Energy-efficient mobile advertising

• US 9,558,508 B2
• Filed: 03/15/2013
• Issued: 01/31/2017
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Nath, Suman Kumar, Riva, Oriana, Burger, Douglas Christopher, Mohan, Prashanth
• Primary Examiner(s): STOLTENBERG, DAVID J
• Application Number:
  US13/831,845
• Publication Number:
  US 20140279026A1
• Time in Patent Office: 1,418 Days
• US Class Current: 1/1
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G06Q 30/0254 based on statistics
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0267 Wireless devices
• CPC Class Codes: G06Q 30/0272 Period of advertisement exposure
• CPC Class Codes: H04L 67/01 Protocols

Abstract

Various technologies described herein pertain to prefetching advertisements for mobile advertising. A prediction model for estimating a number of advertisements that a mobile client is likely to request during an upcoming prediction time period can be employed. An estimated total amount of time of likely interaction with application(s) executed by the mobile client can be predicted; based upon such prediction, a number of advertisement slots likely to be available and a probability of each of the advertisement slots being available can be computed. Moreover, an ad server can allocate advertisements in a pending advertisement queue and/or disparate advertisements collected from an ad exchange to the mobile client based upon the number of advertisement slots likely to be available, the probability of each of the advertisements slots being available, and aggregated probabilities of the pending advertisements in the pending advertisement queue being displayed prior to corresponding deadlines for expiration.

Claims

1. A method performed by at least one processor, comprising:

predicting, utilizing the at least one processor, an estimated total amount of time of likely interaction with one or more applications executed by a mobile client during an upcoming prediction time period based upon features from past behavior data for the mobile client;

computing, utilizing the at least one processor, a number of advertisement slots likely to be available during the upcoming prediction time period based upon the estimated total amount of time of likely interaction with the one or more applications during the upcoming prediction time period; and

determining, utilizing the at least one processor, a probability of each of the advertisement slots being available during the upcoming prediction time period;

wherein prefetched advertisements are assigned to the mobile client for the upcoming prediction time period by an ad server based at least in part upon the number of advertisement slots likely to be available during the upcoming prediction time period and the probability of each of the advertisement slots being available during the upcoming prediction time period, and wherein the prefetched advertisements are retained in an advertisement cache of the mobile client and locally served for display on a display screen of the mobile client during execution of the one or more applications.

2. The method of claim 1, further comprising periodically invoking the predicting of the estimated total amount of time of likely interaction with the one or more applications during the upcoming prediction time period, the computing of the number of the advertisement slots likely to be available during the upcoming prediction time period, and the determining of the probability of each of the advertisement slots being available during the upcoming prediction time period.

3. The method of claim 1, further comprising invoking the predicting of the estimated total amount of time of likely interaction with the one or more applications during the upcoming prediction time period, the computing of the number of the advertisement slots likely to be available during the upcoming prediction time period, and the determining of the probability of each of the advertisement slots being available during the upcoming prediction time period in response to reduction of a number of the prefetched advertisements retained in the advertisement cache of the mobile client below a threshold number.

4. The method of claim 1 being executed by the mobile client, wherein the mobile client comprises the at least one processor, the method further comprising transmitting data indicative of the number of the advertisement slots likely to be available during the upcoming prediction time period and the probability of each of the advertisement slots being available during the upcoming prediction time period to the ad server.

5. The method of claim 4, further comprising opportunistically transmitting status information pertaining to the prefetched advertisements retained in the advertisement cache of the mobile client to the ad server when a radio of the mobile client is active for a disparate transmission, wherein the status information specifies a first subset of the prefetched advertisements retained in the advertisement cache of the mobile client that have been displayed on the display screen of the mobile client and a second subset of the prefetched advertisements retained in the advertisement cache of the mobile client that are unlikely to be displayed on the display screen of the mobile client prior to corresponding deadlines for expiration.

6. The method of claim 1, wherein the features from the past behavior data for the mobile client comprise at least one of a time of day during which a particular application was executed using the mobile client, a day of a week during which the particular application was executed using the mobile client, a length of time that the particular application was executed using the mobile client, a geographic location of the mobile client at which the particular application was executed using the mobile client, or a type of the particular application.

7. The method of claim 1, wherein the ad server assigns at least one of the prefetched advertisements to the mobile client and at least one disparate mobile client.

8. An ad server that allocates advertisements to mobile clients, comprising:

a data repository that retains a pending advertisement queue that comprises pending advertisements, wherein the pending advertisements in the pending advertisement queue are collected from an ad exchange and have yet to be displayed by one or more of the mobile clients, and wherein the pending advertisements in the pending advertisement queue each have a corresponding deadline for expiration;

memory that comprises computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including;

receiving a prefetch request from a given mobile client for a certain number of advertisements to be locally served during an upcoming prediction time period for display by the given mobile client during execution of one or more applications, wherein the mobile clients comprise the given mobile client;

computing an aggregated probability of each of the pending advertisements in the pending advertisement queue being displayed by at least one of the mobile clients prior to the corresponding deadline for expiration;

in response to the prefetch request, assigning at least one of one or more disparate advertisements collected from the ad exchange or one or more of the pending advertisements from the pending advertisement queue to the given mobile client based at least in part upon the aggregated probability of each of the pending advertisements in the pending advertisement queue being displayed prior to the corresponding deadline for expiration, a number of advertisement slots that the given mobile client is likely to have available during the upcoming prediction time period, and a probability of each of the advertisement slots of the given mobile client being available during the upcoming prediction time period; and

transmitting assigned advertisements to the given mobile client.

9. The ad server of claim 8, the memory further comprising computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

receiving, data indicative of the number of the advertisement slots that the given mobile client is likely to have available during the upcoming prediction time period and the probability of each of the advertisement slots of the given mobile client being available during the upcoming prediction time period from at least one of the given mobile client or a disparate server.

10. The ad server of claim 8, the memory further comprising computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

predicting an estimated total amount of time of likely interaction with the one or more applications executed by the given mobile client during the upcoming prediction time period based upon features from past behavior data for the given mobile client; and

computing the number of the advertisement slots that the given mobile client is likely to have available during the upcoming prediction time period and the probability of each of the advertisement slots of the given mobile client being available during the upcoming prediction time period based upon the estimated total amount of time of likely interaction with the one or more applications during the upcoming prediction time period.

11. The ad server of claim 8, the memory further comprising computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

collecting the one or more disparate advertisements from the ad exchange, wherein the one or more disparate advertisements are collected from the ad exchange for assignment to the given mobile client, wherein the ad exchange performs an auction prior to availability of one or more advertisement slots for display of the one or more disparate advertisements by the given mobile client; and

adding the one or more disparate advertisements to the pending advertisement queue.

12. The ad server of claim 8, the memory further comprising computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

receiving, via an opportunistic control channel, status information, the status information specifies;

a first subset of the pending advertisements that have been prefetched by the mobile clients and have been displayed on display screens of the mobile clients; and

a second subset of the pending advertisements that have been prefetched by the mobile clients and are unlikely to be displayed on the display screens of the mobile clients prior to the corresponding deadlines for expiration; and

adjusting assignments of the pending advertisements from the pending advertisement queue to the mobile clients based upon the status information.

13. The ad server of claim 8, the memory further comprising computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

transmitting an indication of a penalty to the ad exchange when one of the pending advertisements in the pending advertisement queue is not displayed by one or more of the mobile clients upon expiration of the corresponding deadline.

14. The ad server of claim 8, the memory further comprising computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

overbooking a particular pending advertisement from the pending advertisement queue by assigning the particular pending advertisement to two or more of the mobile clients upon approach of a corresponding deadline for expiration of the particular pending advertisement.

15. The ad server of claim 8, the memory further comprising computer executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

generating information that controls which of the assigned advertisements is to be displayed in each of the advertisement slots that the given mobile client is likely to have available during the upcoming prediction time period; and

transmitting the information to the given mobile client.

16. The ad server of claim 8, the memory further comprising computer executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

invalidating a particular pending advertisement that has been prefetched by the given mobile client and has yet to be displayed on a display screen of the given mobile client.

17. The ad server of claim 8, wherein the prefetch request from the given mobile client for the certain number of advertisements comprises a single bulk request for one or more advertisements.

18. The ad server of claim 8, wherein the prefetch request from the given mobile client for the certain number of advertisements comprises a plurality of separate requests, wherein at least one of the separate requests is for one advertisement.

19. A client device, comprising:

at least one processor; and

memory that comprises computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including;

predicting an estimated total amount of time of likely interaction with one or more applications executed by the client device during an upcoming prediction time period based upon features from past behavior data for the client device;

computing a number of slots for content units likely to be available during the upcoming prediction time period based upon the estimated total amount of time of likely interaction with the one or more applications during the upcoming prediction time period;

determining a probability of each of the slots being available during the upcoming prediction time period;

transmitting data indicative of the number of the slots likely to be available during the upcoming prediction time period and the probability of each of the slots being available during the upcoming prediction time period to a server;

receiving, from the server, prefetched content units assigned to the client device for the upcoming prediction time period, the prefetched content units being received responsive to the data indicative of the number of the slots likely to be available during the upcoming prediction time period and the probability of each of the slots being available during the upcoming prediction time period;

storing the prefetched content units in a cache of the client device; and

locally serving, from the cache, one or more of the prefetched content units for display on a display screen of the client device during execution of the one or more applications.

20. The client device of claim 19, the memory further comprising computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

opportunistically transmitting status information pertaining to the prefetched content units stored in the cache of the client device to the server when a radio of the client device is active for a disparate transmission, wherein the status information specifies;

a first subset of the prefetched content units stored in the cache of the client device that have been displayed on the display screen of the client device; and

a second subset of the prefetched content units stored in the cache of the client device that are unlikely to be displayed on the display screen of the client device prior to corresponding deadlines for expiration.

21. A method performed by a client device, comprising:

predicting an estimated total amount of time of likely interaction with one or more applications executed by the client device during an upcoming prediction time period based upon features from past behavior data for the client device;

computing a number of slots for content units likely to be available during the upcoming prediction time period based upon the estimated total amount of time of likely interaction with the one or more applications during the upcoming prediction time period;

determining a probability of each of the slots being available during the upcoming prediction time period;

transmitting, from the client device to a server, data indicative of the number of the slots likely to be available during the upcoming prediction time period and the probability of each of the slots being available during the upcoming prediction time period;

receiving, at the client device from the server, prefetched content units assigned to the client device for the upcoming prediction time period, the prefetched content units being received responsive to the data indicative of the number of the slots likely to be available during the upcoming prediction time period and the probability of each of the slots being available during the upcoming prediction time period;

storing the prefetched content units in a cache of the client device; and

locally serving, from the cache, one or more of the prefetched content units for display on a display screen of the client device during execution of the one or more applications.

22. The method of claim 21, further comprising:

opportunistically transmitting, from the client device to the server, status information pertaining to the prefetched content units stored in the cache of the client device when a radio of the client device is active for a disparate transmission, wherein the status information specifies;

a first subset of the prefetched content units stored in the cache of the client device that have been displayed on the display screen of the client device; and

a second subset of the prefetched content units stored in the cache of the client device that are unlikely to be displayed on the display screen of the client device prior to corresponding deadlines for expiration.

23. A system that allocates content units to client devices, comprising:

a data repository that retains a pending queue that comprises pending content units, wherein the pending content units in the pending queue are collected from a differing system and have yet to be displayed by one or more of the client devices, and wherein the pending content units in the pending queue each have a corresponding deadline for expiration;

at least one processor; and

memory that comprises computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including;

receiving a prefetch request from a given client device for a certain number of content units to be locally served during an upcoming prediction time period for display by the given client device during execution of one or more applications, wherein the client devices comprise the given client device;

computing an aggregated probability of each of the pending content units in the pending queue being displayed by at least one of the client devices prior to the corresponding deadline for expiration;

in response to the prefetch request, assigning at least one of one or more disparate content units collected from the differing system or one or more of the pending content units from the pending queue to the given client device based at least in part upon the aggregated probability of each of the pending content units in the pending queue being displayed prior to the corresponding deadline for expiration, a number of slots for content units that the given client device is likely to have available during the upcoming prediction time period, and a probability of each of the slots of the given client device being available during the upcoming prediction time period; and

transmitting assigned content units to the given client device.

24. The system of claim 23, the memory further comprising computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

predicting an estimated total amount of time of likely interaction with the one or more applications executed by the given client device during the upcoming prediction time period based upon features from past behavior data for the given client device; and

computing the number of the slots that the given client device is likely to have available during the upcoming prediction time period and the probability of each of the slots of the given client device being available during the upcoming prediction time period based upon the estimated total amount of time of likely interaction with the one or more applications during the upcoming prediction time period.

25. The system of claim 23, the memory further comprising computer-executable instructions that, when executed by the at least one processor, cause the at least one processor to perform acts including:

receiving, via an opportunistic control channel, status information, the status information specifies;

a first subset of the pending content units that have been prefetched by the client devices and have been displayed on display screens of the client devices; and

a second subset of the pending content units that have been prefetched by the client devices and are unlikely to be displayed on the display screens of the client devices prior to the corresponding deadlines for expiration; and

adjusting assignments of the pending content units from the pending queue to the client devices based upon the status information.

26. A method performed by at least one server, comprising:

receiving, at the at least one server, a prefetch request from a given client device for a certain number of content units to be locally served during an upcoming prediction time period for display by the given client device during execution of one or more applications;

computing aggregated probabilities of pending content units in a pending queue of the at least one server being displayed prior to corresponding deadlines for expiration, wherein the pending content units in the pending queue have yet to be displayed;

in response to the prefetch request, assigning one or more of the pending content units from the pending queue to the given client device based on;

the aggregated probabilities of the pending content units in the pending queue being displayed prior to the corresponding deadlines for expiration;

a number of slots for content units that the given client device is likely to have available during the upcoming prediction time period; and

a probability of each of the slots of the given client device being available during the upcoming prediction time period; and

transmitting, from the at least one server, assigned content units to the given client device.

27. The method of claim 26, further comprising:

receiving, at the at least one server, data indicative of the number of the slots that the given client device is likely to have available during the upcoming prediction time period and the probability of each of the slots of the given client device being available during the upcoming prediction time period.

28. The method of claim 26, further comprising:

predicting an estimated total amount of time of likely interaction with the one or more applications executed by the given client device during the upcoming prediction time period based upon features from past behavior data for the given client device; and

computing the number of the slots that the given client device is likely to have available during the upcoming prediction time period and the probability of each of the slots of the given client device being available during the upcoming prediction time period based upon the estimated total amount of time of likely interaction with the one or more applications during the upcoming prediction time period.

29. The method of claim 26, further comprising:

receiving, via an opportunistic control channel, status information, the status information specifies;

a first subset of the pending content units that have been prefetched and have been displayed; and

a second subset of the pending content units that have been prefetched and are unlikely to be displayed prior to the corresponding deadlines for expiration.

30. The method of claim 26, further comprising:

assigning a particular pending content unit to two or more client devices upon approach of a corresponding deadline for expiration of the particular pending content unit.

31. The method of claim 26, further comprising:

generating information that controls which of the assigned content units is to be displayed in each of the slots that the given client device is likely to have available during the upcoming prediction time period; and

transmitting the information to the given client device.