Media recommendations based on negative feedback

• US 8,387,093 B2
• Filed: 08/23/2002
• Issued: 02/26/2013
• Est. Priority Date: 08/23/2002
• Status: Active Grant

• Current Assignee: Rovi Technologies Corporation (Adeia Inc.)
• Original Assignee: Microsoft Corporation
• Inventors: Danker, Daniel, Barrett, Peter T.
• Primary Examiner(s): Kumar, Pankaj
• Assistant Examiner(s): ANDRAMUNO, FRANKLIN S
• Application Number:
  US10/227,115
• Publication Number:
  US 20040040040A1
• Time in Patent Office: 3,840 Days
• Field of Search: 725/46, 725/47
• US Class Current: 725/46
• CPC Class Codes: H04N 21/44224 Monitoring of user activity on external systems, e.g. Internet browsing
• CPC Class Codes: H04N 21/4532 involving end-user characteristics, e.g. viewer profile, preferences monitoring of user activities for profile generation for accessing a video database G06F16/739; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18
• CPC Class Codes: H04N 21/4668 for recommending content, e.g. movies
• CPC Class Codes: H04N 21/47 End-user applications
• CPC Class Codes: H04N 21/4755 for defining user preferences, e.g. favourite actors or genre retrieval personalisation and generation of user profiles for the retrieval of video data G06F16/739; user profiles in network data switching protocols H04L67/306
• CPC Class Codes: H04N 21/4756 for rating content, e.g. scoring a recommended movie
• CPC Class Codes: H04N 21/482 End-user interface for program selection broadcast systems using EPGs H04H60/72
• CPC Class Codes: H04N 7/163 by receiver means only

Abstract

A media recommendation system gathers negative user feedback associated with media content. Based on the gathered feedback, the media recommendation system extracts a list of recommended media content from program data describing available media content.

Claims

1. A method for recommending media content comprising:

tuning a multimedia tuning device to a first channel that is providing a first media content;

identifying attributes of the first media content;

receiving a user-submitted command that results in the first media content no longer being presented;

based on the user-submitted command, making an inference of user disinterest in media content having at least one attribute in common with the first media content;

storing data representing user disinterest in association with at least one attribute of the first media content;

refining the data representing user disinterest over time as further inferences of user disinterest are made; and

filtering a list of recommended media content to not include media content having attributes with which user disinterest has been inferred.

16. The method as recited in claim 1, wherein the first media content comprises segments, wherein identifying attributes of the first media content comprises identifying a particular segment of the first media content, and wherein storing data representing an indication of the disinterest of the user comprises storing data representing an indication of the disinterest of the user in the particular segment of the first media content.

17. The method as recited in claim 1, the method further comprising:

before presenting the display comprising the plurality of available media content, associating the data representing an indication of the disinterest of the user in the first media content with each of the data associated with the at least one attribute of the first media content; and

presenting, based on a plurality of the at least one attribute of the first media content, the display comprising the plurality of available media content, wherein the at least one attribute is extracted from closed caption data that is broadcast with the first media content.

18. A method as recited in claim 1, wherein the user-submitted command that results in the first media content no longer being presented comprises a channel change command.

19. A method as recited in claim 1, wherein the user-submitted command that results in the first media content no longer being presented comprises a power off command.

2. A method for gathering negative user feedback comprising:

tuning a multimedia tuner to a particular media content through a media channel;

identifying the particular media content being delivered;

receiving a user submitted power off command;

inferring user disinterest in the particular media content based on the user submitted power off command;

associating an indication of negative feedback with one or more attributes of the particular media content, based on the inferred user disinterest;

generating a list of currently available media content to be recommended to the user based on a media content recommendation algorithm, wherein the list of currently available media content to be recommended to the user comprises the particular media content and at least one other item of media content;

generating a filtered list of media content by filtering the list of available media content to be recommended to the user such that media content having attributes with an associated indication of negative feedback is removed from the list of available media content to be recommended to the user, whereby the filtered list does not include the particular media content; and

generating media recommendations based on the filtered list.

3. The method as recited in claim 2 wherein the media channel comprises one of a broadcast channel, a television channel, a radio channel, and a virtual channel associated with stored media content.

4. The method as recited in claim 2 wherein the particular media content comprises one of a broadcast television program, a video-on-demand program, a previously recorded program, a song, and a news program segment.

5. The method as recited in claim 2 wherein the at least one attribute of the particular media content is selected from a plurality of attributes comprising:

a rating;

a release year;

a program description;

closed caption data;

a title;

a rerun indicator;

a genre;

an actor;

an artist;

an album;

a broadcast time; and

a viewing time.

15. The method as recited in claim 2 wherein the negative feedback is the initial user feedback associated with the particular media content.

20. The method as recited in claim 2, further comprising:

making subsequent inferences of user disinterest based on subsequent user-submitted commands to change channels or power off;

refining indications of negative feedback associated with media content attributes based on the subsequent inferences.

6. A method for serving media content recommendations to a client, the method comprising:

tuning to a first channel providing a first media content;

receiving an indication of a user request to tune to a different channel providing a different media content, wherein negative feedback is inferred from the user request to tune to the different channel providing the different media content;

associating an indication of negative user feedback with the first media content based on the user request to tune to the different channel;

filtering a list of available media content to remove from the list, media content associated with the negative user feedback; and

serving the list of media content that is filtered to a client device for presentation to a user.

7. A method for generating media content recommendations, the method comprising:

receiving from a user, a selection to tune away from a particular media content, wherein negative feedback is inferred from the user request to tune away from the particular media content;

in response to receiving the selection to tune away from the particular media content, associating an indication of the negative feedback with at least one attribute of the particular media content; and

using the indication of the negative feedback to filter a list of available media content such that media content in which the user is not interested is removed from the list.

8. The method as recited in claim 7 wherein the media content comprises at least one of a broadcast television program, a purchased video on demand, a movie, a song, and a television program segment.

9. The method as recited in claim 7 wherein the at least one attribute of the media content is selected from a plurality of attributes comprising:

a channel;

a rating;

a release year;

a program description;

closed caption data;

a title;

a rerun indicator;

a genre;

an actor;

an artist;

an album;

a broadcast time;

a viewing time; and

a listening time.

10. A computer system comprising:

a media distribution interface coupled to a data processor, the media distribution interface being configured to;

receive media content from a media distribution system;

tune to a first channel providing a first media content;

receive an indication of a user request to tune to a different channel providing a different media content, wherein negative user feedback associated with at least one attribute of the first media content is inferred from the user request to tune to the different channel providing the different media content;

transmit a representation of the negative user feedback associated with the at least one attribute of the first media content to the media distribution system; and

receive media recommendations from the media distribution system; and

the data processor configured to gather the negative user feedback inferred from the user request to tune to the different channel providing the different media content.

11. The computer system as recited in claim 10, wherein the data processor is further configured to:

generate media content recommendations based on the negative user feedback that is inferred such that media content that is not of interest to a user is not recommended.

12. A computer system as recited in claim 10, implemented as a television set-top box.

13. One or more memory storage devices comprising computer executable instructions that, when executed, direct a computing system to perform operations comprising:

identifying a particular media content on a currently tuned channel;

detecting a user-submitted channel change command, wherein negative feedback is inferred from the user-submitted channel change command;

in response to detecting the user-submitted channel change command, associating an indicator of negative user feedback with a media content attribute that is associated with the particular media content; and

generating media content recommendations based, at least in part, on the indicator of negative user feedback.

14. One or more memory storage devices comprising computer executable instructions that, when executed, direct a computing system to perform operations comprising:

tuning to a first channel providing a first media content;

receiving an indication of a user request to tune to a different channel providing a different media content, wherein negative feedback is inferred from the user request to tune to the different channel providing a different media content;

associating the negative feedback with the first media content, based on the user request to tune to the different channel;

generating media content recommendations based, at least in part, on the negative feedback; and

serving the media content recommendations to one or more client devices for presentation to a user.

21. A method comprising:

tuning to a particular channel that is currently broadcasting a particular program;

receiving a user-submitted command to tune away from the particular channel while the particular program is being broadcast;

inferring, from the user-submitted command to tune away from the particular channel while the particular program is being broadcast, a user disinterest in at least one attribute associated with the particular program, the at least one attribute selected from a group of attributes including;

a television series of which the particular program is a single episode;

a particular episode of a television series;

a broadcast network;

a re-run indicator;

a broadcast time;

a broadcast day; and

a programming content;

analyzing other viewing data to select the at least one attribute associated with the particular program with which to associate the user disinterest, the other viewing data including at least one of;

indicators of preferred programs;

user viewing history;

or

other inferred indicators of user disinterest;

associating an indication of user disinterest with the at least one attribute that is selected;

selecting a list of programs to recommend based on indicators of preferred programs; and

filtering the list of programs to recommend to remove programs having an attribute with which user disinterest is associated.