Content recommendation system

• US 8,671,068 B2
• Filed: 09/22/2011
• Issued: 03/11/2014
• Est. Priority Date: 09/22/2011
• Status: Active Grant

• Current Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Original Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Inventors: Harber, Dustin, Oguchi, Kentaro
• Primary Examiner(s): Holmes, Michael B
• Application Number:
  US13/240,802
• Publication Number:
  US 20130080371A1
• Time in Patent Office: 901 Days
• Field of Search: None
• US Class Current: 706/50
• CPC Class Codes: G06F 16/48 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/686 using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings

Abstract

A system and method for in-vehicle content recommendation is disclosed. The system comprises an aggregator module, a modeler module and a recommendation module. The aggregator module aggregates content data describing content such as a music, a news program, a podcast, an audio book, a movie, a television program, etc. The aggregator module also aggregates situation data describing an environment in which the content was recommended to a user, and feedback data describing the user'"'"'s response to the recommendation. The modeler module generates a content preference model. The content preference model represents the user'"'"'s dynamic content preferences based at least in part on the aggregated data. The recommendation module generates a first content recommendation based at least in part on the content preference model.

Claims

1. A computer-implemented method for recommending content, the method comprising:

aggregating content data describing content, situation data describing an environment in which the content was recommended to a user, and feedback data describing the user'"'"'s response to a content recommendation, the feedback data having a type that is one or more of a user request for a content recommendation;

a user response that indicates that the user accepted a previous content recommendation; and

a user response that indicates that the user rejected a previous content recommendation;

generating a content preference model that represents the user'"'"'s dynamic content preferences based at least in part on the aggregated data; and

generating a first content recommendation based at least in part on the content preference model.

2. The method of claim 1, the method further comprising:

presenting the first content recommendation to the user;

receiving feedback data from the user; and

generating an updated content preference model responsive to receiving the feedback data.

3. The method of claim 2, wherein generating the updated content preference model further comprises:

receiving the feedback data including a type of feedback received from the user and a timestamp indicating when the feedback was received;

collecting new situation data from one or more of a sensor and a camera responsive to the new feedback data being received and having a timestamp proximate to the timestamp of the new feedback data;

receiving the content data associated with the first content recommendation when the new feedback was received;

aggregating the feedback data, the new situation data and the content data associated with the content recommendation to create new aggregated data; and

generating the updated content preference model based at least in part on the new aggregated data.

4. The method of claim 2, the method further comprising:

responsive to the feedback data indicating that the user rejected the first content recommendation, generating a new content recommendation based at least in part on the updated content preference model.

5. The method of claim 1, wherein the situation data is based at least in part on scene category information generated from data collected by a camera.

6. The method of claim 1, wherein the situation data is based at least in part on data collected from one or more of a sensor and a camera, and the situation data has been vectorized to create a situation feature vector.

7. A system for recommending content, the system comprising:

an aggregator module for aggregating content data describing content, situation data describing an environment in which the content was recommended to a user, and feedback data describing the user'"'"'s response to a recommendation;

a modeler module for generating a content preference model that represents the user'"'"'s dynamic content preferences based at least in part on the aggregated data; and

a recommendation module for generating a first content recommendation based at least in part on the content preference model; and

a GUI module for presenting the first content recommendation and receiving feedback data including a timestamp.

8. The system of claim 7, the system further comprising:

one or more of a camera and a sensor for capturing situation data describing the environment of a vehicle;

a situation module for collecting, responsive to the feedback data being received by the GUI module, situation data from one or more of the sensor and the camera having a timestamp proximate to the timestamp of the feedback data;

a content library storing the content data associated with the first content recommendation presented when the feedback was received by the GUI module;

a playback history database for storing aggregated data resulting from the aggregation of the feedback data, the situation data and the content data associated with the first content recommendation presented when the feedback was received by the GUI module by the aggregation module; and

a model database for storing the content preference model based at least in part on the aggregated data and generated by the modeler module.

9. The system of claim 7, wherein the GUI module receives feedback data having a type that is one or more of:

a user request for the content recommendation;

a user response that indicates that the user accepted the first content recommendation; and

a user response that indicates that the user rejected the first content recommendation.

10. The system of claim 7, wherein responsive to feedback data indicating that the user rejected the first content recommendation, the recommendation module generating a new content recommendation based at least in part on the updated content preference model.

11. The system of claim 7, wherein the situation data is based at least in part on scene category information generated by the situation module processing data collected by a camera.

12. The system of claim 7, wherein the situation data is based at least in part on data collected from one or more of a sensor and a camera, which has been vectorized by the situation module to create a situation feature vector.

13. A computer program product comprising a computer usable medium including instructions that, when executed by a computer, cause the computer to perform the steps comprising:

aggregating content data describing content, situation data describing an environment in which the content was recommended to a user, and feedback data describing the user'"'"'s response to a content recommendation, the feedback data having a type that is one or more of a user request for a content recommendation;

a user response that indicates that the user accepted a previous content recommendation; and

a user response that indicates that the user rejected a previous content recommendation;

generating a content preference model that represents the user'"'"'s dynamic content preferences based at least in part on the aggregated data; and

generating a first content recommendation based at least in part on the content preference model.

14. The computer program product of claim 13, further comprise instructions for:

presenting the first content recommendation to the user;

receiving feedback data from the user; and

generating an updated content preference model responsive to receiving the feedback data.

15. The computer program product of claim 14, wherein the instructions for generating the updated content preference model further comprise instructions for:

receiving the feedback data including a type of feedback received from the user and a timestamp indicating when the feedback was received;

collecting new situation data from one or more of a sensor and a camera responsive to the new feedback data being received and having a timestamp proximate to the timestamp of the new feedback data;

receiving the content data associated with the first content recommendation when the new feedback was received;

aggregating the feedback data, the new situation data and the content data associated with the content recommendation to create new aggregated data; and

generating the updated content preference model based at least in part on the new aggregated data.

16. The computer program product of claim 14, the further comprising instructions for:

determining that the feedback data indicates that the user rejected the first content recommendation, and responsive to the determination, generating a new content recommendation based at least in part on the updated content preference model.

17. The computer program product of claim 13, wherein the situation data is based at least in part on scene category information generated from data collected by a camera.