Multiple device intersection using geo-temporal mobile application data
A computer-implemented method for identifying device intersections using application data derived from multiple mobile devices. The method extracts the unique devices identified in a user defined area of interest (AOI) and generates individual geospatial layers for each device. The layers are derived from the metadata containing coordinates and associated timestamps for all available points of the device. Combining the individual layers will produce a convergence surface of polygonal intersections, indicating where devices have shared additional locations outside the initial AOI. Applying the timestamp to the resulting layer with a user defined time interval will generate a probability matrix of association.
1. A method of evaluating associations between electronic devices, comprising the steps of:
defining a geospatial area of interest (AOI) and a time of interest (TOI);
identifying devices contained within the AOI and TOI;
selecting at least a subset of the identified devices;
defining a radius and a time interval proximity;
generating at least one individual geospatial layer for each selected device;
creating a convergence surface for the selected devices;
retrieving a set of timestamps for each selected device in the convergence surface;
identifying a timespan for each of the selected devices in the convergence surface based on a first device timestamp and a last device timestamp;
identifying one or more overlapping timespans for the selected devices; and
generating an association matrix including the one or more overlapping timespans.
2. The method of claim 1 further comprising the step of generating an overall probability of association matrix.
3. The method of claim 1 wherein the convergence surface is defined by polygonal intersections of the geospatial layers.
4. The method of claim 1 further comprising the step of calculating an individual convergence zone C intersection matrix probability score according to:
∑
l
=
1
l
R
ijl
min
(
∑
l
=
1
l
R
t
l
∑
l
=
1
l
R
j
l
)
i
≠
j
.
5. The method of claim 1 further comprising the step of calculating an overall network intersection matrix probability score for each device pair according to:
∑
k
=
1
k
C
k
∑
l
=
1
l
R
ijl
min
(
∑
k
=
1
k
C
k
∑
l
=
1
l
R
il
∑
k
=
1
k
C
k
∑
l
=
1
l
R
jl
)
i
≠
j
.
6. A computer-implemented method for identifying device intersections using application data derived from a plurality of mobile devices, comprising the steps of:
identifying a plurality of mobile devices contained in an initial user defined area of interest (AOI) and generating a set of individual geospatial layers for each of the devices, wherein the set of individual geospatial layers are derived from a set of metadata containing a plurality of coordinate points and a plurality of associated timestamp points for each of the devices;
combining the individual layers to produce a convergence surface of polygonal intersections indicating where devices have shared one or more additional locations outside the initial user defined AOI; and
applying the timestamps to the resulting convergence surface with a user defined time interval to generate a probability matrix of association for the one or more additional locations.
7. The method of claim 6 further comprising the step of calculating an individual convergence zone C intersection matrix probability score for the one or more additional locations according to:
∑
l
=
1
l
R
ijl
min
(
∑
l
=
1
l
R
il
,
∑
l
=
1
l
R
j
l
)
i
≠
j
.
8. The method of claim 6 further comprising the step of calculating an overall network intersection matrix probability score for each device pair having associations in the one or more additional locations according to:
∑
k
=
1
k
C
k
∑
l
=
1
l
R
ijl
min
(
∑
k
=
1
k
C
k
∑
l
=
1
l
R
il
,
∑
k
=
1
k
C
k
∑
l
=
1
l
R
j
l
)
i
≠
j
.