Item recommendations weighted by user-valued features

Patent metadata

Field	Value
Publication number	US-2025259224-A1
Application number	US-202519195078-A
Country	US
Kind code	A1
Filing date	Apr 30, 2025
Priority date	Mar 13, 2019
Publication date	Aug 14, 2025
Grant date	—

Abstract

Computer-implemented methods, computer systems, and computer readable media for determining, sorting, or determining and sorting a set of items. The method includes receiving, via a user device, parameter values from a user and receiving parameter weights. The method further includes determining a sort score for each vehicle of a set of vehicles based on the received parameter values and parameter weights, and sorting the set of vehicles based on the sort score of each vehicle of the set of vehicles.

First claim

What is claimed is:

1 . A computer-implemented method for determining purchase recommendations, comprising:
accessing historical data indicative of one or more of activity or behavior of a user; determining a respective value for each of one or more parameters of an item and one or more parameter weights for the one or more parameters by applying a machine-learning model to the historical data; generating one or more parameter similarity values for each item of a set of items based on the respective parameter values; determining a sort score for each item of the set of items based on the one or more parameter weights and the one or more generated parameter similarity values; sorting the set of items based on the sort score of each item of the set of items; and causing a user interface of a user device associated with the user to output at least a portion of the sorted set of items.

2 . The computer-implemented method of claim 1 , wherein the determining of the respective values and parameter weights for the one or more parameters includes applying the machine-learning model to the historical data so as to learn one or more user preferences over time and using the one or more user preferences to determine the respective values and parameter weights.

3 . The computer-implemented method of claim 1 , wherein the machine-learning model is configured to apply one or more filters to a domain of items based on the historical data.

4 . The computer-implemented method of claim 1 , wherein each of the respective parameter values is indicative of a respective aspect of the item.

5 . The computer-implemented method of claim 1 , wherein the one or more parameter weights are indicative of a ranked order of importance of the one or more parameters to the user.

6 . The computer-implemented method of claim 1 , wherein the one or more parameter weights are associated with less than an entirety of the one or more parameter values.

7 . The computer-implemented method of claim 1 , wherein each parameter weight is indicative of a level of importance of an associated parameter value to the user.

8 . A system for determining purchase recommendations, comprising:
at least one memory storing instructions and a machine-learning model; at least one display; and at least one processor operatively connected with the at least one memory and the at least one display, and configured to execute the instructions to perform operations, including:
accessing historical data indicative of one or more of activity or behavior of a user;
determining a respective value for each of one or more parameters of an item and one or more parameter weights for the one or more parameters by applying a machine-learning model to the historical data;
generating one or more parameter similarity values for each item of a set of items based on the respective parameter values;
determining a sort score for each item of the set of items based on the one or more parameter weights and the one or more generated parameter similarity values; <sorting the set of items based on the sort score of each item of the set of items; and
causing a user interface of a user device associated with the user to output at least a portion of the sorted set of items.

9 . The system of claim 8 , wherein the determining of the respective values and parameter weights for the one or more parameters includes applying the machine-learning model to the historical data so as to learn one or more user preferences over time and using the one or more user preferences to determine the respective values and parameter weights.

10 . The system of claim 8 , wherein the machine-learning model is configured to determine a respective parameter, parameter value, or parameter weight based on the historical data.

11 . The system of claim 8 , wherein the machine-learning model is configured to apply one or more filters to a domain of items based on the historical data.

12 . The system of claim 8 , wherein each of the respective parameter values is indicative of a respective aspect of the item.

13 . The system of claim 8 , wherein the one or more parameter weights are indicative of a ranked order of importance of the one or more parameters to the user.

14 . The system of claim 8 , wherein the one or more parameter weights are associated with less than an entirety of the one or more parameter values.

15 . The system of claim 8 , wherein each parameter weight is indicative of a level of importance of an associated parameter value to the user.

16 . A computer-implemented method for determining purchase recommendations, comprising:
accessing historical data indicative of one or more of activity or behavior of a user; determining a respective value for each of one or more parameters of an item and one or more parameter weights for the one or more parameters by applying a machine-learning model to the historical data wherein:
each of the respective parameter values is indicative of a respective aspect of the item; and
the one or more parameter weights are indicative of one or more of a level of importance of each of the one or more parameters to the user or a ranked order of importance of the one or more parameters to the user;
applying one or more filters to a domain of items based on the historical data to generate a set of items; generating one or more parameter similarity values for each item of the set of items based on the respective parameter values; determining a sort score for each item of the set of items based on the one or more parameter weights and the one or more generated parameter similarity values; sorting the set of items based on the sort score of each item of the set of items; and causing a user interface of a user device associated with the user to output at least a portion of the sorted set of items.

17 . The computer-implemented method of claim 16 , wherein the determining of the respective values and parameter weights for the one or more parameters includes applying the machine-learning model to the historical data so as to learn one or more user preferences over time and using the one or more user preferences to determine the respective values and parameter weights.

18 . The computer-implemented method of claim 16 , wherein each parameter weight is indicative of a level of importance of an associated parameter value to the user.

19 . The computer-implemented method of claim 16 , wherein the machine-learning model is configured to determine a respective parameter, parameter value, or parameter weight based on the historical data.

20 . The computer-implemented method of claim 16 , wherein the one or more parameter weights are associated with less than an entirety of the one or more parameter values.

Assignees

• Capital One Services Llc

Inventors

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

Classifications

• G06Q30/0627

  by specifying product or service characteristics, e.g. product dimensions · CPC title

• G06Q30/0629

  by pre-processing results, e.g. ranking or ordering results · CPC title

• G06Q30/0631 Primary

  Recommending goods or services · CPC title