Discovery of behavior-based item relationships

• US 8,433,621 B2
• Filed: 06/18/2012
• Issued: 04/30/2013
• Est. Priority Date: 09/18/1998
• Status: Expired due to Fees

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon.com, Inc.
• Inventors: Linden, Gregory D., Smith, Brent R., Zada, Nida K.
• Primary Examiner(s): ZARE, SCOTT A
• Application Number:
  US13/526,393
• Publication Number:
  US 20120259729A1
• Time in Patent Office: 316 Days
• Field of Search: 705/7.29, 705/26.7, 705/26.8
• US Class Current: 705/26.7
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0253 During e-commerce, i.e. online transactions
• CPC Class Codes: G06Q 30/0254 based on statistics
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: G06Q 30/0271 Personalized advertisement
• CPC Class Codes: G06Q 30/0277 Online advertisement
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0603 Catalogue ordering
• 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
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

Various processes are disclosed for discovering item relationships between particular items, such as products represented in an electronic catalog, based on monitored user behaviors (e.g., item viewing activities, item purchases, shopping cart activities, etc.). The discovered item relationships may, for example, be used to generate personalized item recommendations for users, and/or to supplement item detail pages of an electronic catalog with lists of related items. Also disclosed are processes for generating personalized item recommendations based on users'"'"' search activities and browse node visits.

Claims

1. A method of discovering item relationships, the method comprising:

storing, in computer storage, records of electronic shopping cart contents of each of a plurality of users of an interactive system that provides electronic shopping carts, said records of shopping cart contents identifying particular items selected by users to add to their electronic shopping carts, wherein the records of electronic shopping cart contents include, for particular users, records of recently removed shopping cart contents, said records of recently removed shopping cart contents specifying items that, within a selected period of time, were removed from the electronic shopping carts without being purchased; and

generating, based at least partly on the records of electronic shopping cart contents, including the records of recently removed shopping cart contents, a mapping that maps items to corresponding sets of related items;

said method performed by a computing system that comprises one or more computers.

2. The method of claim 1, wherein generating said mapping comprises generating data values for specific item pairs, each data value representing a degree of relationship between a first item and a second item and being dependent upon how frequently both the first item and the second item are added to a common electronic shopping cart.

3. The method of claim 2, wherein generating the data values comprises counting a number times that both the first item and the second item were concurrently in a common electronic shopping cart.

4. The method of claim 2, wherein generating the data values comprises counting a number of sessions in which both the first item and the second item were added to a common electronic shopping cart.

5. The method of claim 1, further comprising using the mapping to generate personalized item recommendations for each of a plurality of users.

6. The method of claim 5, wherein using the mapping to generate personalized item recommendations for each of a plurality of users comprises replicating the mapping across a plurality of physical servers that are configured with executable code for using the mapping to generate personalized item recommendations.

7. The method of claim 1, wherein generating the mapping comprises performing a co-occurrence analysis of aggregated records of shopping cart contents of the plurality of users.

8. A system for discovering item relationships, the system comprising:

a computer data repository that contains records of electronic shopping cart contents associated with each of a plurality of users of an interactive system that provides electronic shopping carts, said records of shopping cart contents identifying particular items selected by users to add to their electronic shopping carts, said records of shopping cart contents including, for particular users, records of recently removed shopping cart contents, said records of recently removed shopping cart contents specifying items that, within a selected period of time, were removed from the electronic shopping carts without being purchased; and

a computing system programmed to generate, based at least partly on the records of electronic shopping cart contents, including the records of recently removed shopping cart contents, a mapping that maps specific items to sets of related items.

9. The system of claim 8, wherein the computing system is programmed to generate data values for specific item pairs, each data value representing a degree of relationship between two items of a respective item pair and being dependent upon how many times the two items have co-occurred in an electronic shopping cart.

10. The system of claim 9, wherein the computing system is programmed to generate the data values using a method that comprises, for an item pair consisting of a first item and a second item, counting a number of times that both the first item and the second item were concurrently in a common electronic shopping cart.

11. The system of claim 9, wherein the computing system is programmed to generate the data values using a method that comprises, for an item pair consisting of a first item and a second item, counting a number of sessions in which both the first item and the second item were added to a common electronic shopping cart.

12. The system of claim 9, wherein the computing system, in generating the mapping, is programmed to perform an item co-occurrence analysis of aggregated records of shopping cart contents of a plurality of users.

13. The system of claim 8, further comprising a recommendation system configured to use the mapping to generate personalized item recommendations for users.

14. The system of claim 13, wherein the recommendation system comprises a plurality of physical servers, each of which stores a copy of said mapping and stores executable code for using the mapping to generate personalized item recommendations.

15. The system of claim 8, wherein the computing system is programmed to generate the mapping without considering purchases made by the users.

16. Non-transitory computer storage which stores executable program code that instructs a computing system to discover item relationships by a process that comprises:

storing, in computer storage, records of electronic shopping cart contents of each of a plurality of users, said records of shopping cart contents identifying particular items selected by users to add to their electronic shopping carts, wherein the records of electronic shopping cart contents include, for particular users, records of recently removed shopping cart contents, said records of recently removed shopping cart contents specifying items that, within a selected period of time, were removed from the electronic shopping carts without being purchased; and

generating, based at least partly on an aggregated analysis of the records of electronic shopping cart contents, including the records of recently removed shopping cart contents, a mapping that maps items to related items, wherein generating the mapping comprises generating data values for specific item pairs, each data value representing a degree of relationship between a first item and a second item and being dependent upon how frequently both the first item and the second item are added to a common electronic shopping cart.

17. The non-transitory computer storage of claim 16, wherein generating the data values comprises counting a number times that both the first item and the second item were concurrently in a common electronic shopping cart.

18. The non-transitory computer storage of claim 16, wherein generating the data values comprises counting a number of sessions in which both the first item and the second item were added to a common electronic shopping cart.

19. The non-transitory computer storage of claim 16, wherein the executable program code further instructs the computing system to use the mapping to generate personalized item recommendations for each of a plurality of users.

20. The non-transitory computer storage of claim 16, wherein the executable program code further instructs the computing system to generate the mapping without considering purchases made by the users.