System and method for identifying associated subjects from location histories
Systems and methods to track the respective locations of subjects over time. The system identifies subjects who, over time, were co-located with one another suggesting they are associated with one another, and the pairs are analyzed. For each of the subjects, the system produces a vector that quantifies the subject's location history by including a respective weight for each combination of a time interval with a geographical area. The vectors are compared using a distance metric, and any pair of subjects whose vectors are sufficiently close are flagged as being an associated pair. The respective vector belonging to each subject is normalized to account for the total number of other subjects who were co-located with the subject. For each interval-area pair, the system may compute the frequency of the interval-area pair, and then divide each weight that corresponds to the interval-area pair by the frequency of the interval-area pair.
1. A system, comprising:
a communication interface; and
a processor, executing instructions to:
receive, via the communication interface, tracking data that indicate respective locations of a plurality of subjects within a plurality of geographic areas and during a plurality of time intervals;
calculating a plurality of weights that respectively correspond to each combination of a subject, a geographic area, and a time interval;
select a plurality of interval-area pairs, each of which includes a respective one of the geographic areas and a respective one of the time intervals;
for each of the selected interval-area pairs, normalize a respective subset of the weights to construct a plurality of vectors that correspond to respective combinations of the subject with the selected interval-area pairs to determine measures of similarity between one or more pairs of the vectors; and
in response to the measures of similarity, identify an association of the subject and at least one other subject of interest.
2. The system according to claim 1 , wherein the processor further executes instructions to receive the tracking data from a cellular the network.
3. The system of claim 1 , wherein the processor further executes instructions to maintain a location history for the plurality of subjects.
4. The system according to claim 1 , wherein the processor further executes instructions to calculate each of the weights responsively to a percentage of the time interval during which the subject was located in the geographic area that is indicated by the tracking data.
5. The system according to claim 1 , wherein the processor further executes instructions to calculate each of the weights by:
calculating a level of confidence that the tracking data indicate that the subject was located in the geographic area during the time interval; and
calculating the weight responsively to the level of confidence.
6. The system according to claim 1 , wherein the processor further executes instructions to normalize the respective subset of the weights by:
calculating a normalizing factor as (i) an increasing function of a sum of the subset of the weights, and (ii) a decreasing function of a total sum of the weights; and
normalizing the subset of the weights by the normalizing factor.
7. The system according to claim 1 , wherein the processor further executes instructions to normalize the respective subset of the weights by:
calculating a normalizing factor as (i) an increasing function of a number of those of the weights in the subset that are greater than zero, and (ii) a decreasing function of a total number of the subjects; and
normalizing each of the weights in the subset by the normalizing factor.
8. The system according to claim 1 , wherein the processor further executes instructions to, prior to calculating the measures of similarity:
cluster the vectors into a plurality of different clusters; and
subsequently to clustering the vectors, select the pairs of the vectors in response to each of the pairs of the vectors being contained within a same one of the clusters or within respective ones of the clusters that are within a predefined threshold distance of one another.
9. The system according to claim 1 , wherein the processor further executes instructions to output the association of the subject and the at least one other subject, and to visually display the output.
10. A non-transitory computer readable having computer executable instructions stored thereon that when execute by a processor causes the processor to:
receive, via the communication interface, tracking data that indicate respective locations of a plurality of subjects within a plurality of geographic areas and during a plurality of time intervals;
calculating a plurality of weights that respectively correspond to each combination of a subject, a geographic area, and a time interval;
select a plurality of interval-area pairs, each of which includes a respective one of the geographic areas and a respective one of the time intervals;
for each of the selected interval-area pairs, normalize a respective subset of the weights to construct a plurality of vectors that correspond to respective combinations of the subject with the selected interval-area pairs to determine measures of similarity between one or more pairs of the vectors; and
in response to the measures of similarity, identify an association of the subject and at least one other subject of interest.
11. A method, comprising:
receiving, via a communication interface of a computing device, tracking data that indicate respective locations of a plurality of subjects within a plurality of geographic areas and during a plurality of time intervals;
calculating a plurality of weights that respectively correspond to each combination of a subject, a geographic area, and a time interval;
selecting a plurality of interval-area pairs, each of which includes a respective one of the geographic areas and a respective one of the time intervals;
for each of the selected interval-area pairs, normalizing a respective subset of the weights to construct a plurality of vectors that correspond to respective combinations of the subject with the selected interval-area pairs to determine measures of similarity between one or more pairs of the vectors; and
in response to the measures of similarity, identifying an association of the subject and at least one other subject of interest.
12. The method according to claim 11 , further comprising receiving the tracking data from a cellular the network.
13. The method of claim 11 , further comprising maintaining a location history for the plurality of subjects.
14. The method according to claim 11 , further comprising calculating each of the weights responsively to a percentage of the time interval during which the subject was located in the geographic area that is indicated by the tracking data.
15. The method according to claim 11 , further comprising calculating each of the weights by:
calculating a level of confidence that the tracking data indicate that the subject was located in the geographic area during the time interval, and
calculating the weight responsively to the level of confidence.
16. The method according to claim 11 , further comprising normalizing the respective subset of the weights by:
calculating a normalizing factor as (i) an increasing function of a sum of the subset of the weights, and (ii) a decreasing function of a total sum of the weights, and
normalizing the subset of the weights by the normalizing factor.
17. The method according to claim 11 , further comprising normalizing the respective subset of the weights by:
calculating a normalizing factor as (i) an increasing function of a number of those of the weights in the subset that are greater than zero, and (ii) a decreasing function of a total number of the subjects, and
normalizing each of the weights in the subset by the normalizing factor.
18. The method according to claim 11 , further comprising, prior to calculating the measures of similarity:
clustering the vectors into a plurality of different clusters, and
subsequently to clustering the vectors, selecting the pairs of the vectors in response to each of the pairs of the vectors being contained within a same one of the clusters or within respective ones of the clusters that are within a predefined threshold distance of one another.
19. The method according to claim 11 , further comprising outputting the association of the subject and the at least one other subject, and to visually display the output.
20. The method according to claim 19 , further comprising outputting one or more pairings of other subjects associated with the subject.