and transmitting results in response to trust assessment requests. The patent contains 20 claims and names five inventors.

View original document View source feed page

What changed

The USPTO granted US Patent 12591887B2 to Walmart Apollo, LLC on March 31, 2026. The patent covers machine learning methods for generating trust scores for customer accounts in retail environments, including receiving trust assessment requests, computing trust-related features from historical transaction data, and transmitting trust score data. This is a routine patent grant notice with no associated regulatory requirements or compliance obligations.

Compliance officers and legal professionals should note this patent grant for intellectual property tracking purposes. Companies developing similar trust assessment or fraud detection systems in retail contexts should consider potential licensing implications or design-around needs. No immediate action is required; this is informational regarding the intellectual property landscape in machine learning applications for retail transactions.

Source document (simplified)

← USPTO Patent Grants

Systems and methods for assessing trustworthiness of customer accounts in omnichannel retail transactions using machine learning

Grant US12591887B2 Kind: B2 Mar 31, 2026

Assignee

Walmart Apollo, LLC

Inventors

Jay Shital Shah, Sarthak Jagetiya, Gnanapriya Venkatasubramaniam, Mohammad Javad Mahzoon, Ahmad Khodayari-Rostamabad

Abstract

Systems and methods for assessing trustworthiness of customer accounts in retail and e-commerce systems using machine learning are disclosed. In some embodiments, a disclosed method includes: receiving, from a computing device, a trust assessment request regarding a customer account; generating at least one trust-related feature or quantitative attribute associated with the customer based on historical transaction data and relevant account information of the customer; computing, using at least one machine learning model, trust score data of the customer based on the at least one trust-related feature; and transmitting, in response to the trust assessment request, the trust score data of the customer to the computing device.