Recommendations based on progress data

• US 9,153,141 B1
• Filed: 06/30/2009
• Issued: 10/06/2015
• Est. Priority Date: 06/30/2009
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Kane, Francis J. Jr., Killalea, Tom, Mason, Llewllyn J.
• Primary Examiner(s): Patel, Chirag R
• Application Number:
  US12/495,256
• Time in Patent Office: 2,289 Days
• Field of Search: 709224-226, 706 45- 62, 707711-713
• US Class Current: 1/1
• CPC Class Codes: G09B 7/00 Electrically-operated teaching apparatus or devices working with questions and answers mechanically operated G09B3/00; computing arrangements G06F

Abstract

User content access events pertaining to a content item, such as an ebook, audio, video file, and so on, are collected and analyzed to determine progress data, including abandonment information about when the content item, or a portion thereof, has been abandoned. Once determined, recommendations may be presented based on the progress data from similar users.

Claims

1. A method comprising:

identifying, by one or more computing devices, one or more content items;

generating, by at least one computing device of the one or more computing devices, abandonment information pertaining to the one or more content items based at least in part on access events resulting from one or more users accessing the one or more content items, wherein the abandonment information is further based at least in part on one or more varying abandonment thresholds associated with a user of the one or more users;

ranking, by at least one computing device of the one or more computing devices, the one or more content items at least in part according to the abandonment information; and

recommending, by at least one computing device of the one or more computing devices, a potential content item from the one or more content items to one or more of the accessing users.

2. The method of claim 1, further comprising identifying a sample user similar to an accessing user who accesses the content items and identifying those items accessed by the sample user.

3. The method of claim 2, wherein the identifying a sample user comprises evaluating user profiles of the sample user and the accessing user.

4. The method of claim 2, wherein the identifying a sample user comprises evaluating whether the sample user accesses at least one content item in common with the accessing user.

5. The method of claim 2, wherein the sample user comprises a composite that may be partly or wholly synthesized.

6. The method of claim 1, wherein the generating abandonment information comprises computing, based on the access events, whether the one or more users abandoned at least a portion of the one or more content items.

7. The method of claim 1, wherein the generating abandonment information comprises distinguishing among which of the one or more content items the one or more users abandoned and which of the one or more content items the one or more users completed.

8. The method of claim 1, wherein the ranking comprises grouping at least some of the one or more content items into at least three categories comprised of a first category for any of the one or more content items that are in progress of being consumed by the one or more users, a second category for any of the one or more content items that have been completed by the one or more users, and a third category for any of the one or more content items that have been abandoned by the one or more users, wherein the one or more content items are grouped based at least in part on the one or more varying abandonment thresholds associated with the user.

9. The method of claim 8, wherein the recommending comprises recommending as a potential content item to at least one of the one or more users, one of the content items in the second category.

10. The method of claim 8, further comprising providing a negative recommendation to avoid any of the content items in the third category.

11. The method of claim 8, further comprising filtering the one or more content items to remove any of the content items in the third category.

12. The method of claim 1, wherein the ranking comprises grouping the one or more content items according to a percent completion by the one or more users.

13. The method of claim 1, wherein the recommending comprises recommending a potential content item that is ranked least abandoned.

14. The method of claim 1, wherein the recommending comprises recommending at least a portion of the potential content item.

15. The method of claim 1, further comprising filtering the one or more content items based on a preference of at least one user of the one or more users.

16. The method of claim 1, further comprising inferring a preference of at least one user of the one or more users and filtering the one or more content items based on the preference.

17. The method of claim 1, further comprising presenting a recommendation for the user to consider the potential content item.

18. A method comprising:

collecting, by one or more computing devices, progress data for content items being accessed by a plurality of users, the content items having multiple portions and the progress data including abandonment data that specifies varying abandonment thresholds for the plurality of users to determine abandonment of the content items with respect to the plurality of users; and

recommending, by at least one computing device of the one or more computing devices and from the content items, at least a portion of a potential content item to a target user based on the progress data exhibited by the plurality of users.

19. The method of claim 18, wherein the collecting progress data comprises receiving progress data recorded by an electronic book reader.

20. The method of claim 18, wherein the abandonment data comprises access velocity in at least one of the content items measured relative to at least one of (1) an elapsed time since last access of the content item or (2) a position in the content item.

21. The method of claim 18, wherein the abandonment data comprises total access duration of at least one of the content items measured in relation to at least one of (1) different time periods or (2) positions in the content item.

22. The method of claim 18, wherein the progress data is based on frequency of access of the content items.

23. The method of claim 18, wherein at least one of the plurality of users comprises a synthesized composite.

24. The method of claim 18, wherein the recommending is further based on progress data for content items previously accessed by the target user.

25. The method of claim 18, wherein the recommending comprises identifying the potential content item based on predictive modeling.

26. The method of claim 18, wherein the recommending comprises identifying the potential content item based on user-based collaborative filtering.

27. The method of claim 18, wherein the recommending comprises identifying the potential content item based on item-based collaborative filtering.

28. The method of claim 18, wherein the recommending comprises identifying the potential content item based on past user behavior.

29. The method of claim 18, wherein the recommending comprises recommending at least one chapter of an electronic book.

30. The method of claim 18, wherein the recommending comprises providing a negative recommendation to avoid at least one of the content items.

31. The method of claim 18, further comprising filtering recommendation results to remove at least one of the content items.

32. The method of claim 31, wherein the filtering comprises removing the at least one of the content items that exhibits comparatively higher abandonment.

33. The method of claim 18, further comprising identifying the plurality of users as being similar to the target user according to whether characteristics in an access profile of the target user and access profiles of the plurality of users are within a pre-determined threshold of commonality.

34. The method of claim 33, wherein the access profiles comprise historical content item consumption statistics.

35. A method comprising:

generating, by one or more computing devices, abandonment information for a content item consumed on an electronic device, the content item having multiple portions, the abandonment information including first abandonment information associated with a first path in which first users navigated through the content item and second abandonment information associated with a second path in which second users navigated through the content item, wherein the first path is different from the second path indicating and

recommending, by at least one computing device of the one or more computing devices, at least a portion of one or more potential content items to a user based, at least in part, on the first and second abandonment information.

36. The method of claim 35, wherein the content items comprise electronic books having multiple chapters, and wherein the electronic device comprises an electronic book reader.

37. The method of claim 35, wherein the recommending comprises comparing the abandonment information to historical content item consumption statistics for the user.

38. The method of claim 35, wherein the recommending is further based on abandonment patterns for content items previously accessed by the user.

39. The method of claim 35, wherein the recommending comprises providing a negative recommendation to avoid at least one of the content items.

40. The method of claim 35, wherein the recommending comprises identifying the one or more potential content items based on predictive modeling.

41. The method of claim 35, wherein the recommending comprises identifying the one or more potential content items based on user-based collaborative filtering.

42. The method of claim 35, wherein the recommending comprises identifying the one or more potential content items based on item-based collaborative filtering.

43. The method of claim 35, wherein the recommending comprises identifying the one or more potential content items based on past user behavior.

44. A system comprising:

a processor;

a memory accessible by the processor;

a collection module stored in the memory and executable on the processor to collect content access events indicative of content consumption within a a content item having multiple portions;

an abandonment module stored in the memory and executable on the processor to determine abandonment information of the content item based on the content access events, the abandonment information including abandonment patterns for the content item that indicate a likelihood that the content item is to be abandoned, but has yet to be abandoned, by a user, the likelihood being based at least in part upon varying abandonment thresholds associated with the user;

a recommendation module stored in the memory and executable on the processor to recommend a potential content item according to the abandonment information; and

an interface module stored in the memory and executable on the processor to present the potential content item to at least the user.

45. The system of claim 44, wherein the content access events are collected from an electronic book reader.

46. The system of claim 44, wherein the recommendation module is further executable to recommend a portion of the potential content item.

47. The system of claim 44, wherein the recommendation module is further executable to provide a negative recommendation advising the at least the user to avoid the potential content item.

48. The system of claim 44, wherein the interface module is further executable to list one or more potential content items according to the abandonment information.