Method and apparatus of user recommendation system for in-vehicle apps

• US 9,430,476 B2
• Filed: 03/13/2014
• Issued: 08/30/2016
• Est. Priority Date: 03/13/2014
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations LLC (General Motors Company)
• Inventors: Bai, Fan, Grimm, Donald K., Osella, Massimo, Nieman, Leonard C., Hrabak, Robert A.
• Primary Examiner(s): Corrielus, Jean M
• Application Number:
  US14/210,047
• Publication Number:
  US 20150261754A1
• Time in Patent Office: 901 Days
• Field of Search: 701/533, 701/36, 707/749, 707/740, 707/724, 707/722, 707/769, 707/748
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products

Abstract

A method and system for recommending applications to users of in-vehicle infotainment systems are disclosed. Application rating data from many road vehicle infotainment system users are collected on a central server, including both explicit ratings and implicit ratings. Implicit ratings may be calculated based on application usage data. The user/application rating data is filtered for relevance, and then analyzed to determine inferred ratings for user/application relationships where no rating exists. The inferred ratings are calculated using both a user-driven consensus rating calculation and an application-driven consensus rating calculation. The inferred ratings, along with optional cyberspace-based external inputs, are used to synthesize application recommendations for users. The synthesized recommendations for application consideration are provided to the appropriate user via downloading from the central server to the infotainment system in the user'"'"'s vehicle.

Claims

1. A method for making user recommendations for infotainment system applications, said method comprising:

collecting, on a central server computer including a processor and a memory module, user rating data for in-vehicle infotainment system applications from many infotainment system users, including both explicit ratings and implicit ratings;

filtering the user rating data for the applications based on relevance to produce filtered user/application rating data;

computing, on the central server computer, inferred ratings for user/application relationships where no rating is available, using the filtered user/application rating data in a user-driven consensus calculation;

computing, on the central server computer, inferred ratings for user/application relationships where no rating is available, using the filtered user/application rating data in an application-driven consensus calculation;

synthesizing user application recommendations based on the inferred ratings from the user-driven consensus calculation and the application-driven consensus calculation; and

providing the user application recommendations to the infotainment system users for the user/application relationships where no rating is available and the inferred ratings are high.

2. The method of claim 1 wherein the implicit ratings are calculated based on recency of a user viewing an application, recency of the user using the application, frequency of the user using the application, duration of the user'"'"'s usage of the application and monetary value of the application.

3. The method of claim 1 further comprising combining the explicit ratings and the implicit ratings using a weighted average before filtering the user rating data.

4. The method of claim 1 wherein filtering the user rating data includes filtering the users based on user attributes and filtering the applications based on application attributes.

5. The method of claim 1 wherein the user-driven consensus calculation includes calculating a similarity between a target user and other users based on applications which both the target user and the other users have rated.

6. The method of claim 5 wherein the user-driven consensus calculation also includes identifying nearest neighbor users, which are the other users that are nearest neighbors to the target user in application rating similarity.

7. The method of claim 6 wherein the user-driven consensus calculation also includes calculating an inferred rating of a particular application by the target user based on the similarity between the target user and the nearest neighbor users, ratings of the particular application by the nearest neighbor users, and an average rating of all applications by the target user.

8. The method of claim 1 wherein the application-driven consensus calculation includes calculating a similarity between a target application and other applications based on users that have rated both the target application and the other applications.

9. The method of claim 8 wherein the application-driven consensus calculation also includes identifying nearest neighbor applications, which are the other applications which are nearest neighbors to the target application in user rating similarity.

10. The method of claim 9 wherein the application-driven consensus calculation also includes calculating an inferred rating of the target application by a particular user based on the similarity between the target application and the nearest neighbor applications, ratings by the particular user of the nearest neighbor applications, and an average rating by all users of the target application.

11. The method of claim 1 wherein synthesizing user application recommendations includes using the inferred ratings from the user-driven consensus calculation and the application-driven consensus calculation, along with external cyberspace-based inputs, in a weighted average calculation.

12. A method for making user recommendations for infotainment system applications, said method comprising:

collecting, on a central server computer including a processor and a memory module, user rating data for in-vehicle infotainment system applications from many infotainment system users, including both explicit ratings and implicit ratings, where the implicit ratings are calculated based on recency of a user viewing an application, recency of the user using the application, frequency of the user using the application, duration of the user'"'"'s usage of the application and monetary value of the application;

filtering the user rating data for the applications based on relevance to produce filtered user/application rating data, including filtering the users based on user attributes and filtering the applications based on application attributes;

computing, on the central server computer, inferred ratings for user/application relationships where no rating is available, using the filtered user/application rating data in a user-driven consensus calculation;

computing, on the central server computer, inferred ratings for user/application relationships where no rating is available, using the filtered user/application rating data in an application-driven consensus calculation;

synthesizing user application recommendations based on the inferred ratings from the user-driven consensus calculation and the application-driven consensus calculation, including using the inferred ratings from the user-driven consensus calculation and the application-driven consensus calculation, along with external cyberspace-based inputs, in a weighted average calculation; and

providing the user application recommendations to the infotainment system users for the user/application relationships where no rating is available and the inferred ratings are high.

13. The method of claim 12 wherein:

the user-driven consensus calculation includes calculating an inferred rating of a particular application by a target user based on a similarity between the target user and a set of nearest neighbor users, ratings of the particular application by the nearest neighbor users, and an average rating of all applications by the target user, where the nearest neighbor users are other users that are nearest neighbors to the target user in application rating similarity; and

the application-driven consensus calculation includes calculating an inferred rating of a target application by a particular user based on a similarity between the target application and a set of nearest neighbor applications, ratings by the particular user of the nearest neighbor applications, and an average rating by all users of the target application, where the nearest neighbor applications are other applications that are nearest neighbors to the target application in user rating similarity.

14. A system for making user recommendations for infotainment system applications, said system comprising:

a computer including a processor, a memory module and a network connection, where the computer is configured with an algorithm for recommending infotainment system applications including;

a rating module configured to collect and process user rating data for in-vehicle infotainment system applications from many infotainment system users, including both explicit ratings and implicit ratings, where the user rating data is wirelessly uploaded from a plurality of road vehicles;

a filtering module configured to filter the user rating data for the applications based on relevance to produce filtered user/application rating data;

a user-driven consensus module configured to compute inferred ratings for user/application relationships where no rating is available, using the filtered user/application rating data in a user-driven consensus calculation;

an application-driven consensus module configured to compute inferred ratings for user/application relationships where no rating is available, using the filtered user/application rating data in an application-driven consensus calculation; and

a recommendation synthesizer module configured to synthesize user application recommendations based on the inferred ratings from the user-driven consensus module and the application-driven consensus module, and provide the recommendations to the infotainment system users for the user/application relationships where no rating is available and the inferred ratings are high.

15. The system of claim 14 wherein the implicit ratings are calculated based on recency of a user viewing an application, recency of the user using the application, frequency of the user using the application, duration of the user'"'"'s usage of the application and monetary value of the application.

16. The system of claim 14 wherein the user-driven consensus module calculates a similarity between a target user and other users based on applications which both the target user and the other users have rated, and identifies nearest neighbor users, which are the other users that are nearest neighbors to the target user in application rating similarity.

17. The system of claim 16 wherein the user-driven consensus module also calculates an inferred rating of a particular application by the target user based on the similarity between the target user and the nearest neighbor users, ratings of the particular application by the nearest neighbor users, and an average rating of all applications by the target user.

18. The system of claim 14 wherein the application-driven consensus module calculates a similarity between a target application and other applications based on users that have rated both the target application and the other applications, and identifies nearest neighbor applications, which are the other applications which are nearest neighbors to the target application in user rating similarity.

19. The system of claim 18 wherein the application-driven consensus module also calculates an inferred rating of the target application by a particular user based on the similarity between the target application and the nearest neighbor applications, ratings by the particular user of the nearest neighbor applications, and an average rating by all users of the target application.

20. The system of claim 14 wherein the recommendation synthesizer module includes the inferred ratings from the user-driven consensus module and the application-driven consensus module, along with external cyberspace-based inputs, in a weighted average calculation.