Systems and methods for generating keyword-specific content with category and facet information

Patent metadata

Field	Value
Publication number	US-12099540-B2
Application number	US-202318104091-A
Country	US
Kind code	B2
Filing date	Jan 31, 2023
Priority date	Jan 31, 2023
Publication date	Sep 24, 2024
Grant date	Sep 24, 2024

Abstract

Systems and methods of generating keyword-specific content are disclosed. A request for including a keyword is received and the keyword is classified as one of catalog related or unrelated. When the keyword is catalog related, the keyword is categorized in a category associated with the catalog and at least one term in the keyword is categorized in a facet category associated with the catalog. A content template is obtained. The content template is a category specific template when the keyword is catalog related and a generic template when the keyword is catalog unrelated. The category specific template is populated with the at least one term at a position associated with the one of the plurality of facet categories. Responsive content including the category specific template populated with the at least one term when the keyword is catalog related and the generic template when the keyword is catalog unrelated, is transmitted.

First claim

What is claimed is:

1. A system, comprising:
a non-transitory memory;
a processor communicatively coupled to the non-transitory memory, wherein the processor is configured to read a set of instructions to:
receive a request for responsive content including a keyword;
classify the keyword as one of catalog related or catalog unrelated, wherein catalog related keywords are related to a catalog of items;
in response to classifying the keyword as catalog related:
categorize the keyword in one of a plurality of categories associated with the catalog;
categorize at least one term in the keyword in one of a plurality of facet categories associated with the catalog;

obtain a content template, wherein the content template includes a category specific template when the keyword is classified as catalog related, and wherein the content template includes a generic template when the keyword is classified as catalog unrelated;
in response to obtaining the category specific template, populate the category specific template with the at least one term, wherein the at least one term is inserted into the category specific template at a position associated with the one of the plurality of facet categories; and
transmit responsive content to a system that generated the request, wherein the responsive content includes the category specific template populated with the at least one term when the keyword is classified as catalog related, and wherein the responsive content includes the generic template when the keyword is classified as catalog unrelated.

2. The system of claim 1 , wherein the keyword is classified by a trained Bidirectional Encoder Representations from Transformers (BERT) model.

3. The system of claim 1 , wherein the keyword is categorized in one of the plurality of categories associated with the catalog by a trained semantic similarity categorization model.

4. The system of claim 3 , wherein the trained semantic similarity categorization model comprises a two-tower semantic categorization model.

5. The system of claim 1 , wherein the at least one term in the keyword is categorized in one of a plurality of facet categories by a trained question answer model.

6. The system of claim 5 , wherein the trained question answer model comprises a Robustly Optimized BERT-Pretraining Approach (RoBERTa) model.

7. The system of claim 1 , wherein categorization of the at least one term in the keyword in the one of the plurality of facet categories associated with the catalog is based, in part, on the categorization of the keyword into the one of the plurality of categories.

8. The system of claim 1 , wherein the content template is obtained based, in part, on the categorization of the keyword into the one of the plurality of categories.

9. A computer-implemented method, comprising:
receiving a request for responsive content including a keyword;
classifying the keyword as one of catalog related or catalog unrelated, wherein catalog related keywords are related to a catalog of items;
in response to classifying the keyword as catalog related:
categorizing the keyword in one of a plurality of categories associated with the catalog;
categorizing at least one term in the keyword in one of a plurality of facet categories associated with the catalog;

obtaining a content template, wherein the content template includes a category specific template when the keyword is classified as catalog related, and wherein the content template includes a generic template when the keyword is classified as catalog unrelated;
in response to obtaining the category specific template, populating the category specific template with the at least one term, wherein the at least one term is inserted into the category specific template at a position associated with the one of the plurality of facet categories; and
transmitting responsive content to a system that generated the request, wherein the responsive content includes the category specific template populated with the at least one term when the keyword is classified as catalog related, and wherein the responsive content includes the generic template when the keyword is classified as catalog unrelated.

10. The method of claim 9 , wherein the keyword is classified by a trained Bidirectional Encoder Representations from Transformers (BERT) model.

11. The method of claim 9 , wherein the keyword is categorized in one of the plurality of categories associated with the catalog by a trained semantic similarity categorization model.

12. The method of claim 11 , wherein the trained semantic similarity categorization model comprises a two-tower semantic categorization model.

13. The method of claim 9 , wherein the at least one term in the keyword is categorized in one of a plurality of facet categories by a trained question answer model.

14. The method of claim 13 , wherein the trained question answer model comprises a Robustly Optimized Bidirectional Encoder Representations from Transformers (BERT)-Pretraining Approach (RoBERTa) model.

15. The method of claim 9 , wherein categorization of the at least one term in the keyword in the one of the plurality of facet categories associated with the catalog is based, in part, on the categorization of the keyword into the one of the plurality of categories.

16. The method of claim 9 , wherein the content template is obtained based, in part, on the categorization of the keyword into the one of the plurality of categories.

17. A non-transitory computer-readable medium having instructions stored thereon, wherein the instructions, when executed by at least one processor, cause a device to perform operations comprising:
receiving a request for responsive content including a keyword;
classifying the keyword as one of catalog related or catalog unrelated, wherein catalog related keywords are related to a catalog of items, and wherein the keyword is classified by a trained Bidirectional Encoder Representations from Transformers (BERT) model;
in response to classifying the keyword as catalog related:
categorizing the keyword in one of a plurality of categories associated with the catalog using a trained semantic similarity categorization model;
categorizing at least one term in the keyword in one of a plurality of facet categories associated with the catalog by a trained question answer model;

obtaining a content template, wherein the content template includes a category specific template when the keyword is classified as catalog related, and wherein the content template includes a generic template when the keyword is classified as catalog unrelated;
in response to obtaining the category specific template, populating the category specific template with the at least one term, wherein the at least one term is inserted into the category specific template at a position associated with the one of the plurality of facet categories; and
transmitting responsive content to a system that generated the request, wherein the responsive content includes the category specific template populated with the at least one term when the keyword is classified as catalog related, and wherein the responsive content includes the generic template when the keyword is classified as catalog unrelated.

18. The non-transitory computer-readable medium of claim 17 , wherein the trained semantic similarity categorization model comprises a two-tower semantic categorization model.

19. The non-transitory computer-readable medium of claim 17 , wherein the trained question answer model comprises a Robustly Optimized Bidirectiona

Assignees

• Walmart Apollo Llc

Inventors

• Wang Zigeng
• Kim Jae Young
• Yao Tong
• Shen Wei

Classifications

• G06F40/186

  Templates · CPC title

• G06F16/3347

  using vector based model · CPC title

• G06F16/2423

  Interactive query statement specification based on a database schema · CPC title

• G06F16/334

  Query execution (filtering based on additional data G06F16/335) · CPC title

• G06F16/353 Primary

  into predefined classes · CPC title