Method, system, and computer readable medium for the selection of content items for presentation to users

• US 7,536,320 B2
• Filed: 11/16/2007
• Issued: 05/19/2009
• Est. Priority Date: 08/13/2003
• Status: Expired due to Term

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon.com, Inc.
• Inventors: McQueen, Clyde D III, Brinkerhoff, Richard, Hyman, Michael Irvin
• Primary Examiner(s): Smith; Jeffrey A.
• Assistant Examiner(s): Allen; William J.
• Application Number:
  US11/941,818
• Publication Number:
  US 20080071647A1
• Time in Patent Office: 550 Days
• Field of Search: 705/14, 705 26- 27
• US Class Current: 705/26.7
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0245 Surveys
• CPC Class Codes: G06Q 30/0254 based on statistics
• 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 30/0623 Item investigation
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 30/0641 Shopping interfaces
• CPC Class Codes: G06Q 50/10 Services

Abstract

Various computer-implemented methods are disclosed for selecting content items, including but not limited to user-submitted blurbs, to present to users. One such method comprises storing user-specific item selection history data identifying particular items, such as catalog items, selected by particular users, and recording responses of a plurality of the users to a content item. The method further comprises determining whether the content item should be presented to a target user based, at least in part, on a degree to which the item selection history of the target user is similar to item selection histories of users who responded favorably to the content item.

Claims

1. A computer-implemented method of selecting content items for presentation to users, the method comprising:

storing user-specific item selection history data in computer storage, said item selection history data identifying particular items selected by particular users;

recording user responses of a plurality of said users to a content item, said content item being distinct from the items represented in said item selection history data, said user responses including at least favorable responses to the content item and unfavorable responses to the content item; and

automatically determining, by a personalized selection system that comprises computer hardware, whether the content item should be presented to a target user based, at least in part, on a degree to which an item selection history of the target user is similar to item selection histories of users who responded favorably to the content item, as determined using said item selection history data and said recorded user responses, wherein automatically determining whether the content item should be presented to a target user further comprises taking into consideration a degree to which the target user'"'"'s item selection history is dissimilar to item selection histories of users who responded unfavorably to the content item.

2. The method of claim 1, wherein the user-specific item selection history data comprises purchase history data that identifies catalog items purchased by the users from an electronic catalog of items.

3. The method of claim 1, wherein determining whether the content item should be presented to the target user comprises determining, for each of a plurality of catalog items selected by the target user, an extent to which users who have selected the respective catalog item have tended to respond favorably to the content item.

4. The method of claim 1, wherein the content item is a blurb submitted by a user of said electronic catalog.

5. The method of claim 4, wherein the method further comprises receiving the blurb via a blurb submission pipeline of a web site that hosts the electronic catalog, said blurb submission pipeline enabling users to author and submit blurbs for viewing by other users.

6. The method of claim 1, wherein determining whether the content item should be presented to the target user comprises:

for each of a plurality of catalog items selected by the target user, looking up a corresponding score, said score reflecting an extent to which responses to the content item by other users who have selected the respective catalog item have been favorable responses; and

combining said scores to generate a cumulative score for the content item, said cumulative score reflecting a degree to which the item selection history of the target user is associated with favorable user responses to the content item.

7. The method of claim 1, wherein the item selection history data identifies particular catalog items selected from an electronic catalog by particular users, and the method comprises using the favorable and unfavorable responses in combination to generate, for specific catalog items, respective scores that reflect degrees to which such catalog items are associated with favorable and unfavorable user responses to the content item.

8. The method of claim 1, wherein recording the user responses comprises, in response to a first user responding favorably to the content item, adjusting item-specific scores for each catalog item in an item selection history of the first user, said item-specific scores being non-user-specific.

9. The method of claim 8, wherein automatically determining whether the content item should be presented to the target user comprises combining the item-specific scores associated with each of a plurality of catalog items identified in the target user'"'"'s item selection history to generate a cumulative score that represents a degree of association between the content item and the target user'"'"'s item selection history.

10. The method of claim 1, wherein the content item is a unit of content other than a blurb.

11. The method of claim 1, further comprising, when a determination is made that the item should be presented to the target user, adding the content item to personal log of the target user.

12. The method of claim 1, wherein the method is performed automatically by a system that comprises a plurality of computers interconnected by a network.

13. A computer-readable storage medium that stores executable software that, when executed by a computer system, causes the computer system to perform a method that comprises:

storing user-specific item selection history data in computer storage, said item selection history data identifying particular items selected by particular users;

recording user responses of a plurality of said users to a content item, said content item being distinct from the items represented in said item selection history data, said user responses including at least favorable responses to the content item and unfavorable responses to the content item; and

automatically determining whether the content item should be presented to a target user based, at least in part, on a degree to which an item selection history of the target user is similar to item selection histories of users who responded favorably to the content item, as determined using said item selection history data and said recorded user responses, wherein automatically determining whether the content item should be presented to a target user further comprises taking into consideration a degree to which the target user'"'"'s item selection history is dissimilar to item selection histories of users who responded unfavorably to the content item.

14. A system for selecting content items to present to users, the system comprising:

a data repository that stores user-specific item selection history data identifying particular items selected by particular users, said data repository comprising computer hardware;

a response recording system that records user responses to particular content items, including at least favorable responses, said content items being distinct from the items identified in the item selection history data, said response recording system comprising computer hardware; and

a personalized selection system that selects particular content items to present to particular users, said personalized selection system configured to select particular content items to present to a particular user based, at least in part, on a degree to which an item selection history of the particular user is similar to item selection histories of users that responded favorably to the particular content items, as determined from the user-specific item selection history data and the recorded user responses, said personalized selection system comprising computer hardware;

wherein the personalized selection system, in determining whether a particular content item should be presented to a particular user, additionally takes into consideration a degree to which the particular user'"'"'s item selection history is dissimilar to item selection histories of users that have responded unfavorably to the content item.

15. The system of claim 14, wherein the user-specific item selection history data comprises purchase history data that identifies catalog items purchased by the users from an electronic catalog.

16. The system of claim 15, wherein the content items are blurbs submitted by users of said electronic catalog.

17. The system of claim 16, further comprising a blurb submission system that enables the users of the electronic catalog to author and submit said blurbs for viewing by other users.

18. The system of claim 14, wherein the user responses include positive and negative votes explicitly cast by users on particular content items, and the personalized selection system uses said positive and negative votes in combination to identify content items that tend to be liked by users with item selection histories similar to an item selection history of a target user.

19. The system of claim 14, wherein the personalized selection system is configured to assess whether a candidate content item should be presented to a target user, at least in part, by determining, for each of a plurality of catalog items identified in the target user'"'"'s item selection history, an extent to which users that have selected the respective catalog item have tended to respond favorably to the candidate content item.

20. The system of claim 14, wherein the response recording system is responsive to a first user responding favorably to a first content item by adjusting item-specific scores for each of a plurality of catalog items identified in an item selection history of the first user, said item-specific scores not being specific to any particular user.

21. The system of claim 14, wherein the content item is a unit of content other than a blurb.

22. The system of claim 14, further comprising a personal log component that enables users to view personal logs containing the content items selected for such users by the personalized selection system.

23. The system of claim 14, wherein the item selection history data identifies particular catalog items selected by particular users from an electronic catalog.

24. The system of claim 23, wherein the personalized selection system, in determining whether to select a particular content item to present to a target user, is operative to:

look up, for each of a plurality of catalog items selected by the target user, a respective score, said score reflecting an extent to which responses to the content item by other users who have selected the respective catalog item have been favorable responses; and

combine said scores to generate a cumulative score for the particular content item, said cumulative score reflecting a degree to which the item selection history of the target user is associated with favorable user responses to the content item.

25. The system of claim 14, wherein at least some of the content items are blog entries submitted by users.

26. The system of claim 14, further comprising at least one web server machine that is operative to serve personalized web pages that include personalized content selected by the personalized selection system.