Personalized property recommender system using artificial reality and machine learning

ChangeBridge: Patent Grants - Business Methods (G06Q)

Published March 31st, 2026

Detected April 1st, 2026

Summary

The USPTO granted patent US12591942B1 to United Services Automobile Association (USAA) for a personalized property recommender system using artificial reality and machine learning. The system correlates user preferences with property data to generate ranked recommendations, mitigating information overload for property buyers. The patent contains 20 claims and names seven inventors.

View original document View source feed page

What changed

USPTO granted patent US12591942B1 to USAA on March 31, 2026, covering a personalized property recommender system that integrates artificial reality and machine learning to generate and rank property recommendations based on user preferences. The patent application was filed on January 31, 2024 (Application No. 18428678) and includes CPC classifications related to real estate services, machine learning, and AR/VR technologies.

Compliance teams at financial institutions, real estate technology firms, and software developers should review the patent claims to assess potential infringement risks if developing similar property recommendation or AR/ML-based property search systems. Freedom-to-operate analyses may be warranted for companies in the property technology space.

Source document (simplified)

← USPTO Patent Grants

Personalized property recommender system using artificial reality and machine learning

Grant US12591942B1 Kind: B1 Mar 31, 2026

Assignee

United Services Automobile Association (USAA)

Inventors

Gideon Bowie Luck, Donnette L. Moncrief Brown, Matthew Robert Byrd, Rincy Raju George, Bradly Jay Billman, Huihui Wu, Surender Kumar

Abstract

Implementations generate personalized property recommendations based on experiential data obtained via artificial reality systems and machine learning predictions that correlate user preferences to these properties. A user can be one or more persons interested in purchasing a property, such as a new home. A property buying workflow often includes defining a set of preferences for a desired property and comparing properties for sale with these defined preferences. However, in a conventional buying workflow, defining preferences and doing the comparison is left up to the user and/or the user's real estate agent. Users can often feel overwhelmed by the information, in particular users that do not have previous buying experience. Implementations can integrate machine learning and artificial reality systems into the property buying experience to mitigate the information overload problem and provide meaningful recommendations to the user, such as a ranking of properties that comply with user's preferences.

CPC Classifications

G06Q 50/16 G06Q 50/163 G06Q 50/165 G06Q 50/14 G06Q 10/02 G06Q 10/20 G06F 40/40 G06T 19/006 G06V 20/20 Y10S 707/946 A61B 2090/365 G01C 21/3484 G01C 21/343 G01C 21/2476