Predictive service for third party application developers

• US 8,812,416 B2
• Filed: 11/08/2011
• Issued: 08/19/2014
• Est. Priority Date: 11/08/2011
• Status: Expired due to Fees

• Current Assignee: Nokia Technologies Oy (Nokia Corporation)
• Original Assignee: Nokia Corporation
• Inventors: Dousse, Olivier, Laurila, Juha K., Nefedov, Nikolai, Mielikainen, Taneli
• Primary Examiner(s): Chang, Li-Wu
• Assistant Examiner(s): Misir, Dave
• Application Number:
  US13/291,666
• Publication Number:
  US 20130117208A1
• Time in Patent Office: 1,015 Days
• Field of Search: 706/11, 706/12, 706/21
• US Class Current: 706/21
• CPC Class Codes: G06F 9/44505 Configuring for program initiating, e.g. using registry, configuration files
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

Disclosed is an apparatus, method and computer program device which send a prediction request to a prediction service to construct a prediction result, receives a prediction result and an estimation of accuracy of the prediction result from the prediction service and configures an application to adapt to at least one of user preferences, behavior and habits based upon the prediction result and estimation of accuracy of the prediction result. The prediction service can include an over-the-air server which is coupled to a prediction server or a dedicated API residing on one or more user devices and configured to access a prediction server.

Claims

1. A method comprising:

sending a prediction request to a prediction service to construct a prediction result;

receiving a constructed prediction result and an estimation of accuracy of the prediction result; and

configuring an application to adapt to at least one of user preferences, behavior and habits based upon the prediction result and estimation of accuracy of the prediction result;

wherein the prediction request includes a plurality of uniform profile vectors that are used to protect a user'"'"'s privacy and sensitive personal information, which are created by extracting features from databases of a plurality of user devices, and wherein the profile vectors are computed locally on an electronic device based upon usage and/or context data gathered on the device itself.

2. The method of claim 1, wherein the prediction service is comprised of an over-the-air server which is coupled to a prediction server.

3. The method of claim 1, wherein the prediction service is comprised of a dedicated API residing on one or more user devices and configured to access a prediction server.

4. The method of claim 1, wherein the prediction request further comprises:

a training set corresponding to a subset of a plurality of uniform profile vectors usernames, and a designation of a predictor-type.

5. The method of claim 4, wherein the predictor-type comprises:

a structured induction score including a transition model for a specific user or group of users;

a data summarization including descriptive statistics of the specific user or group of users to indicate a classifier; and

a probability estimation including classification, similarity measurements, regression analysis scoring, or ranking.

6. The method of claim 1, wherein the extracted features comprises the user'"'"'s preferences, behavior and habits and comprise:

a mobility category;

a social behavior category;

a preferences category; and

a demographic data category.

7. The method of claim 1, further comprising:

sending a prediction usage request to the prediction service to construct a prediction usage result;

receiving a prediction usage result and an estimation of accuracy of the prediction usage result; and

configuring the application to adapt to the user preferences, behavior and habits based upon the prediction usage result and estimation of accuracy of the prediction usage result.

8. The method of claim 1, further comprising:

sending a prediction update request to the prediction service to construct a prediction update result;

receiving a prediction update result and an estimation of accuracy of the prediction update result; and

configuring the application to adapt to the user preferences, behavior and habits based upon the prediction update result and estimation of accuracy of the prediction update result.

9. The method of claim 1, further comprising limiting access to a user data to conduct requests for prediction calculation or relevance calculation.

10. The method of claim 9, wherein a requester is limited to those segments of a user profile which are needed from the perspective of a particular application.

11. The method of claim 9, wherein a requester is limited in number and frequency of requests for prediction calculation or relevance calculation.

12. The method of claim 9, wherein a requester is restricted from direct access to a user device or a mobile platform vendor or mobile ecosystem contributor.

13. A non-transitory computer-readable medium that contains software program instructions, where execution of the software program instructions by at least one data processor results in the performance of operations that comprise execution of the method of claim 1.

14. An apparatus comprising a processor and a memory including computer program code, wherein the memory and the computer program code are configured, with the processor, to cause the apparatus to perform at least:

sending a prediction request to a prediction service to construct a prediction result;

receiving a constructed prediction result and an estimation of accuracy of the prediction result; and

configuring an application to adapt to at least one of user preferences, behavior and habits based upon the prediction result and estimation of accuracy of the prediction result;

wherein the prediction request includes a plurality of uniform profile vectors that are used to protect a user'"'"'s privacy and sensitive personal information, which are created by extracting features from databases of a plurality of user devices, and wherein the profile vectors are computed locally on an electronic device based upon usage and/or context data gathered on the device itself.

15. The apparatus of claim 14, wherein, the apparatus is a mobile device such as a smart phone, tablet or portable computer.

16. The apparatus of claim 14, wherein, the apparatus is over the air server connected to a communication network.

17. The apparatus of claim 14, wherein, the apparatus is a set-top box connected to a cable, Internet or satellite telecommunication network.

18. The apparatus of claim 14, wherein, the apparatus is a standalone device capable of connecting to a telecommunication network.

19. An apparatus comprising a processor and a memory including computer program code, wherein the memory and the computer program code are configured, with the processor, to cause the apparatus to perform at least:

receiving one or more prediction requests from one or more third parties;

calculating a prediction result based on the received prediction request and an estimation of accuracy of the prediction result; and

sending the prediction result and the estimation of accuracy to third party to enable the third party to configure an application to adapt to at least one of user preferences, behavior and habits based upon the prediction result and estimation of accuracy of the prediction result;

wherein the prediction request includes a plurality of uniform profile vectors that are used to protect a user'"'"'s privacy and sensitive personal information, which are created by extracting features from databases of a plurality of user devices, and wherein the profile vectors are computed locally on an electronic device based upon usage and/or context data gathered on the device itself.