Recommendations based on items viewed during a current browsing session

• US 8,620,767 B2
• Filed: 03/14/2013
• Issued: 12/31/2013
• 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 Russell, Zada, Nida K.
• Primary Examiner(s): ZARE, SCOTT A
• Application Number:
  US13/828,146
• Publication Number:
  US 20130198030A1
• Time in Patent Office: 292 Days
• Field of Search: 705/26.7, 705/26.8, 705/27.1
• US Class Current: 705/26.1
• 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 system for selecting items to recommend to users, comprising:

a computer data repository that stores a mapping that maps items to related items, said mapping based at least partly on item relationships discovered by analyzing recorded activities of a plurality of users; and

a computer system comprising one or more physical computers, said computer system programmed to maintain a record of a plurality of items selected by a user for viewing during a browsing session of the user, and programmed to generate session-specific recommendations for the user by at least;

identifying, based on the record and the mapping, a plurality of candidate recommendation items, each candidate recommendation item being related to at least one of the plurality of items selected by the user for viewing during the browsing session;

ranking the plurality of candidate recommendation items; and

selecting, based at least partly on the ranking, a subset of the plurality of candidate recommendation items to recommend to the user during the browsing session;

wherein the computer system is additionally programmed to generate a personalized page that includes;

(1) a visual representation of the plurality of items selected for viewing, said visual representation including, for each of the plurality of items, a respective user-selectable link to a corresponding page, and (2) a visual representation of the subset of items selected to recommend to the user.

2. The system of claim 1, further comprising a mapping generation component that generates the mapping at least partly by analyzing recorded item viewing activities of the plurality of users.

3. The system of claim 2, wherein the mapping generation component measures a degree of relationship between two items at least partly by determining how frequently the two items are viewed within a common browsing session.

4. The system of claim 1, wherein the record of items selected by the user for viewing is an anonymous session record that is not linked to an identity of the user.

5. The system of claim 1, wherein the computer system is programmed to maintain the record of items selected for viewing at least partly by recording visits by the user to item detail pages of an electronic catalog, each visit representing a selection of an item for viewing.

6. The system of claim 1, wherein the computer system is programmed to rank the plurality of candidate recommendation items by assigning a respective score to each candidate recommendation item, each score being dependent upon how frequently the respective candidate recommendation item is selected for viewing in a browsing session in which one or more of said plurality of items is selected for viewing.

7. The system of claim 6, wherein each score is additionally dependent upon whether the respective candidate recommendation item is related to more than one of the plurality of items selected for viewing.

8. The system of claim 1, wherein the computer system is programmed to generate the session-specific recommendations substantially in real time in response to a browsing action performed by the user.

9. The system of claim 1, wherein the mapping is based at least partly on purchase activities of users.

10. The system of claim 1, wherein the computer system provides a user interface that includes functionality for the user to deselect one or more of the plurality of items, and the computer system is programmed to generate, for the user, a refined set of item recommendations that reflects the de-selection.

11. The system of claim 1, wherein the items are products represented in an electronic catalog.

12. The system of claim 1, wherein the items are videos.

13. The system of claim 1, wherein each user-selectable link is to a respective item detail page corresponding to a respective one of the items selected for viewing.

14. The system of claim 1, wherein the personalized page includes:

a plurality of user-selectable links to respective catalog pages corresponding to the items selected for viewing; and

a plurality of user-selectable links to respective catalog pages corresponding to the items selected to recommend.

15. A computer-implemented method for selecting items to recommend to users, comprising:

generating a mapping that maps items to related items based at least partly on item relationships discovered by analyzing recorded activities of a plurality of users;

maintaining a record of a plurality of items selected by a user for viewing during a browsing session of the user;

during the browsing session, generating item recommendations for the user by at least;

identifying, based on the record and the mapping, a plurality of candidate recommendation items, each candidate recommendation item being related to at least one of the plurality of items selected by the user for viewing during the browsing session;

ranking the plurality of candidate recommendation items; and

selecting, based at least partly on the ranking, a subset of the plurality of candidate recommendation items to recommend to the user; and

generating a personalized page that includes;

(1) a visual representation of the plurality of items selected for viewing, said visual representation including, for each of the plurality of items, a respective user-selectable link to corresponding content, and (2) a visual representation of the subset of candidate recommendation items selected to recommend to the user;

said method performed programmatically by a computer system that comprises one or more physical computers.

16. The method of claim 15, wherein generating the mapping comprises, for each of a plurality of item pairs, measuring a degree of relationship between the two items of the pair at least partly by determining how frequently the two items are viewed within a common browsing session.

17. The method of claim 15, wherein the record of items selected by the user for viewing is an anonymous session record.

18. The method of claim 15, wherein maintaining the record of items selected by the user for viewing comprises recording visits by the user to item detail pages of an electronic catalog, each visit representing a selection of an item for viewing.

19. The method of claim 15, wherein ranking the plurality of candidate recommendation items comprises assigning a respective score to each candidate recommendation item, each score being dependent upon how frequently the respective candidate recommendation item is selected for viewing in a browsing session in which one or more of said plurality of items is selected for viewing.

20. The method of claim 19, wherein each score is additionally dependent upon whether the respective candidate recommendation item is related to more than one of the plurality of items selected for viewing.

21. The method of claim 15, wherein the recommendations are generated substantially in real time in response to a browsing action performed by the user during the browsing session.

22. The method of claim 15, wherein the mapping is based at least partly on item purchases of users.

23. The method of claim 15, wherein the personalized page includes functionality for the user to deselect one or more of the plurality of items and to initiate generation of refined recommendations that reflect said de-selection.

24. Non-transitory computer storage that stores executable program instructions that direct a computer system to at least:

generate a mapping that maps items to related items based at least partly on item relationships discovered from recorded activities of a plurality of users;

store a record of a plurality of items selected by a user for viewing during a browsing session of the user;

during said browsing session, generate item recommendations for the user by at least;

identifying, based on the record and the mapping, a plurality of candidate recommendation items, each candidate recommendation item being related to at least one of the plurality of items selected by the user for viewing during the browsing session;

ranking the plurality of candidate recommendation items; and

selecting, based at least partly on the ranking, a subset of the plurality of candidate recommendation items to recommend to the user; and

generate a personalized page that includes;

(1) a visual representation of the plurality of items selected for viewing, said visual representation including, for each of the plurality of items, a respective user-selectable link to corresponding content, and (2) a visual representation of the subset of candidate recommendation items selected to recommend to the user.

25. The non-transitory computer storage of claim 24, wherein each user-selectable link points to a respective item detail page for a respective item selected for viewing.

26. The non-transitory computer storage of claim 25, wherein the visual representation of the subset of candidate recommendation items comprises, for each item selected to recommend, a respective user-selectable link to a corresponding item detail page.

27. The non-transitory computer storage of claim 24, wherein the personalized page includes:

a plurality of user-selectable links to respective catalog pages corresponding to the items selected for viewing; and

a plurality of user-selectable links to respective catalog pages corresponding to the items selected to recommend.

28. The non-transitory computer storage of claim 24, wherein the executable instructions additionally direct the computer system to provide a user interface that includes functionality for the user to deselect one or more of the plurality of items selected for viewing and to access a refined set of item recommendations that reflects the de-selection.

29. A computer-implemented method, comprising:

maintaining a record of a plurality of catalog items selected by a user for viewing in an electronic catalog;

generating a set of item recommendations for the user based on the plurality of catalog items selected for viewing, and based on item-to-item relationship mappings;

generating a personalized page for the user, said personalized page including a listing of the plurality of catalog items selected for viewing and including the set of item recommendations;

providing an option, via a user interface, for the user to deselect one or more of the catalog items selected for viewing;

detecting de-selection by the user of one or more of the catalog items selected for viewing; and

generating and outputting a refined set of item recommendations for the user that reflects the de-selection;

said method performed in its entirety by a computer system that comprises a physical server.

30. The method of claim 29, wherein the method is performed in its entirety during a single browsing session of the user.

31. The computer-implemented method of claim 29, wherein the personalized page includes hyperlinks to product detail pages corresponding to (1) the catalog items selected for viewing and (2) items included in the set of item recommendations.

32. The computer-implemented method of claim 29, wherein the option to deselect one or more items is provided on said personalized page.