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