PRODUCT AND CONTENT ASSOCIATION

• US 20200134689A1
• Filed: 11/04/2019
• Published: 04/30/2020
• Est. Priority Date: 12/12/2013
• Status: Active Grant

• Current Assignee: Transform Sr Brands, LLC (ESL Investments, Inc.)
• Original Assignee: Transform Sr Brands, LLC (ESL Investments, Inc.)
• Inventors: Chung, Eui, Sridharan, Bharath, Ebrahimi, Seyed Alireza Forouzan
• Granted Patent: US 11,127,063 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0627 using item specifications

Abstract

Methods and apparatus are disclosed regarding an e-commerce system that maintains references between products and relevant content. In some embodiments, methods and/or apparatus obtain content from one or more content providers via a computer network, identify a product from a product catalog of an electronic database that is related to the obtained content; and update references to relevant content maintained in an electronic database for the product to include a reference to the obtained content.

Claims

1. A computer-implemented method, comprising:

obtaining content from one or more content providers via a computer network;

identifying a product from a product catalog of an electronic database that is related to the obtained content; and

updating the electronic database to include for the product a reference to the obtained content.

2. The computer-implemented method of claim 1, further comprising presenting a customer with a product listing for the product that comprises the reference to the obtained content.

3. The computer-implemented method of claim 1, wherein said identifying comprises:

extracting relevant phrases from the content;

ranking the phrases based on weighted term frequency;

selecting phrases based on their weighted term frequency; and

selecting the product from the product catalog based on the selected phrases.

4. The computer-implemented method of claim 3, further comprising removing blacklisted phrases from the extracted phrases prior to said ranking.

5. The computer-implemented method of claim 1, wherein said identifying comprises:

extracting context from the content based on natural language processing and a set of topics to obtain a distribution for the content across the set of topics;

extracting context for the product based on applying natural language processing and a set of topics to its product listing to obtain a distribution for the product listing across the set of topics;

obtaining a distance measure between the distribution for the content and the distribution for the product listing; and

selecting the product based on the distance measure.

6. The computer-implemented method of claim 1, wherein the natural language processing uses Latent Dirichlet Allocation to obtain the distribution for the content and the distribution for the product.

7. A non-transitory computer-readable medium, comprising a plurality of instructions, that in response to being executed, result in a computing device:

obtaining content from one or more content providers via a computer network;

identifying a product from a product catalog of an electronic database that is related to the obtained content; and

updating the electronic database to include for the product a reference to the obtained content.

8. The non-transitory computer-readable medium of claim 7, further comprising instructions that result in the computing device presenting a customer with a product listing for the product that comprises the reference to the obtained content.

9. The non-transitory computer-readable medium of claim 7, further comprising instructions that result in the computing device:

extracting relevant phrases from the content;

ranking the phrases based on weighted term frequency;

selecting phrases based on their weighted term frequency; and

selecting the product from the product catalog based on the selected phrases.

10. The non-transitory computer-readable medium of claim 9, further comprising instructions that result in the computing device removing blacklisted phrases from the extracted phrases prior to ranking the phrases.

11. The non-transitory computer-readable medium of claim 7, further comprising instructions that result in the computing device:

extracting context from the content based on natural language processing and a set of topics to obtain a distribution for the content across the set of topics;

extracting context for the product based on applying natural language processing and a set of topics to its product listing to obtain a distribution for the product listing across the set of topics;

obtaining a distance measure between the distribution for the content and the distribution for the product listing; and

selecting the product based on the distance measure.

12. The non-transitory computer-readable medium of claim 11, further comprising instructions that result in the computing device performing the natural language processing in accordance with Latent Dirichlet Allocation to obtain the distribution for the content and the distribution for the product.

13. A computing device, comprising

a network interface to a computer network;

an electronic database comprising a product catalog having a plurality of product listings for a plurality of products; and

a processor configured to;

obtain content from one or more content providers via the network interface;

identify a product from the product catalog that is related to the obtained content; and

update the electronic database to include for the identified product a reference to the obtained content.

14. The computing device of claim 13, wherein the processor is further configured to present, via the network interface, a product listing for the product that comprises the reference to the obtained content.

15. The computing device of claim 13, wherein the processor is further configured to:

extract relevant phrases from the content;

rank the phrases based on weighted term frequency;

select phrases based on their weighted term frequency; and

select the product from the product catalog based on the selected phrases.

16. The computing device of claim 15, wherein the processor is further configured to remove blacklisted phrases from the extracted phrases prior to ranking the phrases.

17. The computing device of claim 13, wherein the processor is further configured to:

extract context from the content based on natural language processing and a set of topics to obtain a distribution for the content across the set of topics;

extract context for the product based on applying natural language processing and a set of topics to its product listing to obtain a distribution for the product listing across the set of topics;

obtain a distance measure between the distribution for the content and the distribution for the product listing; and

select the product based on the distance measure.

18. The computing device of claim 17, wherein the processor is further configured to performing the natural language processing in accordance with Latent Dirichlet Allocation to obtain the distribution for the content and the distribution for the product.