ADAPTIVE RECOMMENDATION SYSTEM

• US 20130066819A1
• Filed: 09/09/2011
• Published: 03/14/2013
• Est. Priority Date: 09/09/2011
• Status: Active Grant

• Current Assignee: Rovi Technologies Corporation (Adeia Inc.)
• Original Assignee: Microsoft Corporation
• Inventors: Nice, Nir, Kremer, Dror, Sitton, Daniel, Feldman, Michael, Shlevich, Shimon, Folger, Ori
• Granted Patent: US 9,208,155 B2
• Time in Patent Office: Days
• US Class Current: 706/14
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0631 Item recommendations

Abstract

A recommendation system for optimizing content recommendation lists is disclosed. The system dynamically tracks a list interaction history of a user, which details that user'"'"'s interactions with a plurality of different lists presenting different recommended items to that user. The system automatically correlates one or more list preferences with that user based on the list interaction history, and builds a recommendation list with a plurality of candidate items having different recommendation confidences. The recommendation list is built such that each candidate item with a higher recommendation confidence is prioritized over each candidate item with a lower recommendation confidence according to the one or more list preferences correlated to that user.

Claims

1. A recommendation system, comprising:

a logic subsystem; and

a data-holding subsystem holding instructions executable by the logic subsystem to;

dynamically track a list interaction history of a user, the list interaction history detailing that user'"'"'s interactions with a plurality of different lists presenting different recommended items to that user;

automatically correlate one or more list preferences with that user based on the list interaction history; and

build a recommendation list with a plurality of candidate items having different recommendation confidences such that each candidate item with a higher recommendation confidence is prioritized over each candidate item with a lower recommendation confidence according to the one or more list preferences correlated to that user.

2. The system of claim 1, wherein the instructions are further executable to correlate the list preferences to the user by tracking list attributes of recommended items with which the user interacts.

3. The system of claim 2, wherein the instructions are further executable to track list attributes of recommended items with which the user interacts by tracking one or more of user clicks on the recommended items, user hovers over the recommended items, and user purchases of the recommended items.

4. The system of claim 1, wherein the recommendation confidences are determined by a collaborative and/or features filter.

5. The system of claim 1, wherein the list preferences are independent of content attributes.

6. The system of claim 1, wherein the list preferences include content attributes.

7. A recommendation system, comprising:

a logic subsystem; and

a data-holding subsystem holding instructions executable by the logic subsystem to;

dynamically track a list interaction history of a user, the list interaction history detailing list attributes of recommended items with which that user interacts, the list attributes being independent of content attributes of such recommended items; and

build a recommendation list with a plurality of candidate items having different recommendation confidences such that a candidate item having a relatively higher recommendation confidence than another candidate item is given list attributes towards which the interaction history indicates that user has previously demonstrated relatively higher interaction.

8. The system of claim 7, wherein the list attributes for each recommended item comprise one or more of a size of that item on a user interface, a color of that item on the user interface, and a font of that item on the user interface.

9. The system of claim 7, wherein the list attributes for each recommended item comprise a number of columns of a list in which that item is presented.

10. The system of claim 7, wherein the list attributes for each recommended item comprise a number of subheadings on a list in which that item is presented.

11. The system of claim 7, wherein the list attributes for each recommended item comprise a position of that item on a user interface.

12. The system of claim 7, wherein the list attributes for each recommended item comprise whether that item includes a picture.

13. The system of claim 7, wherein the list attributes for each recommended item comprise whether that item includes animation.

14. The system of claim 7, wherein the list attributes for each recommended item comprise an amount of textual information displayed on the user interface along with that item.

15. The system of claim 7, wherein the list attributes for each recommended item comprise a length of a list in which that item is presented.

16. The system of claim 7, wherein the list attributes for each recommended item comprise a relative position of that item within a list in which that item is presented.

17. The system of claim 7, wherein the list attributes for each recommended item comprise whether that item is from a “

recommended for you”

subgroup, from a “

social”

subgroup, or from a “

popular”

subgroup.

18. The system of claim 7, wherein the recommendation confidences are determined by a collaborative filter.

19. The system of claim 7, wherein the recommendation confidences are determined by a features filter.

20. A recommendation system, comprising:

a logic subsystem; and

a data-holding subsystem holding instructions executable by the logic subsystem to;

dynamically track a list interaction history of a user, the list interaction history detailing content-independent list attributes of recommended items with which that user interacts, the list attributes for each recommended item comprising a position of that item on the user interface; and

build a recommendation list with a plurality of candidate items having different recommendation confidences such that a candidate item having a relatively higher recommendation confidence than another candidate item is given a position on the user interface towards which the interaction history indicates that user has previously demonstrated relatively higher interaction, the relatively higher interaction demonstrated by one or more of user clicks on a recommended item having that position on the user interface, user hovers over a recommended item having that position on the user interface, and user purchases of a recommended item having that position on the user interface.