Methods for providing a recommendation

Patent metadata

Field	Value
Publication number	US-12211084-B2
Application number	US-202218146104-A
Country	US
Kind code	B2
Filing date	Dec 23, 2022
Priority date	Jun 18, 2020
Publication date	Jan 28, 2025
Grant date	Jan 28, 2025

Abstract

A computer-implemented method for providing a vehicle recommendation to a user may include: obtaining one or more vehicle images via a device associated with the user; identifying one or more user-selected images of the one or more vehicle images based on user interaction with the one or more vehicle images performed by the user via a user interface; identifying one or more first-level attributes from the one or more user-selected images; obtaining one or more vehicle identifications from the one or more user-selected images; determining one or more second-level attributes based on the one or more vehicle identifications; determining a value of each of the one or more first-level attributes and the one or more second-level attributes; determining the vehicle recommendation based on the value; and transmitting, to the device associated with the user, a notification indicating the vehicle recommendation.

First claim

What is claimed is:

1. A computer-implemented method for providing a vehicle recommendation to a user, the method comprising:
providing a first set of vehicle images from a plurality of vehicle images for display on a user interface of a device associated with a user;
receiving an indication of one or more first user-selected vehicle images from the first set via the user interface;
identifying a first plurality of vehicle attributes from the one or more first user-selected vehicle images;
determining a frequency that each vehicle attribute of the first plurality of vehicle attributes appears in the one or more first user-selected vehicle images from the first set;
determining a value of each vehicle attribute of the first plurality of vehicle attributes based on the frequency;
providing a second set of vehicle images from the plurality of vehicle images for display on the user interface, the providing causing one or more non-user-selected vehicle images from the first set that are displayed on the user interface to be replaced, on the user interface, with one or more new vehicle images included in the second set;
receiving an indication of one or more second user-selected vehicle images from the one or more new vehicle images via the user interface, wherein a second plurality of vehicle attributes identified from the one or more second user-selected vehicle images include at least a portion of the first plurality of vehicle attributes;
based on the selection of the one or more second user-selected vehicle images, updating the value of each vehicle attribute of the portion of the first plurality of vehicle attributes;
determining, using a trained neural network, a vehicle recommendation based, at least in part, on the updated value of each vehicle attribute of the portion of the first plurality of vehicle attributes, wherein the trained neural network is a convolutional neural network having a plurality of layers, including one or more convolutional layers that each apply a convolution operation to a respective input received by the convolutional layer for output to a next layer of the plurality of layers; and
transmitting, to the device associated with the user, a notification indicating the vehicle recommendation.

2. The computer-implemented method of claim 1 , wherein identifying the first plurality of vehicle attributes from the one or more first user-selected vehicle images comprises:
identifying one or more first-level vehicle attributes and one or more second-level vehicle attributes.

3. The computer-implemented method of claim 2 , wherein identifying the one or more first-level vehicle attributes comprises:
identifying the one or more first-level vehicle attributes from the one or more first user-selected vehicle images, wherein the one or more first-level vehicle attributes comprise at least one of a make, a model, a color, a door count, a seat count, a weight, a mileage, or a height of one or more vehicles indicated via the one or more first user-selected vehicle images.

4. The computer-implemented method of claim 2 , wherein identifying the one or more second-level vehicle attributes comprises:
obtaining one or more vehicle identifications from the one or more first user-selected vehicle images; and
determining the one or more second-level vehicle attributes based on the one or more vehicle identifications, wherein the one or more second-level vehicle attributes comprise at least one of an engine type, a manufacturing region, a manufacturing year, or a vehicle price of one or more vehicles indicated via the one or more first user-selected vehicle images.

5. The computer-implemented method of claim 1 , further comprising:
determining the one or more new vehicle images based on the value of each vehicle attribute of the first plurality of vehicle attributes and a predetermined threshold value.

6. The computer-implemented method of claim 1 , wherein determining the vehicle recommendation comprises:
determining the vehicle recommendation further based on one or more of: the first set of vehicle images, the one or more first user-selected vehicle images, the first plurality of vehicle attributes, the value for each vehicle attribute of the first plurality of vehicle attributes, the one or more new vehicle images, the one or more second user-selected vehicle images, the second plurality of vehicle attributes, or a value determined for each vehicle attribute of the second plurality of vehicle attributes.

7. The computer-implemented method of claim 1 , wherein the first plurality of vehicle attributes include the portion of the first plurality of vehicle attributes and one or more additional attributes, and determining the vehicle recommendation comprises:
determining the vehicle recommendation further based on the value of the one or more additional attributes.

8. The computer-implemented method of claim 1 , wherein the second plurality of vehicle attributes include a first portion of vehicle attributes corresponding to the portion of the first plurality of vehicle attributes, and a second portion of vehicle attributes including one or more different attributes from the first plurality of vehicle attributes, and determining the vehicle recommendation comprises:
determining a value of each vehicle attribute of the first portion of vehicle attributes and the second portion of vehicle attributes, wherein the value of each vehicle attribute of the portion of the first plurality of vehicle attributes is updated based on the value determined for the corresponding attribute in the first portion of vehicle attributes; and
determining the vehicle recommendation further based on the value determined for each vehicle attribute in the second portion of vehicle attributes.

9. The computer-implemented method of claim 1 , further comprising:
determining the first set of vehicle images from the plurality of vehicle images to provide for display based on user interactions with the first set.

10. The computer-implemented method of claim 1 , wherein at least a portion of the plurality of vehicle images are images captured by and obtained from the device associated with the user.

11. The computer-implemented method of claim 1 , wherein the one or more first user-selected vehicle images and the one or more non-user-selected vehicle images from the first set that are displayed on the user interface are caused to be replaced, on the user interface, with the one or more new vehicle images included in the second set.

12. The computer-implemented method of claim 1 , further comprising:
receiving user feedback data associated with the vehicle recommendation; and
updating the trained neural network based on the user feedback data.

13. A computer-implemented method for providing a vehicle recommendation to a user, the method comprising:
providing a first set of vehicle images from a plurality of vehicle images for display on a user interface of a device associated with the user;
receiving an indication of one or more first user-selected vehicle images from the first set via the user interface;
identifying a first plurality of vehicle attributes from the one or more first user-selected vehicle images;
determining a value of each vehicle attribute of the first plurality of vehicle attributes;
generating user preference data based on the value of each vehicle attribute of the first plurality of vehicle attributes;
providing a second set of vehicle images from the plurality of vehicle images for display on the user interface, the providing causing one or more non-user-selected vehicle images from the first set that are displayed on the user interface to

Assignees

• Capital One Services Llc

Inventors

• Tang Qiaochu
• Emamdie Michelle
• Guo Jiaxin
• Huang Jeremy
• Bumpas Derek

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N20/00

  Machine learning · CPC title

• G06Q30/0643 Primary

  graphically representing goods, e.g. 3D product representation · CPC title

• G06Q10/087

  Inventory or stock management, e.g. order filling, procurement or balancing against orders · CPC title