Location Data for Defining Places and Traffic

One or more non-transitory, computer-readable storage media embodying software that is operable when executed to:

• select a plurality of location updates corresponding to a place, the plurality of location updates being received from one or more users, wherein each location update comprises:
• data indicating a geographic location that a user was at;
• and
• a time stamp indicating a time when the user was at the geographic location;
• select a first subset of the location updates, wherein the location updates in the first subset correspond to geographic locations within a geographic area;
• for each of one or more of the location updates in the first subset:
• determine the user corresponding to the location update;
• and
• determine the time corresponding to the location update;
• select a second subset of the location updates, wherein each of the location updates in the second subset corresponds to:
• a user corresponding to a location update in the first subset;
• and
• a time within a threshold amount of time of the time corresponding to the location update in the first subset;
• generate a probability map based on a distribution of geographic locations corresponding to the location updates in the second subset;
• and
• define one or more boundaries corresponding to the place based at least in part on the probability map.

A system comprising:

• one or more processors;
• and
• a memory coupled to the processors comprising instructions executable by the processors, the processors being operable when executing the instructions to:
• select a plurality of location updates corresponding to a place, the plurality of location updates being received from one or more users, wherein each location update comprises:
• data indicating a geographic location that a user was at;
• and
• a time stamp indicating a time when the user was at the geographic location;
• select a first subset of the location updates, wherein the location updates in the first subset correspond to geographic locations within a geographic area;
• for each of one or more of the location updates in the first subset:
• determine the user corresponding to the location update;
• and
• determine the time corresponding to the location update;
• select a second subset of the location updates, wherein each of the location updates in the second subset corresponds to:
• a user corresponding to a location update in the first subset;
• and
• a time within a threshold amount of time of the time corresponding to the location update in the first subset;
• generate a probability map based on a distribution of geographic locations corresponding to the location updates in the second subset;
• and
• define one or more boundaries corresponding to the place based at least in part on the probability map.