Intelligent dealership recommendation engine

Patent metadata

Field	Value
Publication number	US-12346919-B2
Application number	US-202418437882-A
Country	US
Kind code	B2
Filing date	Feb 9, 2024
Priority date	May 11, 2020
Publication date	Jul 1, 2025
Grant date	Jul 1, 2025

Abstract

Aspects described herein may provide an interface and/or search functionality for a dealership to determine vehicles a customer is most likely to purchase. A recommender system may generate vehicle recommendations for a dealership to sell to a customer based on customer information, vehicle information, and dealership information. Machine learning may be used to generate the recommendations. The recommendations may be based on the vehicle preferences of a customer.

First claim

What is claimed is:

1. A computer-implemented method comprising:
receiving, from a user device, customer information corresponding to a customer;
determining a set of vehicles, which is in an inventory of a vehicle dealership, that the customer is prequalified to purchase based on the customer information and a pricing model for financing the set of vehicles;
determining, using a machine learning model trained to approximate the pricing model based on historical output of the pricing model, a score for each vehicle within the set of vehicles based on the customer information and vehicle information for the set of vehicles, wherein the machine learning model is trained to approximate the pricing model with a threshold accuracy of less than 100% and prevented from training beyond the threshold accuracy, and wherein the score is determined based on an objective maximization function that maximizes both:
a first probability that the customer will purchase the vehicle, and
a second probability that a profitability of a sale price of the vehicle is agreeable to the vehicle dealership; and

generating a list of vehicles based on the set of vehicles, sorted by the score for each vehicle.

2. The method of claim 1 , wherein the first probability is associated with one or more customer preferences that the vehicle matches.

3. The method of claim 1 , wherein the pricing model indicates a loan structure that is acceptable to the vehicle dealership.

4. The method of claim 1 , further comprising:
calculating, based on customer purchase history, an effectiveness of the machine learning model; and
based on the effectiveness calculation, switching to an alternative machine learning model.

5. The method of claim 1 , further comprising:
calculating, based on customer purchase history, an effectiveness of the machine learning model; and
based on the effectiveness calculation, adjusting one or more weights of the machine learning model associated with one or more customer preferences.

6. The method of claim 1 , wherein the machine learning model is trained based on historical vehicle information, historical customer information, and historical dealership information generated by the pricing model.

7. The method of claim 1 , wherein the threshold accuracy used to train the machine learning model is less than 95% accuracy.

8. The method of claim 1 , further comprising:
receiving information indicating a vehicle, from the list of vehicles, that is associated with the customer;
receiving product information indicating front end purchases for the vehicle; and
generating, based on the product information, a list of product combinations for the vehicle, wherein the list of product combinations is sorted by a likelihood that the customer will purchase each product combination within the list of product combinations.

9. The method of claim 8 , wherein the product information further indicates back end purchases for the vehicle.

10. The method of claim 1 , wherein a vehicle's position within the list of vehicles sorted by the score value is adjusted based on a random variable.

11. A computing system comprising:
one or more processors; and
memory storing instructions that, when executed by the one or more processors, cause the computing system to:
receive, from a user device, customer information corresponding to a customer;
determine a set of vehicles, which is in an inventory of a vehicle dealership, that the customer is prequalified to purchase based on the customer information and a pricing model for financing the set of vehicles;
determine, using a machine learning model trained to approximate the pricing model based on historical output of the pricing model, a score for each vehicle within the set of vehicles based on the customer information and vehicle information for the set of vehicles, wherein the machine learning model is trained to approximate the pricing model with a threshold accuracy of less than 100% and prevented from training beyond the threshold accuracy, and wherein the score is determined based on an objective maximization function that maximizes both:
a first probability that the customer will purchase the vehicle, and
a second probability that a profitability of a sale price of the vehicle is agreeable to the vehicle dealership; and

generate a list of vehicles based on the set of vehicles, sorted by the score for each vehicle.

12. The computing system of claim 11 , wherein the first probability is associated with one or more customer preferences that the vehicle matches.

13. The computing system of claim 11 , wherein the pricing model indicates a loan structure that is acceptable to the vehicle dealership.

14. The computing system of claim 11 , wherein the instructions further cause the computing system to:
calculate, based on customer purchase history, an effectiveness of the machine learning model; and
based on the effectiveness calculation, switch to an alternative machine learning model.

15. The computing system of claim 11 , wherein the instructions further cause the computing system to:
calculate, based on customer purchase history, an effectiveness of the machine learning model; and
based on the effectiveness calculation, adjust one or more weights of the machine learning model associated with one or more customer preferences.

16. The computing system of claim 11 , wherein the machine learning model is trained based on historical vehicle information, historical customer information, and historical dealership information generated by the pricing model.

17. The computing system of claim 11 , wherein the threshold accuracy used to train the machine learning model is less than 95% accuracy.

18. One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to perform steps comprising:
receiving, from a user device, customer information corresponding to a customer;
determining a set of vehicles, which is in an inventory of a vehicle dealership, that the customer is prequalified to purchase based on the customer information and a pricing model for financing the set of vehicles;
determining, using a machine learning model trained to approximate the pricing model based on historical output of the pricing model, a score for each vehicle within the set of vehicles based on the customer information and vehicle information for the set of vehicles, wherein the machine learning model is trained to approximate the pricing model with a threshold accuracy of less than 100% and prevented from training beyond the threshold accuracy, and wherein the score is determined based on an objective maximization function that maximizes both:
a first probability that the customer will purchase the vehicle, and
a second probability that a profitability of a sale price of the vehicle is agreeable to the vehicle dealership; and

generating a list of vehicles based on the set of vehicles, sorted by the score for each vehicle.

19. The computer readable media of claim 18 , wherein the machine learning model is trained based on historical vehicle information, historical customer information, and historical dealership information generated by the pricing model.

20. The computer readable media of claim 18 , wherein the threshold accuracy used to train the machine learning model is less than 95% accuracy.

Assignees

• Capital One Services Llc

Inventors

• Price Micah
• Tang Qiaochu
• Dagley Geoffrey
• Ghamsari Avid

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N5/025

  Extracting rules from data · CPC title

• G06Q30/0202 Primary

  Market predictions or forecasting for commercial activities · CPC title