LEVERAGING PATTERNS IN GEOGRAPHIC POSITIONS OF MOBILE DEVICES AT A LOCALE
Embodiments are disclosed for a method that may include identifying a mobile device based on events including interactions between the mobile device and one or more network devices on a network at a locale, generating geographic position patterns for the mobile device using the interactions, determining a characteristic of the mobile device based on the geographic position patterns, and assigning the mobile device to a cluster sharing the characteristic.
1 . A method, comprising:
identifying a mobile device based on a plurality of events comprising a plurality of interactions between the mobile device and at least one network device on a network at a locale;
generating a plurality of geographic position patterns for the mobile device using the plurality of interactions;
determining a characteristic of the mobile device based on the plurality of geographic position patterns; and
assigning the mobile device to a cluster sharing the characteristic.
2 . The method of claim 1 , wherein the characteristic is further based on a sales transaction obtained from a point-of-sale (POS) system at the locale.
3 . The method of claim 1 , wherein the characteristic is further based on data obtained from a sensor at the locale.
4 . The method of claim 1 , further comprising:
identifying, as an anomaly, a first geographic position pattern of the plurality of geographic position patterns; and
excluding, in response to identifying the first geographic position pattern as an anomaly, the first geographic position pattern from the plurality of geographic position patterns.
5 . The method of claim 1 , further comprising:
obtaining user data corresponding to a user from an application executing on the mobile device;
correlating the plurality of geographic position patterns with the user data; and
assigning the mobile device to the user in response to the correlating.
6 . The method of claim 1 , further comprising:
correlating the characteristic with a purchasing preference for a product;
identifying, in response to correlating the characteristic, a promotion for the product targeting the cluster; and
predicting an impact of the promotion on a metric of the locale.
7 . The method of claim 1 , further comprising:
obtaining inventory data from an inventory system;
identifying, using the inventory data, a product with expiring inventory;
correlating, in response to identifying the product, the characteristic with a purchasing preference for the product; and
transmitting a promotion for the product to the mobile device.
8 . The method of claim 1 , further comprising:
obtaining user data corresponding to a user from a mobile application executing on the mobile device;
correlating the plurality of geographic position patterns with the user data;
assigning the mobile device to the user in response to the correlating; and
determining, using the plurality of geographic position patterns, an average number of days between visits to the locale by the user,
wherein the characteristic is further based on the average number of days between visits.
9 . The method of claim 1 , further comprising:
determining, using the plurality of geographic position patterns, an average number of days between visits to the locale by the mobile device; and
transmitting a promotion for a product to the mobile device in response to a current number of days following the most recent visit to the locale by the mobile device being within a threshold of the average number of days between visits, wherein the characteristic indicates a purchasing preference for the product.
10 . The method of claim 1 , further comprising:
obtaining user data corresponding to a user from a mobile application executing on the mobile device;
correlating the plurality of geographic position patterns with the user data;
assigning the mobile device to the user in response to the correlating; and
determining, using the plurality of geographic position patterns, an average amount of time spent at the locale by the user,
wherein the characteristic is further based on the average amount of time spent.
11 . A computer system, comprising:
a data store comprising a plurality of events comprising a plurality of interactions between a mobile device and at least one network device on a network at a locale; and
circuitry configured to:
identify the mobile device based on the plurality of events;
generate a plurality of geographic position patterns for the mobile device using the plurality of interactions;
determine a characteristic of the mobile device based on the plurality of geographic position patterns; and
assign the mobile device to a cluster sharing the characteristic.
12 . The system of claim 11 , wherein the characteristic is further based on a sales transaction obtained from a point-of-sale (POS) system at the locale.
13 . The system of claim 11 , wherein the characteristic is further based on data obtained from a sensor at the locale.
14 . The system of claim 11 , wherein the circuitry is further configured to:
identify, as an anomaly, a first geographic position pattern of the plurality of geographic position patterns; and
exclude, in response to identifying the first geographic position pattern as an anomaly, the first geographic position pattern from the plurality of geographic position patterns.
15 . The system of claim 11 , wherein the circuitry is further configured to:
obtain user data corresponding to a user from an application executing on the mobile device;
correlate the plurality of geographic position patterns with the user data; and
assign the mobile device to the user in response to the correlating.
16 . The system of claim 11 , wherein the circuitry is further configured to:
correlate the characteristic with a purchasing preference for a product;
identify, in response to correlating the characteristic, a promotion for the product targeting the cluster; and
predict an impact of the promotion on a metric of the locale.
17 . The system of claim 11 , wherein the circuitry is further configured to:
obtain inventory data from an inventory system;
identify, using the inventory data, a product with expiring inventory;
correlate, in response to identifying the product, the characteristic with a purchasing preference for the product; and
transmit a promotion for the product to the mobile device.
18 . The system of claim 11 , wherein the circuitry is further configured to:
obtain user data corresponding to a user from a mobile application executing on the mobile device;
correlate the plurality of geographic position patterns with the user data;
assign the mobile device to the user in response to the correlating; and
determine, using the plurality of geographic position patterns, an average number of days between visits to the locale by the user,
wherein the characteristic is further based on the average number of days between visits.
19 . The system of claim 11 , wherein the circuitry is further configured to:
determine, using the plurality of geographic position patterns, an average number of days between visits to the locale by the mobile device; and
transmit a promotion for a product to the mobile device in response to a current number of days following the most recent visit to the locale by the mobile device being within a threshold of the average number of days between visits, wherein the characteristic indicates a purchasing preference for the product.
20 . The system of claim 11 , wherein the circuitry is further configured to:
obtain user data corresponding to a user from a mobile application executing on the mobile device;
correlate the plurality of geographic position patterns with the user data;
assign the mobile device to the user in response to the correlating; and
determine, using the plurality of geographic position patterns, an average amount of time spent at the locale by the user,
wherein the characteristic is further based on the average amount of time spent.
21 . A non-transitory computer-readable medium comprising instructions, execution of which in a computer system causes the computer system to:
identify a mobile device based on a plurality of events comprising a plurality of interactions between the mobile device and at least one network device on a network at a locale;
generate a plurality of geographic position patterns for the mobile device using the plurality of interactions;
determine a characteristic of the mobile device based on the plurality of geographic position patterns; and
assign the mobile device to a cluster sharing the characteristic.
22 . The non-transitory computer-readable medium of claim 21 , wherein the characteristic is further based on a sales transaction obtained from a point-of-sale (POS) system at the locale.
23 . The non-transitory computer-readable medium of claim 21 , wherein the characteristic is further based on data obtained from a sensor at the locale.
24 . The non-transitory computer-readable medium of claim 21 , wherein the instructions, upon execution, further cause the computer system to:
identify, as an anomaly, a first geographic position pattern of the plurality of geographic position patterns; and
exclude, in response to identifying the first geographic position pattern as an anomaly, the first geographic position pattern from the plurality of geographic position patterns.
25 . The non-transitory computer-readable medium of claim 21 , wherein the instructions, upon execution, further cause the computer system to:
obtain user data corresponding to a user from an application executing on the mobile device;
correlate the plurality of geographic position patterns with the user data; and
assign the mobile device to the user in response to the correlating.
26 . The non-transitory computer-readable medium of claim 21 , wherein the instructions, upon execution, further cause the computer system to:
correlate the characteristic with a purchasing preference for a product;
identify, in response to correlating the characteristic, a promotion for the product targeting the cluster; and
predict an impact of the promotion on a metric of the locale.
27 . The non-transitory computer-readable medium of claim 21 , wherein the instructions, upon execution, further cause the computer system to:
obtain inventory data from an inventory system;
identify, using the inventory data, a product with expiring inventory;
correlate, in response to identifying the product, the characteristic with a purchasing preference for the product; and
transmit a promotion for the product to the mobile device.
28 . The non-transitory computer-readable medium of claim 21 , wherein the instructions, upon execution, further cause the computer system to:
obtain user data corresponding to a user from a mobile application executing on the mobile device;
correlate the plurality of geographic position patterns with the user data;
assign the mobile device to the user in response to the correlating; and
determine, using the plurality of geographic position patterns, an average number of days between visits to the locale by the user,
wherein the characteristic is further based on the average number of days between visits.
29 . The non-transitory computer-readable medium of claim 21 , wherein the instructions, upon execution, further cause the computer system to:
determine, using the plurality of geographic position patterns, an average number of days between visits to the locale by the mobile device; and
transmit a promotion for a product to the mobile device in response to a current number of days following the most recent visit to the locale by the mobile device being within a threshold of the average number of days between visits, wherein the characteristic indicates a purchasing preference for the product.
30 . The non-transitory computer-readable medium of claim 21 , wherein the instructions, upon execution, further cause the computer system to:
obtain user data corresponding to a user from a mobile application executing on the mobile device;
correlate the plurality of geographic position patterns with the user data;
assign the mobile device to the user in response to the correlating; and
determine, using the plurality of geographic position patterns, an average amount of time spent at the locale by the user,
wherein the characteristic is further based on the average amount of time spent.