Determining targeting information based on a predictive targeting model
A targeting system based on a predictive targeting model based on observed behavioral data including visit data, user profile and/or survey data, and geographic features associated with a geographic region. The predictive targeting model analyzes the observed behavioral data and the geographic features data to predict conversion rates for every cell in a square grid of predefined size on the geographic region. The conversion rate of a cell indicates a likelihood that any random user in that cell will perform a targeted behavior.
1. A method for use by at least one data processing device, the method comprising:
receiving targeting criteria, including a targeted behavior and a geographic region;
segmenting the geographic region using a grid into cells, wherein each cell has a cell identifier;
receiving behavioral information associated with multiple users, wherein the behavioral information includes time-stamped place visit data corresponding to visits to places by the multiple users;
calculating a behavior match metric for one or more cells based on the behavioral information;
receiving feature data for one or more cells;
labeling the feature data for the one or more cells using the behavior match metric for the corresponding cell to obtain labeled feature data;
training a model for predicting a conversion rate of each cell based on a set of the labeled feature data, wherein the conversion rate provides a probability of a user in a cell performing the targeted behavior;
applying the model to the feature data to predict the conversion rate of each cell; and
presenting targeting information based on the conversion rates of the cells to a client device.
2. The method of claim 1 , wherein the targeting information comprises latitude/longitude coordinates of locations having one of:
greater than or equal to a specified conversion rate; or
users identified as likely engage in targeted behavior.
3. The method of claim 1 , wherein the behavioral information is projected onto the one or more cells.
4. The method of claim 1 , wherein the behavioral information comprises one of survey data or demographic profile data associated with users.
5. The method of claim 1 , wherein training the model comprises:
training a classifier using the set of labeled feature data to predict a visit probability that provides an indication of a likelihood that the user in the cell would perform the targeted behavior; and
training a statistical model using the visit probability.
6. The method of claim 5 , wherein training the statistical model further comprises training the statistical model using an observed conversion rate aggregated across a plurality of user in the cell to predict the conversion rate for the cell.
7. The method of 6 , wherein applying the model comprises:
applying the classifier on the feature data to predict a visit probability; and
applying the statistical model on the visit probability to predict the conversion rate.
8. The method of claim 1 , wherein the targeted behavior comprises on or more of:
visiting a place;
signing up for an event; or
performing an activity.
9. A non-transitory computer-readable medium storing computer-executable instructions that, when executed by at least one processor, perform a method comprising:
receiving targeting criteria, including a targeted behavior and a geographic region; segmenting the geographic region using a grid into cells, wherein each cell has a cell identifier;
receiving behavioral information associated with multiple users, wherein the behavioral information includes time-stamped place visit data corresponding to visits to places by the multiple users;
calculating a behavior match metric for one or more cells based on the behavioral information;
receiving feature data for one or more cells;
labeling the feature data for the one or more cells using the behavior match metric for the corresponding cell to obtain labeled feature data;
training a model for predicting a conversion rate of each cell based on a set of the labeled feature data, wherein the conversion rate provides a probability of a user in a cell performing the targeted behavior;
applying the model to the feature data to predict the conversion rate of each cell; and
presenting identifying targeting information based on the conversion rates of the cells to a client device.
10. The non-transitory computer-readable medium of claim 9 , wherein the targeted behavior comprises on or more of:
visiting a place;
signing up for an event; or
performing an activity.
11. The non-transitory computer-readable medium of claim 9 , wherein the targeting information comprises latitude/longitude coordinates of locations having one of:
greater than or equal to a specified conversion rate; or
users identified as likely engage in targeted behavior.
12. The non-transitory computer-readable medium of claim 9 , wherein the behavioral information is projected onto the one or more cells.
13. The non-transitory computer-readable medium of claim 9 , wherein the behavioral information comprises one of survey data or demographic profile data associated with users.
14. The non-transitory computer-readable medium of claim 9 , wherein training the model comprises:
training a classifier using the set of labeled feature data to predict a visit probability that provides an indication of a likelihood that the user in the cell would perform the targeted behavior; and
training a statistical model using the visit probability.
15. The non-transitory computer-readable medium of claim 14 , wherein training the statistical model further comprises training the statistical model using an observed conversion rate aggregated across a plurality of user in the cell to predict the conversion rate for the cell.
16. A system comprising:
at least one processor; and
memory encoding computer-executable instructions that, when executed by the at least one processor, perform a method comprising:
receiving targeting criteria, including a targeted behavior and a geographic region; segmenting the geographic region using a grid into cells, wherein each cell has a cell identifier;
receiving behavioral information associated with multiple users, wherein the behavioral information includes time-stamped place visit data corresponding to visits to places by the multiple users;
calculating a behavior match metric for one or more cells based on the behavioral information;
receiving feature data for one or more cells;
labeling the feature data for the one or more cells using the behavior match metric for the corresponding cell to obtain labeled feature data;
training a model for predicting a conversion rate of each cell based on a set of the labeled feature data, wherein the conversion rate provides a probability of a user in a cell performing the targeted behavior;
applying the model to the feature data to predict the conversion rate of each cell; and
presenting identifying targeting information based on the conversion rates of the cells to a client device.
17. The system of claim 16 , wherein training the model comprises:
training a classifier using the set of labeled feature data to predict a visit probability that provides an indication of a likelihood that the user in the cell would perform the targeted behavior; and
training a statistical model using the visit probability.
18. The system of claim 17 , wherein training the statistical model further comprises training the statistical model using an observed conversion rate aggregated across a plurality of user in the cell to predict the conversion rate for the cell.
19. The system of 18 , wherein applying the model comprises:
applying the classifier on the feature data to predict a visit probability; and
applying the statistical model on the visit probability to predict the conversion rate.
20. The system of claim 16 , wherein the targeted behavior comprises on or more of:
visiting a place;
signing up for an event; or
performing an activity.