User localization using friend location information
View Patent ↗A method, system, and medium are provided for estimating a location of a user using location information of the user's friends. The user's friends are identified, and a social-interaction strength between the user and each of the friends is determined. This results in a set of weights of social ties that respectively indicate a level of social interaction between the user and the user's friends. The location of the friends is determined. The user's location is estimated based on the set of weights of social ties and the friend locations.
1. One of more computer storage devices having computer-executable instructions embodied thereon that, when executed, enable a computing device to perform a method of estimating a location of a user of a mobile device using locations of one or more friends of the user, the method comprising:
identifying the one or more friends of the user;
determining a social-interaction strength between the user and each of the one or more friends, resulting in a set of weights of social ties that respectively indicate a level of social interaction between the user and each of the one or more friends;
determining a location of each of the one or more friends, resulting in a set of friend locations; and
estimating the location of the user based on the set of friend locations and the set of weights.
2. The one or more computer storage devices of claim 1 , wherein identifying the one or more friends of the user comprises at least one of identifying friends in the user's online social network, accessing call data records of the user, or accessing messaging records of the user.
3. The one or more computer storage devices of claim 1 , wherein determining a social-interaction strength between the user and each of the one or more friends of the user comprises at least one of the following:
(A) mobile device talk time between the user and each of the one or more friends,
(B) a number of mobile device calls between the user and each of the one or more friends,
(C) a number of electronic mail messages between the user and each of the one or more friends,
(D) a number of visits to a same Web page by the user and each of the one or more friends,
(E) a number of text messages between the user and each of the one or more friends,
(F) a number of times the user recommends a uniform resource locator to each of the one or more friends,
(G) a number of times the user plays a multiplayer game with each of the one or more friends,
(H) a number of times the user visited a geographical location with each of the one or more friends, or
(I) a fraction of any of the above as compared to a respective total of each of the above.
4. The one or more computer storage devices of claim 1 , wherein determining a location of each of the one or more friends of the user includes localizing an event generated by each of the one or more friends.
5. The one or more computer storage devices of claim 4 , wherein the event generated by each of the one or more friends includes making a call on a mobile device.
6. The one or more computer storage devices of claim 1 , wherein the set of weights is time dependent.
7. The one or more computer storage devices of claim 1 , wherein the location of the user is estimated by utilizing conditional probability based on the set of friend locations.
8. The one or more computer storage devices of claim 7 , wherein the location of the user is a distribution of user locations, and wherein a most probable location of the user is within the distribution of user locations.
9. The one or more computer storage devices of claim 8 , wherein the location of the user is based on a maxima of the distribution of user locations, wherein the maxima of the distribution of user locations indicates a set of probable locations.
10. The one or more computer storage devices of claim 1 , wherein given that the location of each of the one or more friends is within a close geographic proximity of one another, estimating the location of the user by using a centroid of the set of friend locations.
11. A computerized method carried out by at least one server having at least one processor for performing a method of estimating a location of a user of a mobile device using locations associated with one or more friends of the user, the method comprising:
identifying the one or more friends of the user;
determining, using the at least one processor, a location of each of the one or more friends, resulting in a set of friend locations;
determining mutual information that is shared between the user and the one or more friends, resulting in a set of weights of mutual information that respectively indicate a level of shared information between the user and each of the one or more friends; and
estimating the location of the user based on the set of friend locations and the set of weights.
12. The computerized method of claim 11 , wherein the location of each of the one or more friends is determined using at least one of:
(A) Global Positioning System,
(B) call sector information,
(C) triangulation information,
(D) round-trip-time information,
(E) signal-to-noise-and-interference ration, or
(F) network topology.
13. The computerized method of claim 11 , wherein the mutual information is determined using a pair-wise approach between the user and each of the one or more friends.
14. The computerized method of claim 11 , wherein the mutual information is determined based on involvement of the user and the one or more friends with one or more social groups.
15. The computerized method of claim 11 , wherein the mutual information is time dependent.
16. The computerized method of claim 15 , wherein the time depends upon at least one of the following:
(A) daytime versus nighttime,
(B) weekend versus weekday,
(C) holiday versus nonholiday,
(D) time of year, or
(E) workday hours versus nonworkday hours.