Automated discovery of items associated with a context based on monitored actions of users

• US 7,953,739 B2
• Filed: 07/26/2010
• Issued: 05/31/2011
• Est. Priority Date: 10/30/2001
• Status: Active Grant

• Current Assignee: A9.com Incorporated (Amazon.com, Inc.)
• Original Assignee: A9.com Incorporated (Amazon.com, Inc.)
• Inventors: York, Jeremy C., Smith, Brent, Cottrell, Robert
• Primary Examiner(s): Ali; Mohammad
• Assistant Examiner(s): Shanmugasundaram; Kannan
• Application Number:
  US12/843,705
• Publication Number:
  US 20100293474A1
• Time in Patent Office: 309 Days
• Field of Search: 707/749, 707999001-999005
• US Class Current: 707/749
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• 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: Y10S 707/948 Product or catalog
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

Computer processes are disclosed for automatically refining the order and/or content of context-specific lists of items (e.g., products) based on monitored actions of users. One such process involves outputting context-specific lists of items to users together with selectable controls for initiating displays of similar items. When users opt to view similar items, the users'"'"' actions performed in connection with the similar items are recorded. If the resulting level of user activity recorded for a particular similar item is sufficiently high, the item is automatically added to the associated context-specific list.

Claims

1. A computer-implemented method of discovering items associated with particular contexts, the comprising:

outputting, for presentation to each of a plurality of users, a context-specific list of items in which each of a plurality of items on the list is presented together with a respective control for initiating a display of similar items, said context-specific list corresponding to a particular context, said particular context being one of a plurality of defined contexts in which items are capable of being displayed;

recording, in association with the particular context, user activity associated with user selections of the controls for initiating displays of similar items, said user activity including user selections of particular similar items, wherein recording said user activity comprises recording an indication that said user selections of similar items were made in association with said user selections of the controls for initiating displays of similar items; and

determining whether the recorded user activity associated with a similar item is sufficient to add the similar item to the context-specific list, said determining comprising calculating a weight value for the similar item, and using the weight value to determine whether to add the similar item to the context-specific list, said weight value revealing a degree to which the similar item is associated with the particular context;

said method performed by one or more computers of a multi-user interactive system.

2. The method of claim 1, wherein the weight value is dependent upon (1) the recorded user activity associated with the similar item, and (2) a number of times the similar item has been exposed in association with the particular context.

3. The method of claim 1, further comprising recording exposure events in which the similar item is exposed for viewing as a result of user selections of said controls, and taking the recorded exposure events into consideration in determining whether the recorded user activity is sufficient to add the similar item to the context-specific list.

4. The method of claim 1, wherein outputting the context-specific list comprises outputting the list together with a textual identifier of said particular context.

5. The method of claim 1, wherein the items are products represented in an electronic catalog, and the particular context is a particular gift-giving context that is selectable by users.

6. The method of claim 1, further comprising recording, in association with the particular context, user activity data for each of the items on the context-specific list, and using said user activity data for each of the items to determine whether to remove any of the items from the context-specific list.

7. The method of claim 1, further comprising recording, in association with the particular context, user activity data for each of the items on the context-specific list, and using said user activity data for each of the items to rank the items for presentation on said list.

8. The method of claim 1, wherein determining whether the recorded user activity is sufficient comprises giving different amounts of weight to different types of user actions performed in connection with the similar item.

9. The method of claim 1, further comprising responding to user selection of said control, as displayed together with a first item on said list, by outputting a list of items that are purchased by users who purchase the first item, as determined from recorded purchase histories of users.

10. The method of claim 1, further comprising responding to user selection of said control, as displayed together with a first item on said list, by outputting a list of items that are viewed by users who view the first item, as determined from recorded item viewing activities of users.

11. A computerized system capable of discovering items associated with particular contexts, the system comprising:

a server system that is responsive to page requests received over a network from user computing devices by outputting pages that include a context-specific list of items, and that include display elements that are selectable by users to initiate displays of corresponding similar items, said context-specific list corresponding to a particular context among a plurality of defined contexts in which items are capable of being displayed, said server system comprising a processor;

an activity tracking system that maintains, in association with the particular context, user activity data reflective of user selections of particular similar items from said displays of similar items, said user activity data stored with an indication that said user selections of the similar items were made in association with user selections of said display elements; and

a list refinement system that uses the user activity data to identify particular similar items to add to the context-specific list, said list refinement system comprising an item weighting service that generates item-specific weights for particular similar items, wherein an item-specific weight for a similar item reveals a degree to which the similar item is behaviorally associated with a particular context, wherein the list refinement system uses the item-specific weights to select particular similar items to add to the context-specific list.

12. The system of claim 11, further comprising an exposure tracking system that tracks exposure events in which particular similar items are exposed as a result of user selections of said display elements, wherein the list refinement system uses the tracked exposure events, in combination with said user activity data, to select particular similar items to add to the context-specific list.

13. The system of claim 11, wherein the activity tracking system maintains context-specific user activity data for each of the plurality of defined contexts, and the list refinement system uses the context-specific user activity data to refine context-specific lists corresponding to each of said contexts.

14. The system of claim 11, wherein the items are products represented in an electronic catalog, and the particular context is a particular gift-giving context that is selectable by users.

15. The system of claim 11, wherein the activity tracking system tracks a plurality of different types of item selection actions performed by the users, and the list refinement system gives different amounts of weight to different types of item selection actions.

16. The system of claim 11, wherein the server system generates said displays of corresponding similar items based at least partly on purchase-based relationships between items.

17. The system of claim 11, wherein the server system generates said displays of corresponding similar items based at least partly on item-viewing-based relationships between items.

18. The system of claim 11, wherein the list refinement system additionally removes items from the context-specific list based on user activity data maintained by the activity tracking system.

19. The system of claim 11, wherein the list refinement system. additionally determines a display order for displaying the items on the context-specific list, said display order being dependent upon user activity data maintained by the activity tracking system.

20. Physical, non-transitory computer storage that stores executable code modules that instruct one or more computers to perform a method that comprises:

outputting, for presentation to each of a plurality of users, a context-specific list of items in which each of a plurality of items on the list is presented together with a respective display element for initiating a display of similar items, said context-specific list corresponding to a particular context, said particular context being one of a plurality of defined contexts in which items are capable of being displayed;

recording, in association with the particular context, user activity associated with user selections of the display elements, said user activity including user selections of particular similar items, said user activity recorded with an indication that said user selections of similar items were made in association with said user selections of the display elements for initiating displays of similar items; and

determining whether the recorded user activity associated with a similar item reveals a sufficient level of association between the particular context and the similar item to add the similar item to the context-specific list, said determining comprising calculating a value representing a strength of an association between the similar item and the particular context, and using the value to determine whether to add the similar item to the context-specific list.

21. The physical computer storage of claim 20, wherein the method further comprises recording, in association with the particular context, user activity data for each of the items on the context-specific list, and using said user activity data for each of the items to determine whether to remove any of the items from the context-specific list.

22. The physical computer storage of claim 20, wherein the method further comprises recording, in association with the particular context, user activity data for each of the items on the context-specific list, and using said user activity data for each of the items to rank the items for presentation on said list.

23. The physical computer storage of claim 20, wherein the value is dependent upon (1) the recorded user activity associated with the similar item, and (2) context-specific exposure data reflective of how many times the similar item has been exposed to users in association with the particular context.

24. The physical computer storage of claim 20, in combination with a computer system that comprises said one or more computers, the computer system programmed with the executable code modules to perform said method.