User description based on contexts of location and time
View Patent ↗A description of a user is estimated based on the context of a user's past and present locations. A disclosed data-processing system continually receives data points for each user that represent spatial and/or temporal events. These events represent, for example, presence of a person at a specific geographic location such as a geographic area or point of interest (POI). The data-processing system evaluates the received data points in relation to one or more of the geographic locations, yielding results that are also based on the demographic characteristics of each visited location and the commercial characteristics of each visited location. The data-processing system evaluates the data points also to determine patterns exhibited in each user's activity or inactivity, and patterns exhibited in the distance traveled and the type of travel. The data-processing system bases the user descriptions on the results of these evaluations.
1. A method comprising:
receiving, by a data-processing system, a plurality of data points D that correspond to a first user, wherein each of the data points in the plurality represents at least one of i) a spatial event and ii) a temporal event, and wherein the plurality comprises at least i) a first data point that occurs at a first time t 1 , d(t 1 ), and ii) a second data point that occurs at a second time t 2 , d(t 2 );
determining a difference in time between a first time that is represented by the first data point d(t 1 ) and a second time that is represented by the second data point d(t 2 );
evaluating
i) a first characteristic c 1 for a first geographic location I 1 , yielding a first value c 1 (I 1 ), and
ii) the first data point d(t 1 ) in relation to the first geographic location I 1 , yielding a first result that is based on the first value c 1 (I 1 ); and
estimating a description of the first user, wherein the description is based on i) the determined difference in time and ii) the first result.
2. The method of claim 1 wherein the first geographic location I 1 is a geographic area.
3. The method of claim 1 wherein the first geographic location I 1 is a geographic point of interest.
4. The method of claim 3 wherein evaluating the first data point d(t 1 ) comprises determining proximity of a geolocation represented by d(t 1 ) to the geographic point of interest represented by the first geographic location I 1 .
5. The method of claim 1 further comprising assessing inactivity of the first user based on the difference in time between d(t 1 ) and d(t 2 ), wherein d(t 1 ) and d(t 2 ) represent a pair of consecutive events, and wherein the description of the first user is based also on the inactivity.
6. The method of claim 5 wherein assessing inactivity comprises assigning the pair of consecutive events to a predetermined bin.
7. The method of claim 6 wherein the predetermined bin is one of a plurality of bins for tracking user inactivity.
8. The method of claim 1 further comprising:
determining a difference in geolocation between a first geolocation that is represented by the first data point d(t 1 ) and a second geolocation that is represented by the second data point d(t 2 );
wherein the description of the first user is also based on the determined difference in geolocation.
9. The method of claim 8 further comprising assessing distance traveled of the first user based on the difference in geolocation between d(t 1 ) and d(t 2 ), wherein the description of the first user is based also on the distance traveled.
10. The method of claim 9 wherein d(t 1 ) and d(t 2 ) represent a pair of consecutive events.
11. A method comprising:
receiving, by a data-processing system, a plurality of data points D that correspond to a first user, wherein each of the data points in the plurality represents at least one of i) a spatial event and ii) a temporal event, and wherein the plurality comprises at least i) a first data point that occurs at a first time t 1 , d(t 1 ), and ii) a second data point that occurs at a second time t 2 , d(t 2 );
determining a difference in geolocation between a first geolocation that is represented by the first data point d(t 1 ) and a second geolocation that is represented by the second data point d(t 2 );
evaluating
i) a first characteristic c 1 for a first geographic location I 1 , yielding a first value c 1 (I 1 ), and
ii) the first data point d(t 1 ) in relation to the first geographic location I 1 , yielding a first result that is based on the first value c 1 (I 1 ); and
estimating a description of the first user, wherein the description is based on i) the determined difference in geolocation and ii) the first result.
12. The method of claim 11 wherein the first geographic location I 1 is a geographic area.
13. The method of claim 11 wherein the first geographic location I 1 is a geographic point of interest.
14. The method of claim 13 wherein evaluating the first data point d(t 1 ) comprises determining proximity of the first geolocation represented by d(t 1 ), to the geographic point of interest represented by the first geographic location I 1 .
15. The method of claim 14 wherein determining the proximity to the geographic point of interest comprises weighting the distances of the first geolocation to multiple points of interest.
16. The method of claim 11 further comprising assessing distance traveled of the first user based on the difference in geolocation between d(t 1 ) and d(t 2 ), wherein the description of the first user is based also on the distance traveled.
17. The method of claim 16 wherein d(t 1 ) and d(t 2 ) represent a first pair of consecutive events.
18. The method of claim 17 wherein assessing the distance traveled comprises summing i) the difference in geolocation of the first pair of consecutive events and ii) a difference in geolocation of a second pair of consecutive events, wherein d(t 2 ) is also an event in the second pair.
19. The method of claim 16 wherein the description of the first user is also based on time of day.
20. A method comprising:
receiving, by a data-processing system, a plurality of data points D that correspond to a first user, wherein each of the data points in the plurality represents at least one of i) a spatial event and ii) a temporal event, and wherein the plurality comprises at least i) a first data point that occurs at a first time t 1 , d(t 1 ), and ii) a second data point that occurs at a second time t 2 , d(t 2 );
determining a level of entropy of the plurality of data points D, wherein the level of entropy characterizes a randomness of the plurality;
evaluating
i) a first characteristic c 1 for a first geographic location I 1 , yielding a first value c 1 (I 1 ), and
ii) the first data point d(t 1 ) in relation to the first geographic location I 1 , yielding a first result that is based on the first value c 1 (I 1 ); and
estimating a description of the first user, wherein the description is based on i) the determined level of entropy and ii) the first result.
21. The method of claim 20 wherein the first geographic location I 1 is a geographic area.
22. The method of claim 20 wherein the first geographic location I 1 is a geographic point of interest.
23. The method of claim 20 wherein evaluating the first data point d(t 1 ) comprises determining proximity of a geolocation represented by d(t 1 ) to the geographic point of interest represented by the first geographic location I 1 .
24. The method of claim 20 wherein determining the level of entropy comprises determining whether the first data point d(t 1 ) and the second data point d(t 2 ) correspond to different places visited by the first user.
25. The method of claim 24 wherein determining the level of entropy further comprises tracking the number of places visited by the first user.
26. The method of claim 20 further comprising assessing inactivity of the first user based on the difference in time between d(t 1 ) and d(t 2 ), wherein d(t 1 ) and d(t 2 ) represent a pair of consecutive events, and wherein the description of the first user is based also on the inactivity.
27. The method of claim 26 wherein assessing inactivity comprises assigning the pair of consecutive events to a predetermined bin.
28. The method of claim 27 wherein the predetermined bin is one of a plurality of bins for tracking user inactivity.