Methods, systems, and media for determining location information from real-time bid requests

• US 9,014,717 B1
• Filed: 03/14/2013
• Issued: 04/21/2015
• Est. Priority Date: 04/16/2012
• Status: Active Grant

• Current Assignee: Dstillery, Inc.
• Original Assignee: Dstillery, Inc.
• Inventors: Provost, Foster J., Moores, Lauren S., Eliassi-Rad, Tina
• Primary Examiner(s): Jaroenchonwanit, Bunjob
• Application Number:
  US13/830,495
• Time in Patent Office: 768 Days
• Field of Search: 705/14.64, 705/14.67, 705/26.1, 705/14.49, 455/404.2, 455456-457, 455/517, 709/206
• US Class Current: 455/456.1
• CPC Class Codes: H04W 4/029 Location-based management or tracking services
• CPC Class Codes: H04W 64/006 with additional information processing, e.g. for direction or speed determination
• CPC Class Codes: H04W 8/005 Discovery of network devices, e.g. terminals
• CPC Class Codes: H04W 8/02 Processing of mobility data, e.g. registration information at HLR [Home Location Register] or VLR [Visitor Location Register]; Transfer of mobility data, e.g. between HLR, VLR or external networks

Abstract

Methods, systems, and media for determining location information from real-time bid requests are provided. In some implementations, a method for determining device locations is provided, the method comprising: receiving a real-time data stream that includes location proxies corresponding to devices; generating a movement graph of interconnected nodes and edges, wherein each node represents one of the location proxies and wherein each edge represents movement information between the location proxies; calculating, for a node in the movement graph, location information by applying a predictive model having weights based on the movement information; and assigning, for the node in the movement graph, a census-based identifier to the calculated location information. The census-based identifier can be used, for example, to supplement or enrich bid requests within the real-time data stream.

Claims

1. A method for determining device locations, the method comprising:

receiving, using a hardware processor, real-time bid requests;

determining, from IP addresses associated with the real-time bid requests, mobile IP addresses and non-mobile IP addresses based on a radius parameter;

monitoring movement information between the mobile IP addresses and the non-mobile IP addresses;

generating a movement graph of interconnected nodes and edges, wherein each node represents one of the mobile IP addresses and the non-mobile IP addresses and wherein each edge represents the movement information between nodes;

calculating, for a node in the movement graph, location information by applying a predictive model having weights based on the movement information;

assigning, for the node in the movement graph, a census block group information to the calculated location information; and

supplementing the real-time bid request corresponding to the node with the census block group information.

2. The method of claim 1, wherein each real-time bid request is associated with a hashed public IP address.

3. The method of claim 1, further comprising anonymizing the IP addresses associated with the real-time bid requests by applying one or more hash functions.

4. The method of claim 1, wherein each of the mobile IP addresses in the movement graph is associated with a time stamp.

5. The method of claim 1, wherein the movement information includes at least one of:

movement between a non-mobile IP address and another non-mobile IP address, movement between a mobile IP address and another non-mobile IP address, and movement between the non-mobile IP address and the mobile IP address.

6. The method of claim 1, wherein the movement information includes the number of movements between two nodes and the inter-arrival times between two nodes.

7. The method of claim 6, wherein the predictive model is a weighted-vote relational neighbor classifier with the number of movements and the inter-arrival times as the weights.

8. The method of claim 1, wherein the calculated location information includes inferring latitude and longitude information for the node without actual location information.

9. The method of claim 1, wherein the census block group is assigned from a plurality of census block groups by:

generating a plurality of centroids for each of the plurality of census block groups, wherein a centroid describes the census block group;

determining, for a centroid, a distance between the calculated location information and the centroid;

comparing the distance with a radius of the census block group corresponding to the centroid; and

determining, for the plurality of centroids, a census block group identifier based on the comparison.

10. The method of claim 9, wherein the comparison further comprises a ratio of the distance to the radius of the census block group and wherein the census block group identifier corresponds to the census block group that provided the ratio having a minimum value.

11. A system for determining device locations, the system comprising:

at least one hardware processor that is configured to;

receive real-time bid requests;

determine, from IP addresses associated with the real-time bid requests, mobile IP addresses and non-mobile IP addresses based on a radius parameter;

monitor movement information between the mobile IP addresses and the non-mobile IP addresses;

generate a movement graph of interconnected nodes and edges, wherein each node represents one of the mobile IP addresses and the non-mobile LP addresses and wherein each edge represents the movement information between nodes;

calculate, for a node in the movement graph, location information by applying a predictive model having weights based on the movement information;

assign, for the node in the movement graph, a census block group information to the calculated location information; and

supplement the real-time bid request corresponding to the node with the census block group information.

12. The system of claim 11, wherein each real-time bid request is associated with a hashed public IP address.

13. The system of claim 11, further comprising anonymizing the IP addresses associated with the real-time bid requests by applying one or more hash functions.

14. The system of claim 11, wherein each of the mobile IP addresses in the movement graph is associated with a time stamp.

15. The system of claim 11, wherein the movement information includes a movement between a first non-mobile IP address and a second non-mobile IP address.

16. The system of claim 11, wherein the movement information includes the number of movements between two nodes and the inter-arrival times between two nodes.

17. The system of claim 16, wherein the predictive model is a weighted-vote relational neighbor classifier with the number of movements and the inter-arrival times as the weights.

18. The system of claim 11, wherein the calculated location information includes inferring latitude and longitude information for the node without actual location information.

19. The system of claim 11, wherein the processor is further configured to:

generating a plurality of centroids for each of the plurality of census block groups, wherein a centroid describes the census block group;

determining, for a centroid, a distance between the calculated location information and the centroid;

comparing the distance with a radius of the census block group corresponding to the centroid; and

determining, for the plurality of centroids, a census block group identifier based on the comparison.

20. The system of claim 19, wherein the processor is further configured to:

determine a ratio of the distance to the radius of the census block group; and

determine the census block group identifier corresponding to the census block group that provided the ratio having a minimum value.

21. A method for determining device locations, the method comprising:

receiving, using a hardware processor, a real-time data stream that includes location proxies corresponding to devices;

generating a movement graph of interconnected nodes and edges, wherein each node represents one of the location proxies and wherein each edge represents movement information between the location proxies;

calculating, for a node in the movement graph, location information by applying a predictive model having weights based on the movement information; and

assigning, for the node in the movement graph, a census-based identifier to the calculated location information.