Granted Patent B1

US 10,979,848 · App. 17/142,144 · Granted Apr 13, 2021

Inventors: Hari Palappetty (Bangalore, IN); Sumanth N (Bangalore, IN)

Assignee: Near Pte. Ltd.

H04W4/02G06N20/00H04L9/3247H04W64/003

Abstract

Claims (46)

1. A method for identifying an entity device using a device signature of the entity device and a location signature of a location, where the location signature is based on a database, said method comprising:

generating at least one device signature for the entity device based on a plurality of device attributes, connection attributes and user agent strings obtained from independently controlled data sources;

generating at least one location signature for the location based on a at least one of latitude-longitude pair and altitude, a shape or size of the location, and connection attributes of devices connecting from the location;

receiving, a plurality of location data streams from the entity device, wherein the plurality of location data streams comprises a timestamp data and at least one location attribute;

generating a cohort of device signatures for the location;

generating an indexed data stream for the location using a combination of the at least one location signature and the cohort of device signatures for the location;

building a machine learning model based on the indexed data stream for the location;

assigning a unique generated identifier for the entity device based on at least one of a set of rules or the machine learning model; and

identifying, the entity device using the unique generated identifier from a selected data stream associated with the entity device, wherein the selected data stream comprises a device information selected from the plurality of device attributes, connection attributes and user agent strings, and a location information of the entity device at a time, wherein the selected data stream does not include a device identifier of the entity device.

2. The method of claim 1 , wherein the database comprises an annotated point of interest database, and the location signature is determined from the annotated point of interest database.

3. The method of claim 1 , wherein a dynamic score of confidence is assigned to the unique generated identifier based on at least one of the set of rules or the machine learning model.

4. The method of claim 3 , wherein the dynamic score of confidence is validated by correlating a plurality of attributes of the at least one device signature and at least one location signature with a historical labelled data set.

5. The method of claim 1 , further comprising determining at least one attribute of the location by mapping a data stream associated with the location to the entity device.

6. The method of claim 1 , wherein the method further comprises combining a plurality of attributes to identify the entity device, wherein the plurality of attributes comprises device attributes and attributes of a user associated with the entity device.

7. The method of claim 1 , wherein a plurality of pre-computed weights is assigned to the independently controlled data sources to increase accuracy of the unique generated identifier for the entity device.

8. A unique generated identifier server that causes identifying an entity device using a device signature of the entity device and a location signature of a location, where the location signature is based on a database, said server comprising:

a processor; and

a memory that stores a set of instructions, which when executed by the processor, causes to perform:

generating at least one device signature for the entity device based on a plurality of device attributes, connection attributes and user agent strings obtained from independently controlled data sources;

generating at least one location signature for a location based on a at least one of latitude-longitude pair and altitude, a shape or size of the location, and connection attributes of devices connecting from the location;

receiving, a plurality of location data streams from the entity device, wherein the plurality of location data streams comprises a timestamp data and at least one location attribute;

generating a cohort of device signatures for the location;

generating an indexed data stream for the location using a combination of the at least one location signature and the cohort of device signatures for the location;

building a machine learning model based on the indexed data stream for the location;

assigning a unique generated identifier for the entity device based on at least one of a set of rules or the machine learning model; and

identifying, the entity device using the unique generated identifier from a selected data stream associated with the entity device, wherein the selected data stream comprises a device information selected from the plurality of device attributes, connection attributes and user agent strings, and a location information of the entity device at a time, wherein the selected data stream does not include a device identifier of the entity device.

9. The unique generated identifier server of claim 8 , wherein the database comprises an annotated point of interest database, and the location signature is determined from the annotated point of interest database.

10. The unique generated identifier server of claim 8 , wherein a dynamic score of confidence is assigned to the unique generated identifier based on at least one of the set of rules or the machine learning model.

11. The unique generated identifier server of claim 10 , wherein the dynamic score of confidence is validated by correlating a plurality of attributes of the at least one device signature and at least one location signature with a historical labelled data set.

12. The unique generated identifier server of claim 8 , wherein the unique generated identifier server determines at least one attribute of the location by mapping a data stream associated with the location to the entity device.

13. The unique generated identifier server of claim 8 , wherein the unique generated identifier server combines a plurality of attributes to identify the entity device, wherein the plurality of attributes comprises device attributes and attributes of a user associated with the entity device.

14. The unique generated identifier server of claim 8 , wherein a plurality of pre-computed weights is assigned to the independently controlled data sources to increase accuracy of the unique generated identifier.

15. A non-transitory computer readable storage medium storing a sequence of instructions, which when executed by a processor, causes identifying an entity device using a device signature of the entity device and a location signature of a location, where the location signature is based on a database, said sequence of instructions comprising:

generating at least one device signature for the entity device based on a plurality of device attributes, connection attributes and user agent strings obtained from independently controlled data sources;

generating at least one location signature for a location based on a at least one of latitude-longitude pair and altitude, a shape or size of the location, and connection attributes of devices connecting from the location;

receiving, a plurality of location data streams from the entity device, wherein the plurality of location data streams comprises a timestamp data and at least one location attribute;

generating a cohort of device signatures for the location;

generating an indexed data stream for the location using a combination of the at least one location signature and the cohort of device signatures for the location;

building a machine learning model based on the indexed data stream for the location;

assigning a unique generated identifier for the entity device based on at least one of a set of rules or the machine learning model; and

identifying, the entity device using the unique generated identifier from a selected data stream associated with the entity device, wherein the selected data stream comprises a device information selected from the plurality of device attributes, connection attributes and user agent strings, and a location information of the entity device at a time, wherein the selected data stream does not include a device identifier of the entity device.

16. The non-transitory computer readable storage medium storing a sequence of instructions of claim 15 , wherein the database comprises an annotated point of interest database, and the location signature is determined from the annotated point of interest database.

17. The non-transitory computer readable storage medium storing a sequence of instructions of claim 15 , wherein a dynamic score of confidence is assigned to the unique generated identifier based on at least one of the set of rules or the machine learning model.

18. The non-transitory computer readable storage medium storing a sequence of instructions of claim 17 , wherein the dynamic score of confidence is validated by correlating a plurality of attributes of the at least one device signature and at least one location signature with a historical labelled data set.

19. The non-transitory computer readable storage medium storing a sequence of instructions of claim 15 , which when executed by the said process, further combines a plurality of attributes to identify the entity device, wherein the plurality of attributes comprises device attributes and attributes of a user associated with the entity device.

20. The non-transitory computer readable storage medium storing a sequence of instructions of claim 15 , wherein a plurality of pre-computed weights is assigned to the independently controlled data sources to increase accuracy of the unique generated data stream that does not include device identifier of the entity device.

Assignments (10)

Cited By (8)

US 12,386,947 US 12,388,782 US 12,402,064 US 12,470,593 US 12,572,846 US 12,574,399 US 12,640,995 US 12,695,752