Data mining of user activity data to identify related items in an electronic catalog

• US 20050071251A1
• Filed: 11/19/2004
• Published: 03/31/2005
• Est. Priority Date: 09/18/1998
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Brent R. Smith, Gregory D. Linden, Nida K. Zada
• Inventors: Smith, Brent R., Linden, Gregory D., Zada, Nida K.
• Granted Patent: US 7,685,074 B2
• Time in Patent Office: Days
• US Class Current: 705/26
• 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 methods are disclosed for monitoring user browsing activities that indicate user interests in particular products, or other items, represented in an electronic catalog, and for using such information to identify items that are related to one another. In one embodiment, relationships between items within an electronic catalog are determined by identifying items that are frequently viewed by users within the same browsing session (e.g., items A and B are related because a significant portion of those who viewed A also viewed B). The resulting item relatedness data may be stored in a table that maps items to sets of related items. The table may be used to provide personalized item recommendations to users, and/or to supplement item detail pages of the electronic catalog with lists of related items. In one embodiment, the table is used to provide session-specific item recommendations to users.

Claims

1. A data mining method, comprising:

programmatically generating, for each of a plurality of pairs of items represented in a browsable electronic catalog, a respective data value representing a degree to which the items of the pair are related to one another, said respective data value being based, at least in-part, on a frequency with which items of the pair both each have been selected for viewing within a common user browsing session; and

generating in computer storage a data structure that associates items with related items, wherein the data structure is generated based, at least in-part, on the data values.

2. The method of claim 1, wherein the data value for a given item pair is based, at least in-part, on a frequency with which item detail pages of both items of the given item pair have been accessed within a common browsing session.

3. The method of claim 1, wherein the method is performed without requiring users to explicitly indicate preferences for specific items.

4. The method of claim 1, further comprising using the data structure to generate personalized item recommendations for each of a plurality of target users.

5. The method of claim 4, wherein the personalized item recommendations include session-specific item recommendations that are based, at least in-part, on a plurality of item detail pages viewed by a target user during a current browsing session.

6. The method of claim 1, further comprising accessing a session record to identifying a plurality of items viewed by a user during a current browsing session;

using the data structure to select a set of additional items that are collectively related to the plurality of items viewed by the user, and recommending the additional items to the user during the current browsing session.

7. The method of claim 1, further comprising using the data structure to supplement item detail pages of the electronic catalog with corresponding lists of related items.

8. The method of claim 1, wherein the items include video titles.

9. A computer-readable medium having stored thereon an executable computer program that embodies the method of claim 1.

10. A computer storage medium having stored therein a data structure generated according to the method of claim 1.

11. A data mining method, comprising:

maintaining, in computer storage, item viewing histories of a plurality of users of an electronic catalog of items, said item viewing histories identifying catalog items selected for viewing by corresponding users;

programmatically analyzing the item viewing histories to generate data values reflective of item co-occurrences within the item viewing histories; and

generating a data structure that associates specific catalog items with corresponding sets of related catalog items, wherein the data structure is generated based, at least in-part, on said data values.

12. The method of claim 11, wherein each item viewing history is specific to a respective browsing session.

13. The method of claim 11, wherein analyzing the item viewing histories comprises determining frequencies with which specific catalog items have been viewed within a common browsing session.

14. The method of claim 11, wherein the data values are reflective of frequencies with which specific items co-occur within the item viewing histories.

15. The method of claim 11 wherein analyzing the item viewing histories comprises, for each of a plurality of item pairs, each pair comprising a first catalog item and a second catalog item, determining a number of item viewing histories in which both the first catalog item and the second catalog item occur.

16. The method of claim 11, wherein analyzing the item viewing histories comprises, for each of a plurality of item pairs, each pair comprising a first catalog item and a second catalog item, determining a number of browsing sessions in which the first and second catalog items both each were selected for viewing.

17. The method of claim 11, wherein analyzing the item viewing histories comprises determining, for an item pair comprising a first catalog item and a second catalog item, a number of browsing sessions in which item detail pages of the first and the second catalog items both each were accessed.

18. The method of claim 11, wherein at least some of the item viewing histories respectively reflect item viewing events over multiple sessions of the same user.

19. The method of claim 11, wherein generating the data structure comprises analyzing the data values to select pairs of catalog items that are sufficiently related to be stored in association with each other in the data structure.

20. The method of claim 19, wherein generating the data structure further comprises storing the data values associated with the selected pairs of catalog items in the data structure.

21. The method of claim 11, further comprising using the data structure to provide personalized item recommendations to each of a plurality of target users.

22. The method of claim 11, further comprising using the data structure, in combination with a record of a plurality of catalog items viewed by a target user during a current browsing session, to generate personalized, session-specific item recommendations for the target user.

23. The method of claim 11, further comprising using the data structure to supplement an item detail page of the electronic catalog with a list of related catalog items.

24. A supplemented item detail page generated according to the method of claim 23 represented in computer storage, said item detail page including said list of related catalog items.

25. A computer-readable medium having stored thereon an executable computer program that embodies the method of claim 11.

26. A computer memory having stored therein a data structure generated according to the method of claim 11.

27. A data mining method, comprising:

maintaining, in computer storage, session records of a plurality of users of an electronic catalog of items, said session records identifying catalog items selected by users within corresponding user sessions;

programmatically analyzing the session records to generate data values reflective of item co-occurrences within the session records; and

generating a data structure that associates catalog items with corresponding sets of related catalog items, wherein the data structure is generated based, at least in-part, on said data values.

28. The method of claim 27, wherein analyzing the session records comprises determining frequencies with which particular catalog items are viewed within a common session.

29. The method of claim 27, wherein the data values are reflective of frequencies with which particular items co-occur within the session records.

30. The method of claim 27, wherein analyzing the session records comprises, for each of a plurality of item pairs of a first catalog item and a second catalog item, determining a number of sessions in which the first catalog item and the second catalog both each were selected.

31. The method of claim 27, wherein analyzing the session records comprises determining, for an item pair consisting of a first catalog item and a second catalog item, a number of sessions in which item detail pages of the first and the second catalog items both each were accessed.

32. The method of claim 27, wherein the catalog items are products represented in the electronic catalog.

33. The method of claim 27, wherein generating the data structure comprises analyzing the data values to select pairs of catalog items that are sufficiently related to be stored in association with each other in the data structure.

34. The method of claim 33, wherein generating the data structure further comprises storing the data values associated with the selected pairs of catalog items in the data structure.

35. The method of claim 27, further comprising using the data structure to provide personalized item recommendations to each of a plurality of target users.

36. The method of claim 27, further comprising using the data structure, in combination with a record of a plurality of catalog items viewed by a target user during a current session, to generate personalized, session-specific item recommendations for the target user.

37. The method of claim 27, further comprising using the data structure to supplement item detail pages of the electronic catalog with lists of related catalog items.

38. A computer-readable medium having stored thereon an executable computer program that embodies the method of claim 27.

39. A computer memory having stored therein a data structure generated according to the method of claim 27.

40. A data mining method, comprising:

maintaining, in computer storage, histories of item-affinity-evidencing actions performed by users during interaction with an electronic catalog of items;

programmatically analyzing the histories of item-affinity-evidencing actions, in an off-line processing mode, to measure degrees to which catalog items are related to one another; and

generating a mapping of catalog items to related catalog items, said mapping including data values reflective of said degrees to which catalog items are related.

41. The method of claim 40, wherein the item-affinity-evidencing actions analyzed to measure item relatedness include user selections of catalog items for viewing.

42. The method of claim 41, further comprising using the mapping, in combination with identifiers of a plurality of catalog items selected for viewing by a target user during a current browsing session, to select a set of additional catalog items to recommend to the target user during the current browsing session.

43. The method of claim 40, wherein the item-affinity-evidencing actions analyzed to measure item relatedness include item purchases.

44. The method of claim 40, wherein the item-affinity-evidencing actions analyzed to measure item relatedness include selection of a catalog item to add to an electronic shopping cart.

45. The method of claim 40, wherein the item-affinity-evidencing actions analyzed to measure item relatedness include selection of a catalog item to add to an electronic wish list.

46. The method of claim 40, wherein the item-affinity-evidencing actions analyzed to measure item relatedness include user selections of items to obtain from an online business entity.

47. The method of claim 40, wherein the item-affinity-evidencing actions analyzed to measure item relatedness include placement of a bid on a catalog item.

48. The method of claim 40, wherein the method is performed without requiring the users to explicitly rate any catalog items, and without requiring users to explicitly indicate preferences for any catalog items.

49. The method of claim 40, wherein programmatically analyzing the histories of item-affinity-evidencing actions comprises measuring frequencies with which particular catalog items have been viewed within a common browsing session.

50. The method of claim 40, wherein programmatically analyzing the histories of item-affinity-evidencing actions comprises measuring frequencies with which particular item detail pages of the electronic catalog have been viewed within a common browsing session.

51. The method of claim 40, wherein programmatically analyzing the histories of item-affinity-evidencing actions comprises, for each of a plurality of item pairs of a first catalog item and a second catalog item, determining a number of browsing sessions in which the first catalog item and the second catalog item both each were selected by a user.

52. The method of claim 40, wherein generating the mapping comprises selecting catalog items that are sufficiently related to be stored in association with each other in said mapping.

53. The method of claim 40, further comprising using the mapping, including the data values recorded therein, to provide personalized item recommendations to each of a plurality of users.

54. The method of claim 40, wherein the catalog items are products that are available for purchase from the electronic catalog.

55. A computer-readable medium having stored thereon an executable computer program which, when executed by a computer system, causes the computer system to perform the method of claim 40.

56. A computer storage medium having stored therein a mapping generated according to the method of claim 40.