Techniques to leverage machine learning for search engine optimization

Patent metadata

Field	Value
Publication number	US-12061656-B2
Application number	US-202217859758-A
Country	US
Kind code	B2
Filing date	Jul 7, 2022
Priority date	Apr 30, 2019
Publication date	Aug 13, 2024
Grant date	Aug 13, 2024

Abstract

Embodiments are directed to systems, devices, methods, and techniques to determine words or word combinations of search engine queries and selected items corresponding to the words or word combinations. Embodiments also include applying a machine learning model to the word or word combinations and the selected items to determine a mapping between a word or word combination and a particular selected item, and updating attribute data in a web document with the word or word combination, wherein the web document is associated with the particular item such that a search including the word or word combination returns a result including the web document.

First claim

The invention claimed is:

1. An online platform system, comprising:
a processing circuit; and logic stored in computer memory and executed on the processing circuit, the logic operative to cause the processing circuit to:
provide a user interface (UI) comprising content of a webpage, and at least one of a search engine or chatbot, wherein the content of the webpage being separate from content of the at least one of the search engine or chatbot; both configured to receive search engine queries;
receive, via the UI, a set of search engine queries sent to the at least one of search engine or chatbot;
process the set of search engine queries, by the at least one of search engine or chatbot, to identify search terms and select items from the content of the webpage returned as results to the set of search engine queries based on the identified search terms;

determine, for a particular item from the content of the webpage, a set of descriptive words using a trained machine learning model; and
update the content of the webpage with the set of descriptive words, wherein the results of the search engine are updated in response to the content of the webpage being updated.

2. The online platform system of claim 1 , the memory comprising instructions to cause the processing circuit to train the machine learning model on the search terms and the items selected to identify relevant items for specific queries.

3. The online platform system of claim 1 , wherein the set of descriptive words comprise one or more search terms for the particular item identified by the model.

4. The online platform system of claim 1 , the memory comprising instructions to cause the processing circuit to determine the set of descriptive words for the particular item when a relevance value for the descriptive words exceeds a threshold value.

5. The online platform system of claim 1 , wherein the search queries are natural language search queries;
wherein the processing circuit to update the content of the webpage includes the processing circuit to update attribute data for an image or a title associated with the image on the webpage; and
wherein the content being updated is content in proximity to the particular item when displayed on the webpage.

6. The online platform system of claim 1 , the memory comprising instructions to cause the processing circuit to select the set of descriptive words from a plurality of sets of descriptive words based on the set of descriptive words having a highest likelihood of causing a positive change in a search engine result position for a web document corresponding to the particular item.

7. The online platform system of claim 6 , the memory comprising instructions to cause the processing circuit to set the selected set of descriptive words as attribute data for the web document providing the UI.

8. A computer-implemented method, comprising:
generating, by a processor of a system, a user interface (UI) comprising content of a webpage, and at least one of a search engine or chatbot, wherein the content of the webpage being separate from content of the at least one of the search engine or chatbot communications;
receiving, via the UI, a set of queries sent to the at least one of search engine or chatbot;
processing, by the at least one of a search engine or chatbot, the set of queries to identify search terms and select items from the content of the webpage returned as results of the set of queries;
determining, by the system and for a particular item from the content of the webpage, a set of descriptive words using a trained machine learning model; and
updating, by the system, the content of the webpage with the set of descriptive words, wherein the results of the search engine are updated in response to the content of the webpage being updated.

9. The computer-implemented method of claim 8 , comprising training, by the system, the machine learning model on the search terms and the items identified to identify relevant items for specific queries.

10. The computer-implemented method of claim 8 , wherein the set of descriptive words comprise one or more terms for the particular item identified by the model.

11. The computer-implemented method of claim 8 , comprising determining, by the system, the set of descriptive words for the particular item when a relevance value for the descriptive words exceeds a threshold value.

12. The computer-implemented method of claim 8 , wherein the queries are natural language queries;
wherein the processor to update the content of the webpage includes updating attribute data for an image or a title associated with the image on the webpage; and
wherein the content being updated is content in proximity to the particular item when displayed on the webpage.

13. The computer-implemented method of claim 8 , comprising selecting, by the system, the set of descriptive words from a plurality of sets of descriptive words based on the set of descriptive words having a highest likelihood of causing a positive change in a search engine result position for a web document corresponding to the particular item.

14. The computer-implemented method of claim 13 , comprising providing, by the system, the selected set of descriptive words as attribute data for the web document providing the UI.

15. A non-transitory computer-readable medium comprising logic stored in computer memory and executed on a processing circuit, the logic operative to cause the processing circuit to:
provide a user interface (UI) comprising content of a webpage, and at least one of a search engine or chatbot, wherein the content of the webpage being separate from content of the at least one of the search engine or chatbot;
receive, via the UI, a set of search engine queries sent to the at least one of search engine or chatbot;
process the set of search engine queries, by the at least one of search engine or chatbot, to identify search terms and select items from the content of the webpage returned as results to the search engine queries based on the identified search terms;
determine, for a particular item from the content of the webpage, a set of descriptive words using a trained machine learning model; and
update the content of the webpage with the set of descriptive words, wherein the results of the search engine are updated in response to the content of the webpage being updated.

16. The non-transitory computer-readable medium of claim 15 , comprising the processing circuit to determine the set of descriptive words for the particular item when a relevance value for the descriptive words exceeds a threshold value.

17. The non-transitory computer-readable medium of claim 15 , comprising the processing circuit to select the set of descriptive words from a plurality of sets of descriptive words based on the set of descriptive words having a highest likelihood of causing a positive change in a search engine result position for a web document corresponding to the particular item.

18. The online platform system non-transitory computer-readable medium of claim 17 , comprising the system processing circuit to provide the selected set of descriptive words as attribute data for the web document providing the Ul.

19. The non-transitory computer-readable medium of claim 15 , comprising the processing circuit to train the machine learning model on the search terms and the items identified to identify relevant items for specific queries.

20. The non-transitory computer-readable medium of claim 15 , wherein the set of descriptive words c

Assignees

• Capital One Services Llc

Inventors

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

Classifications

• G06F40/237

  Lexical tools · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F16/9538

  Presentation of query results · CPC title

• G06N7/01

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

• G06N3/006

  based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO] · CPC title