Content recommendations using deep data

• US 9,374,411 B1
• Filed: 03/21/2013
• Issued: 06/21/2016
• Est. Priority Date: 03/21/2013
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Goetz, Jeromey Russell
• Primary Examiner(s): Moise, Emmanuel L
• Assistant Examiner(s): Chang, Julian
• Application Number:
  US13/848,305
• Time in Patent Office: 1,188 Days
• Field of Search: 709/231, 725 14- 21, 725 44- 47, 705 1449- 1454, 705/26.7
• US Class Current: 1/1
• CPC Class Codes: G06F 16/447 Temporal browsing, e.g. timeline
• 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: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/125 involving control of end-device applications over a network
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

Disclosed are various embodiments for recommending content for users. A user is associated with a user preferred character that is represented in a first network content. A library is referenced to identify a recommended character based at least upon the user preferred character, the recommended character being represented in a second network content, the library associating the preferred character to the recommended character according to a relationship score. A recommendation is sent for presentation of the second network content.

Claims

1. A non-transitory computer-readable medium having a plurality of computer instructions executable by at least one computing device, wherein, upon execution, the plurality of computer instructions cause the at least one computing device to:

stream a first video content feature from at least one server to a client associated with a user;

cause the first video content feature to be rendered in a user interface for display on the client;

obtain first user interaction data from the client based at least in part upon a first interaction with the first video content feature;

cause a timeline user interface that segments the first video content feature into a plurality of scenes to be rendered in response to obtaining the first user interaction data;

obtain second user interaction data from the client based at least in part upon a second interaction with at least one character feedback interface rendered on the timeline user interface, the at least one character feedback interface being associated with a current scene included in the plurality of scenes;

associate a preferred character with the user based at least upon the second user interaction data, the preferred character being represented by a corresponding first cast member of the first video content feature;

reference a library to identify a recommended character based at least upon the preferred character, the recommended character being represented by a corresponding second cast member of a second video content feature, the library associating the preferred character to the recommended character according to a relationship score, the relationship score being calculated based at least in part upon a proportion of users who have expressed a preference towards both the preferred character and the recommended character; and

send to the user a recommendation to play the second video content feature.

2. The non-transitory computer-readable medium of claim 1, wherein the plurality of computer instructions further cause the at least one computing device to:

obtain additional user interaction data from the user based at least upon an interaction with the second video content feature; and

update the relationship score based at least upon the additional user interaction data.

3. The non-transitory computer-readable medium of claim 2, wherein the additional user interaction data comprises user feedback expressing that the user favors at least one scene associated with the second video content feature.

4. The non-transitory computer-readable medium of claim 1, wherein the plurality of computer instructions further cause the at least one computing device to:

identify an attribute associated with the recommended character; and

send the attribute to the client to indicate a basis for the recommendation to the user.

5. The non-transitory computer-readable medium of claim 1, wherein the user interaction data is obtained during a playback of the first video content feature via the client.

6. A system, comprising:

at least one computing device; and

at least one application executable in the at least one computing device, the at least one application causing the at least one computing device to;

cause a first video content feature to be rendered in a user interface for display on a client;

determine that a user prefers a first character based at least upon a first interaction data, the first character being represented by a corresponding first cast member of the first video content feature;

cause a timeline user interface that segments the first video content feature into a plurality of scenes to be rendered in response to obtaining the first interaction data;

obtain second user interaction data from the client based at least in part upon an interaction with at least one character feedback interface rendered on the timeline user interface, the at least one character feedback interface being associated with a current scene included in the plurality of scenes;

determine that the user prefers a second character based at least upon the second interaction data, the second character being represented by a corresponding second cast member of a second video content feature;

determine a relationship score based at least upon the first interaction data and the second interaction data, the relationship score quantifying a degree of commonality between the first character and the second character and being calculated based at least in part upon a proportion of users who have expressed a preference towards both the first character and the second character; and

generate a recommendation to play at least one of the first video content feature or the second video content feature based at least upon the relationship score.

7. The system of claim 6, wherein the at least one application further causes the at least one computing device to:

store a first set of attributes associated with the first character;

store a second set of attributes associated with the second character; and

compare the first set of attributes to the second set of attributes.

8. The system of claim 6, wherein the first interaction data comprises explicit user feedback expressing that the user favors the first character.

9. The system of claim 6, wherein the first interaction data comprises at least one of consumption history data relating to a product associated with the first character or browse history data relating to a product associated with the first character.

10. The system of claim 6, wherein the first interaction data comprises user feedback expressing that the user favors at least one scene associated with the first video content feature, and wherein the at least one application further causes the at least one computing device to identify the first character in the at least one scene.

11. The system of claim 6, wherein the at least one application further causes the at least one computing device to generate the recommendation for another user in response to the another user specifying a preference for at least one of the first character or the second character.

12. The system of claim 11, wherein the at least one application further causes the at least one computing device to:

obtain additional user interaction data from the another user, the additional user interaction data being associated with a playback of at least one of the first video content feature or the second video content feature; and

update the relationship score based at least upon the additional user interaction data.

13. The system of claim 6, wherein the relationship score is further calculated based at least in part upon a quantity of users who have expressed a preference towards both the first character and the second character.

14. A method, comprising:

causing, by at least one computing device, a first network content to be rendered in a user interface for display on a client;

obtaining, by the at least one computing device, first user interaction data from the client based at least in part upon a first interaction with the first network content;

associating, by the at least one computing device, a user with a user preferred character represented in a first network content based at least in part upon the first user interaction data;

causing, by the at least one computing device, a timeline user interface that segments the first network content into a plurality of scenes to be rendered in response to obtaining the first user interaction data;

obtaining, by the at least one computing device, second user interaction data from the client based at least in part upon a second interaction with at least one character feedback interface rendered on the timeline user interface, the at least one character feedback interface being associated with a current scene among the plurality of scenes;

referencing, by the at least one computing device, a library to identify a recommended character based at least upon the user preferred character and the second interaction data, the recommended character being represented in a second network content, the library associating the user preferred character to the recommended character according to a relationship score, the relationship score being calculated based at least in part upon a proportion of users who have expressed a preference towards both the user preferred character and the recommended character; and

sending, by the at least one computing device, a recommendation for presentation of the second network content.

15. The method of claim 14, wherein the user preferred character is a preferred fictional character and wherein the recommended character is a recommended fictional character.

16. The method of claim 14, wherein the second network content comprises a video content feature, an electronic book (e-book), or an audio book.

17. The method of claim 14, wherein associating the user with the user preferred character comprises obtaining explicit user feedback expressing that the user favors the user preferred character.

18. The method of claim 14, further comprising obtaining, by the at least one computing device, user interaction data from the user based at least upon the presentation of the second network content.

19. The method of claim 18, further comprising updating, by the at least one computing device, the relationship score based at least upon the user interaction data.

20. The method of claim 14, further comprising:

storing, by the at least one computing device, a first set of attributes associated with the user preferred character; and

storing, by the at least one computing device, a second set of attributes associated with the recommended character; and

determining, by the at least one computing device, the relationship score based at least upon a comparison between the first set of attributes and the second set of attributes.

21. A non-transitory computer-readable medium having a plurality of computer instructions executable by at least one computing device, wherein upon execution, the plurality of computer instructions cause the at least one computing device to:

cause a first media content to be rendered in a user interface for display on a client;

obtain first interaction data from the client based at least upon a first interaction with the first media content;

determine that a user prefers a first fictional character based at least upon the first interaction data, the first fictional character being associated with the first media content;

cause a timeline user interface that segments the first media content into a plurality of scenes to be rendered in response to obtaining the first interaction data;

obtain second interaction data from the client based at least upon a second interaction with at least one character feedback interface rendered on the timeline user interface, the at least one character feedback interface being associated with a current scene included in the plurality of scenes;

determine that the user prefers a second fictional character based at least upon the second interaction data, the second fictional character being associated with a second media content;

determine a relationship score based at least upon the first interaction data and the second interaction data, the relationship score quantifying a degree of commonality between the first fictional character and the second fictional character and being calculated based at least in part upon a proportion of users who have expressed a preference towards both the first fictional character and the second fictional character; and

generate a recommendation to present at least one of the first media content or the second media content to another user based at least upon the relationship score.

22. The non-transitory computer-readable medium of claim 21, wherein the first media content comprises an electronic book (e-book).

23. The non-transitory computer-readable medium of claim 21, wherein the plurality of computer instructions further causes the at least one computing device to obtain explicit user feedback expressing that the user favors the first fictional character via a user interface.

24. The non-transitory computer-readable medium of claim 21, wherein the relationship score is further calculated based at least in part upon a quantity of users who have expressed a preference towards the first fictional character.

25. The non-transitory computer-readable medium of claim 21, wherein the relationship score is based at least in part upon whether the first fictional character and the second fictional character are associated with a similar genre.