Recommendations based on item tagging activities of users

• US 8,577,880 B1
• Filed: 02/21/2012
• Issued: 11/05/2013
• Est. Priority Date: 11/17/2005
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Donsbach, Aaron, Gregov, Andrej, Few, Jeffrey D., Lehman, Jeffrey D., Brownell, Jeffrey T.
• Primary Examiner(s): Richardson, James E
• Application Number:
  US13/401,480
• Time in Patent Office: 623 Days
• Field of Search: 707732-734, 707/765, 707/766, 707/948, 705/26.7, 705/27.1
• US Class Current: 707/732
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 30/0641 Shopping interfaces
• CPC Class Codes: Y10S 707/948 Product or catalog

Abstract

A system provides a user interface through which users can flexibly tag individual items represented in an electronic catalog with user-defined tags, such as text strings, and obtain recommendations that are specific to particular tags. The tags and tag-item assignments created by each user are stored persistently in association with the user, and may be kept private to the user or exposed to others. Once a user has assigned a tag to a number of items, the user (or another user in some embodiments) can request and obtain recommendations that are specific to this tag. These recommendations may be generated in real time by a recommendation service that identifies items that are collectively similar or related to the items associated with the tag.

Claims

1. A system for recommending items to users, the system comprising:

a tagging system that provides functionality for a user to;

create tags via entry of text strings into tag entry fields of item pages of an electronic data repository, and

associate the tags with selected items of the electronic data repository, wherein the tagging system is configured to store user-specific tag data, the user-specific tag data reflecting associations created by the user between the tags and the selected items;

the tagging system further configured to output for presentation to the user one or more of the tags created by the user; and

a recommendation system comprising computer hardware, the recommendation system configured to;

receive a user selection of a selected tag of the one or more tags output for presentation to the user, and

in response to receiving the user selection and without requiring user interaction;

identify at least one of the selected items previously tagged with the selected tag,

programmatically access item relationship data to identify additional items that are related to the at least one selected item previously tagged by the user, and

select one or more of the additional items as recommendations for the user.

2. The system of claim 1, wherein the additional items are related to the at least one selected item based at least in part on item-to-item similarity mappings.

3. The system of claim 2, wherein the item-to-item similarity mappings are based, at least in part, on tag-based item similarities detected by analyzing aggregated tag data of a population of users.

4. The system of claim 1, wherein the item pages each comprise a picture of one of the items.

5. The system of claim 1, wherein the recommendation system is further configured to generate the recommendations based at least in part on a second tag created by a second user.

6. The system of claim 1, wherein the selected tag is private to the user.

7. A method of recommending items to users, the method comprising:

providing a user interface that provides functionality for a user to create tags via entry of text strings into tag entry fields of item pages of an electronic data repository, and associate the tags with selected items of the electronic data repository, wherein the tagging system is configured to store user-specific tag data, the user-specific tag data reflecting associations created by the user between the tags and the selected items;

outputting for presentation to the user one or more of the tags created by the user;

receiving a user selection of a selected tag of the one or more tags output for presentation to the user; and

in response to receiving the user selection and without requiring user interaction;

identifying at least one of the selected items previously tagged with the selected tag;

programmatically accessing item relationship data to identify additional items that are related to the at least one selected item previously tagged by the user; and

selecting one or more of the additional items as recommendations for the user;

wherein at least said programmatically accessing the item relationship data is performed by a computer system comprising computer hardware.

8. The method of claim 7, wherein the additional items are related to the at least one selected item based at least in part on item-to-item similarity mappings.

9. The method of claim 8, wherein the item-to-item similarity mappings are based, at least in part, on tag-based item similarities detected by analyzing aggregated tag data of a population of users.

10. The method of claim 8, wherein the item pages each comprise a picture of one of the items.

11. The method of claim 8, further comprising generating the recommendations based at least in part on a second tag created by a second user.

12. The method of claim 8, wherein the selected tag is private to the user.

13. Non-transitory computer storage comprising instructions stored thereon that, when executed in one or more processors, are configured to implement a system for recommending items to users, the system comprising:

a tagging system that provides functionality for a user to;

create tags via entry of text strings into tag entry fields of item pages of an electronic data repository, and

associate the tags with selected items of the electronic data repository, wherein the tagging system is configured to store user-specific tag data, the user-specific tag data reflecting associations created by the user between the tags and the selected items;

the tagging system further configured to output for presentation to the user one or more of the tags created by the user; and

a recommendation system configured, to;

receive a user selection of a selected tag of the one or more tags output for presentation to the user, and

in response to receiving the user selection and without requiring user interaction;

identify at least one of the selected items previously tagged with the selected tag,

programmatically access item relationship data to identify additional items that are related to the at least one selected item previously tagged by the user, and

select one or more of the additional items as recommendations for the user.

14. The non-transitory computer storage of claim 13, wherein the additional items are related to the at least one selected item based at least in part on item-to-item similarity mappings.

15. The non-transitory computer storage of claim 14, wherein the item-to-item similarity mappings are based, at least in part, on tag-based item similarities detected by analyzing aggregated tag data of a population of users.

16. The non-transitory computer storage of claim 13, wherein the item pages each comprise a picture of one of the items.

17. The non-transitory computer storage of claim 13, wherein the recommendation system is further configured to generate the recommendations based at least in part on a second tag created by a second user.

18. The non-transitory computer storage of claim 13, wherein the selected tag is private to the user.