Predictive geo-temporal advertisement targeting

• US 8,265,995 B2
• Filed: 11/14/2011
• Issued: 09/11/2012
• Est. Priority Date: 09/26/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Blegen, Darrell
• Primary Examiner(s): Dagnew, Saba
• Assistant Examiner(s): Stibley, Michael
• Application Number:
  US13/295,944
• Publication Number:
  US 20120059721A1
• Time in Patent Office: 302 Days
• Field of Search: 705/14.58, 705/14.64
• US Class Current: 705/14.58
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0261 based on user location
• CPC Class Codes: G06Q 30/0267 Wireless devices

Abstract

A method, system, and medium are provided for targeting advertisements to users of mobile devices based on geo-temporal models. Time-stamped location information is collected for a mobile device and a dynamic geo-temporal model is constructed and updated when new data is collected according to a sample design. Using the geo-temporal model, device location and instances of user interaction with the device can be predicted, and advertisements can be provided based on the predicted location. Advertisements can be cached on the mobile device for later presentation, and the sample design can be updated to improve efficiency and accuracy in the modeling system.

Claims

1. One or more computer-storage media having computer-executable instructions embodied thereon for performing a method of targeting advertising content to a user of a mobile device based on predicted device location, the method comprising:

collecting time-stamped device location data corresponding to a plurality of instances of user interaction with the mobile device;

updating a dynamic geo-temporal model that represents a hypothesized relationship between user interaction with the mobile device and the location of the mobile device when the user interacts with the mobile device;

specifying a first time period;

predicting, before the first time period, that the user will interact with the mobile device during the first time period, wherein predicting comprises referencing the dynamic geo-temporal model;

predicting, before the first time period, that the mobile device will be at a first location during the first time period; and

providing, before the first time period, a selected advertisement to the mobile device such that the advertisement can be stored in a cache on the mobile device and presented to the user during the first time period, wherein the advertisement is selected based on the first location.

2. The media of claim 1, wherein providing the advertisement further comprises including an indication that the advertisement should be presented during the first time period when the user interacts with the mobile device.

3. The media of claim 1, wherein the dynamic geo-temporal model comprises at least one probability density function associated with device location data.

4. The media of claim 1, wherein updating the geo-temporal model comprises modifying one or more parameters associated with the model.

5. The media of claim 1, wherein updating the geo-temporal model comprises discarding data collected before a specified time period.

6. The media of claim 1, wherein the advertisement is further selected based on identifying an attribute, associated with the advertisement, that indicates that the advertisement should be presented to a user when the user interacts with the mobile device at the first location.

7. The media of claim 6, wherein the first location comprises a designated geographic region.

8. The media of claim 1, wherein the advertisement is further selected based on

determining available storage space associated with a cache that is maintained on the mobile device;

and

selecting an advertisement having an attribute that indicates an amount of memory required to store the advertisement, wherein the amount of memory required to store the advertisement is less than the determined available storage space associated with the cache.

9. The media of claim 1, wherein collecting time-stamped device location data comprises determining location information associated with the mobile device during each of the plurality of instances of user interaction with the mobile device.

10. The media of claim 9, wherein location information comprises addressing information.

11. The media of claim 9, wherein location information comprises a designated geographic location.

12. The media of claim 1, wherein providing the advertisement to the mobile device further comprises providing an executable script to the mobile device, wherein the executable script is configured to cause the advertisement to be presented to the user during the first time period.

13. One or more computer-storage media having computer-executable instructions embodied thereon for performing a method of targeting advertising content to a user of a mobile device based on predicted device location, the method comprising:

updating a dynamic geo-temporal model comprising a representation of a plurality of locations of the mobile device during a plurality of time periods, wherein updating the dynamic geo-temporal model comprises incorporating time-stamped device location data collected according to a dynamically updateable sample design;

specifying a first time period in the future;

predicting, before the first time period, a location of the device during the first time period using the dynamic geo-temporal model;

selecting an advertisement to provide to the device based on the predicted location;

providing the advertisement to the device such that the advertisement can be presented to the user during the first time period;

determining that the sample design should be updated; and

updating the sample design.

14. The media of claim 13, wherein the dynamic geo-temporal model comprises at least one probability density function associated with device location data.

15. The media of claim 13, wherein selecting an advertisement comprises identifying an advertisement having an associated attribute that indicates that the advertisement should be presented to a user when the user interacts with the mobile device at the predicted location.

16. The media of claim 15, wherein the predicted location comprises a designated geographic region.

17. The media of claim 15, wherein selecting the advertisement further comprises:

determining available storage space associated with a cache that is maintained on the mobile device; and

selecting an advertisement having an attribute that indicates an amount of memory required to store the advertisement, wherein the amount of memory required to store the advertisement is less than the determined available storage space associated with the cache.

18. The media of claim 17, wherein providing the advertisement such that the advertisement can be presented to the user during the first time period comprises providing the advertisement to the mobile device such that the mobile device stores the advertisement in the cache.

19. The media of claim 18, wherein providing the advertisement further comprises including an indication that the advertisement should be presented during the first time period when the user interacts with the mobile device.

20. One or more computer-storage media having computer-executable instructions embodied thereon for performing a method of targeting advertising content to a user of a mobile device based on predicted device location, the method comprising:

referencing a dynamically updateable sample design for identifying a plurality of sampling time periods during which to collect location information associated with the mobile device;

detecting an instance of user interaction with the mobile device during each of a subset of the plurality of sampling time periods;

collecting time-stamped device location data associated with the mobile device in response to detecting each instance of user interaction with the device;

updating a dynamic geo-temporal model from the collected time-stamped device location data;

specifying a first time period in the future;

based on the dynamic geo-temporal model, determining, before the first time period,

(1) a first probability that user interaction with the mobile device will occur during the first time period; and

(2) a second probability that the mobile device will be at a first location during the first time period;

providing a selected advertisement to the mobile device such that the advertisement is capable of being presented to the user during the first time period, wherein the advertisement is selected based on the first location, the first probability, and the second probability.