Dynamic evaluation and use of global and contextual personas

Patent metadata

Field	Value
Publication number	US-2023419362-A1
Application number	US-202318462588-A
Country	US
Kind code	A1
Filing date	Sep 7, 2023
Priority date	Jan 21, 2020
Publication date	Dec 28, 2023
Grant date	—

Abstract

Systems and methods for content selection and presentation are disclosed. Training data including indicating one or more interactions with one or more content elements and associated with one of a plurality of individual contexts is received. A selection model is trained by applying a reinforcement learning mechanism and an individual explore-exploit mechanism. A context for a user is selected by applying the selection model, which is configured to determine an expected future reward value of at least one of the plurality of individual contexts, determine an expected future reward value of a global context based on a past click-through rate, a reward value, and a future click-through rate, and, select the global context or one of the one or more individual contexts based on a comparison of the expected future reward values.

First claim

What is claimed is:

1 . A system comprising:
a memory resource storing instructions; and a processor coupled to the memory resource, the processor being configured to execute the instructions to:
receive training data including one or more interactions with one or more content elements and associated with one of a plurality of individual contexts;
train a selection model by applying a reinforcement learning mechanism and an individual explore-exploit mechanism;
select a context to be assigned to a user by applying the selection model, wherein the selection model is configured to:
determine an expected future reward value of at least one of the plurality of individual contexts;
determine an expected future reward value of a global context based on a past click-through rate, a reward value for a selected interaction, and a future click-through rate; and
based on a comparison of the expected future reward value of the at least one of the plurality of individual contexts and the expected future reward value of the global context, select the global context or the one of the plurality of individual contexts.

2 . The system of claim 1 , wherein the processor executes the instructions further to:
obtain, from a database, a plurality of content elements configured for presentation in a first content container; select and present one of the plurality of content elements in the first content container of the user interface on a display of a device of the user based on the selected one of the global context or one of the plurality of individual contexts.

3 . The system of claim 1 , wherein the expected future reward value of at least one of the plurality of individual contexts is associated with a click-through rate.

4 . The system of claim 1 , wherein the expected future reward value of at least one of the plurality of individual contexts is determined using Thompson sampling.

5 . The system of claim 1 , wherein the training data includes data characterizing a click-through rate, data characterizing a user return rate, or data characterizing historic purchase data.

6 . The system of claim 1 , wherein the trained selection model is trained by combining two or more individual contexts into a single individual context.

7 . The system of claim 6 , wherein the two or more individual contexts are grouped based on a pairwise distance.

8 . The system of claim 1 , wherein the expected future reward value of at least one of the plurality of individual contexts is determined based on a sampled click-through rate, a future click-through rate for each item associated with at least the plurality of individual contexts, and a proportion of impressions for the at least one of the plurality of individual contexts.

9 . A system comprising:
a memory resource storing instructions; and a processor coupled to the memory resource, the processor being configured to execute the instructions to:
receive training data including impression data indicating one or more interactions with one or more content elements and associated with one of a plurality of individual contexts;
train a selection model by applying a reinforcement learning mechanism and an individual explore-exploit mechanism;
select a context to be assigned to a user by applying the selection model, wherein the selection model is configured to:
determine an expected future reward value of at least one of the plurality of individual contexts based on a sampled click-through rate, a future click-through rate for each item associated with at least one of one or more contexts in a training dataset, and a proportion of impressions for the at least one of the plurality of individual contexts;
determine an expected future reward value of a global context; and
based on a comparison of the expected future reward value of the at least one of the plurality of individual contexts and the expected future reward value of the global context, select the global context or one of the one or more individual contexts.

10 . The system of claim 9 , wherein the processor executes the instructions further to:
obtain, from a database, a plurality of content elements configured for presentation in a first content container; select and present one of the plurality of content elements in the first content container of the user interface on a display of a device of the user based on the selected one of the global context or the one of the plurality of individual contexts.

11 . The system of claim 9 , wherein the expected future reward value of at least one of the plurality of individual contexts is associated with a click-through rate.

12 . The system of claim 9 , wherein the expected future reward value of at least one of the plurality of individual contexts is determined using Thompson sampling.

13 . The system of claim 9 , wherein the impression data includes data characterizing a click-through rate, data characterizing a user return rate, or data characterizing historic purchase data.

14 . The system of claim 9 , wherein the trained selection model is trained by combining two or more of the plurality of individual contexts into a single individual context. The system of claim 9 , wherein the two or more of the plurality of individual contexts are grouped based on a pairwise distance.

16 . A computer-implemented method, comprising:
receiving training data including impression data indicating one or more interactions with one or more content elements and associated with one of a plurality of individual contexts; training a selection model by applying a reinforcement learning mechanism and an individual explore-exploit mechanism; selecting a context to be assigned to a user by applying the selection model to implement a context selection process comprising:
determining an expected future reward value of at least one of the plurality of individual contexts;
determining an expected future reward value of a global context based on a past click-through rate, a reward value for a selected interaction, and a future click-through rate; and
based on a comparison of the expected future reward value of the at least one of the plurality of individual contexts and the expected future reward value of the global context, selecting the global context or one of the one or more individual contexts.

17 . The computer-implemented method of claim 16 , wherein the expected future reward value of at least one of the plurality of individual contexts is determined based on a sampled click-through rate, a future click-through rate for each item associated with at least the plurality of individual contexts, and a proportion of impressions for the at least one of the plurality of individual contexts.

18 . The computer-implemented method of claim 16 , wherein the trained selection model is trained by combining two or more of the plurality of individual contexts into a single individual context.

19 . The computer-implemented method of claim 18 , wherein the two or more of the plurality of individual contexts are grouped based on a pairwise distance.

20 . The computer-implemented method of claim 17 , comprising:
obtaining, from a database, a plurality of content elements configured for presentation in a first content container; selecting and presenting one of the plurality of content elements in the first content container of the user interface on a display of the device of the user based on the selected global context or one of the one or more individual contexts.

Assignees

• Walmart Apollo Llc

Inventors

• Deng Yang
• Deshpande Omkar
• Ali Afroza
• Mitra Abhimanyu
• Achan Kannan

Classifications

• G06Q30/0255 Primary

  based on user history · CPC title

• G06Q30/0246

  Traffic · CPC title

• G06Q30/0268

  at point-of-sale [POS] · CPC title