Location context inference based on user mobile data with uncertainty

1. A method performed by a computer server for inferring location context categories for a set of mobile users having at least two members, comprising:

for each mobile user in the set, obtaining at least one location context category therefor from publically available information responsive to uncertain mobile device location data; and

applying multi-user collaborative machine learning with an objective function to the at least one location context category for each mobile user in the set to infer a single refined location context category for each mobile user in the set to form a matrix.

2. The method of claim 1 , wherein the uncertain mobile device location data comprises cellular telephone location data generated by a cellular telephone locator.

3. The method of claim 1 , wherein said obtaining step comprises:

for each mobile user in the set, sending a respective query to one or more mobile web applications that return the at least one location context category applicable to a submitting one of the mobile users responsive to the uncertain mobile device location data specified in the respective query; and

for each mobile user in the set, receiving the at least one location context category from the mobile web application responsive to the respective query.

4. The method of claim 1 , further comprising categorizing the mobile users into multiple groups based on similarities between the mobile users.

5. The method of claim 1 , wherein the multi-user collaborative machine learning is performed using an objective function expressed using a nuclear norm.

6. The method of claim 1 , wherein said matrix has each one of columns and rows correspond to a respective one of the mobile users, and each of the other one of the columns and the rows correspond to a respective one of the multiple location context categories.

7. The method of claim 1 , wherein the objective function is expressed using a convex envelope of a non-convex rank function subject to a row constraint on the matrix.

8. The method of claim 1 , further comprising, for at least one mobile user in the set, inferring at least one activity performed by or of interest to the at least one mobile user based on the single refined location context category.

9. The method of claim 1 , further comprising, for at least one mobile user in the set, inferring at least one future intention of the at least one mobile user based on the single refined location context category.

10. A non-transitory article of manufacture tangibly embodying a computer readable program which when executed causes a computer to perform the steps of claim 1 .

11. A method performed by a computer server for inferring location context categories for a set of mobile users having at least two members, comprising:

for each mobile user in the set, sending a respective query to one or more mobile web applications that return at least one location context category applicable to a submitting one of the mobile users responsive to a mobile user location input specified in the respective query;

for each mobile user in the set, receiving the at least one location context category from the mobile web application responsive to the respective query; and

applying multi-user collaborative machine learning with an objective function to the at least one location context category for each mobile user in the set to infer a single refined location context category for each mobile user in the set to form a matrix.

12. The method of claim 11 , wherein at least one of the one or more mobile web applications at least one of use or provide publically available location context category information.

13. The method of claim 11 , further comprising categorizing the mobile users into multiple groups based on similarities between the mobile users.

14. The method of claim 11 , further comprising, for at least one mobile user in the set, inferring at least one activity performed by or of interest to the at least one mobile user based on the single refined location context category.

15. The method of claim 11 , further comprising, for at least one mobile user in the set, inferring at least one future intention of the at least one mobile user based on the single refined location context category.

16. A non-transitory article of manufacture tangibly embodying a computer readable program which when executed causes a computer to perform the steps of claim 11 .

17. A system for inferring location context categories for a set of mobile users having at least two members, comprising:

a computer server, having a processor and a memory, configured to:

for each mobile user in the set, obtain at least one location context category therefor from publically available information responsive to uncertain mobile device location data; and

apply multi-user collaborative machine learning with an objective function to the at least one location context category for each mobile user in the set to infer a single refined location context category for each mobile user in the set to form a matrix.

18. The system of claim 17 , wherein the computer server is implemented using a cloud computing configuration.

19. The system of claim 17 , wherein the computer server is further configured to, for at least one mobile user in the set, infer at least one activity performed by or of interest to the at least one mobile user based on the single refined location context category.

20. The system of claim 17 , wherein the computer server is further configured to, for at least one mobile user in the set, infer at least one future intention of the at least one mobile user based on the single refined location context category.