Product recommendations based on collaborative filtering of user data

• US 7,590,562 B2
• Filed: 06/29/2005
• Issued: 09/15/2009
• Est. Priority Date: 06/29/2005
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Stoppelman, Michael
• Primary Examiner(s): Smith; Jeffrey A
• Assistant Examiner(s): Shah; Amee A
• Application Number:
  US11/168,561
• Publication Number:
  US 20070005437A1
• Time in Patent Office: 1,539 Days
• Field of Search: 705/26, 705/27
• US Class Current: 705/26.7
• CPC Class Codes: G06F 16/904 Browsing; Visualisation therefor for navigating the web G06F16/954; browsing optimisation for the web G06F16/957
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 30/0633 Lists, e.g. purchase orders, compilation or processing
• CPC Class Codes: G06Q 30/0641 Shopping interfaces

Abstract

A system gathers user behavior data from a group of web retailers and/or non-web retailers, analyzes the user behavior data to identify product recommendations for products offered by the web retailers, and provides one of the identified product recommendations in connection with a product page associated with one of the web retailers.

Claims

1. A method, comprising:

gathering, by a server device, user behavior data relating to user purchase activity and user non-purchase activity that are both associated with a plurality of web retailers;

analyzing, by the server device, the user behavior data to identify product recommendations that correspond to products offered for sale by the web retailers;

receiving, by the server device and from a client device, a request identifying one of the web retailers and a particular product, offered for sale by the identified web retailer, that is currently being accessed by the client device;

identifying, by the server device and in response to the request, a plurality of the identified product recommendations relating to the particular product;

determining, by the server device, products offered for sale by the identified web retailer;

identifying, by the server device, a particular one or more of the plurality of the identified product recommendations that correspond to the products offered for sale by the identified web retailer; and

providing, by the server device, the particular one or more of the plurality of the identified product recommendations in connection with a product page, served by the identified web retailer, that presents information regarding the particular product.

2. The method of claim 1, where gathering user behavior data includes:

obtaining, from the web retailers, data relating to the user purchase activity associated with the web retailers.

3. The method of claim 2, where analyzing the user behavior data includes:

identifying products that were purchased together, based on the data relating to user purchase activity, to determine product recommendations for the products offered by the web retailers.

4. The method of claim 1, where gathering user behavior data includes:

obtaining, from a plurality of client devices associated with a plurality of users, data relating to the user non-purchase activity associated with the web retailers.

5. The method of claim 4, where analyzing the user behavior data includes:

identifying products that the users spent at least a particular amount of time accessing, based on the data relating to user non-purchase activity, to determine product recommendations for the products offered by the web retailers.

6. The method of claim 1, where analyzing the user behavior data includes:

identifying products that were purchased together, based on data relating to the user purchase activity, and products that users spent at least a particular amount of time accessing, based on data relating to the user non-purchase activity, to determine product recommendations for the products offered by the web retailers.

7. The method of claim 1, where analyzing the user behavior data includes:

determining that a first product is related to a second product when the first product occurs in conjunction with the second product in the user behavior data at least a particular number of times.

8. The method of claim 1, further comprising:

creating, by the server device, a database based on the identified product recommendations.

9. The method of claim 8, where creating a database includes:

storing a product name associated with a product in the database; and

storing a list of one or more products that are recommended for the product in the database.

10. The method of claim 1, further comprising:

normalizing, by the server device, names of products in the user behavior data prior to identifying product recommendations for products offered by the web retailers.

11. The method of claim 1, where analyzing the user behavior data includes:

determining that a first product is a recommended product for a second product when the user behavior data indicates that users purchased the first product and the second product together.

12. The method of claim 1, where analyzing the user behavior data includes:

determining that a first product is a recommended product for a second product when the user behavior data indicates that users spent at least a particular amount of time accessing a product page associated with the first product before or after purchasing the second product.

13. The method of claim 1, where analyzing the user behavior data includes:

determining that a first product is a recommended product for a second product when the user behavior data indicates that users spent at least a particular amount of time accessing product pages associated with both the first product and the second product during a same online session.

14. The method of claim 1, where gathering user behavior data further includes:

gathering user behavior data from at least one non-web retailer.

15. A system, comprising:

one or more devices comprising;

means for receiving, from a client, a request that identifies a web retailer and a particular product of a plurality of products offered for sale by the web retailer;

means for identifying, in response to the request, a plurality of product recommendations for the particular product based on data relating to user activity associated with a plurality of different web retailers and a plurality of different non-web retailers;

means for determining products offered for sale by the web retailer;

means for identifying a particular one or more of the plurality of product recommendations that correspond to the products offered for sale by the web retailer; and

means for supplying the particular one or more of the plurality of product recommendations with a product page served by the web retailer.

16. A system, comprising:

one or more devices comprising;

a data collector to obtain information relating to user purchases from a plurality of web retailers by a plurality of users;

a recommended products identifier to identify one of a plurality of first products as a recommended product for a second product offered for sale by one of the web retailers based on the information relating to user purchases; and

a recommended products supplier to;

receive a request identifying the one of the web retailers and the second product offered for sale by the one of the web retailers,

identify, in response to the request, the plurality of first products as recommended products for the second product,

identify a plurality of products offered for sale by the one of the web retailers,

identify the one of the plurality of first products as one of the plurality of products that are offered for sale by the one of the web retailers, and

present information regarding the one of the plurality of first products for display with a product page associated with the second product and the one of the web retailers.

17. The system of claim 16, where the recommended products identifier is configured to identify the one of the plurality of first products as a recommended product for the second product when the information relating to user purchases indicates that the one of the plurality of first products and the second product were purchased together.

18. The system of claim 16, where the data collector is further configured to obtain data relating to an amount of time that users spent accessing products associated with the web retailers;

and

where the recommended products identifier is further configured to identify the one of the plurality of first products as a recommended product for the second product when the information relating to an amount of time that users spent accessing products indicates that users spent at least a particular amount of time accessing both the one of the plurality of first products and the second product.

19. The system of claim 16, where the recommended products identifier is configured to identify the one of the plurality of first products as a recommended product for the second product when the one of the plurality of first products occurs in conjunction with the second product in the information relating to user purchases at least a predetermined number of times.

20. The system of claim 16, further comprising:

a database to store a product name associated with the second product and a list of one or more other products, including the one of the plurality of first products, that are recommended for the second product.

21. The system of claim 16, wherein where the data collector is farther configured to obtain information relating to user non-purchase activity from at least one non-web retailer and

where the recommended products identifier is configured to identify the one of the plurality of first product products as a recommended product for the second product based on the information relating to user non-purchase activity from the at least one non-web retailer.

22. A memory device that stores instructions executable by at least one processor, comprising:

instructions for identifying product recommendations for products offered by a plurality of web retailers on corresponding web sites based on user behavior data relating to user activity associated with the web retailers and a plurality of users;

instructions for storing the identified product recommendations in a database;

instructions for receiving a request identifying one of the web retailers and a particular product offered for sale by the identified web retailer;

instructions for identifying, in response to the request, a plurality of the identified product recommendations in the database that correspond to the particular product;

instructions for identifying products offered for sale by the identified web retailer

instructions for identifying a particular one or more of the plurality of the identified product recommendations that correspond to the products offered for sale by the identified web retailer; and

instructions for providing the particular one or more of the identified product recommendations in connection with a product page, associated with the particular product, from a web site corresponding to the identified web retailer.

23. The memory device of claim 22, further comprising:

instructions for charging the identified web retailer a fee based on receiving the request.

24. The memory device of claim 22, further comprising:

instructions for receiving information regarding a selection of one of the particular one or more of the plurality of the identified product recommendations; and

instructions for charging the identified web retailer a fee based on the received information.

25. The memory device of claim 22, further comprising:

instructions for receiving first information regarding a selection of one of the particular one or more of the plurality of the identified product recommendations;

instructions for receiving second information that indicates that the particular one or more of the plurality of the identified product recommendations has been purchased; and

instructions for charging the identified web retailer a fee based on the received second information.

26. A method, comprising:

gathering, by a server device, user behavior data relating to user activity, of a plurality of users, in connection with at least one web retailer and at least one non-web retailer;

analyzing, by the server device, the user behavior data to identify product recommendations for products offered by a plurality of web retailers;

receiving, by the server device and from a client device, a request identifying a particular web retailer of the plurality of web retailers and a particular product offered by the particular web retailer;

identifying, by the server device and in response to the recluest, a plurality of the product recommendations relating to the particular product;

determining, by the server device, products offered by the particular web retailer;

identifying, by the server device, a particular one or more of the plurality of the product recommendations that correspond to the products offered by the particular web retailer; and

providing, by the server device, the particular one or more of the plurality of the product recommendations in connection with a product page associated with the particular web retailer and the particular product offered by the particular web retailer.

27. The method of claim 26, where the user activity includes user purchase activity and user non-purchase activity that occur on web sites of the plurality of web retailers at least one web retailer and the at least one non-web retailer.

28. The method of claim 26, where the particular product is currently being accessed by the client device on a web site of the particular web retailer.