CLICK PREDICTION USING BIN COUNTING

• US 20120323677A1
• Filed: 06/20/2011
• Published: 12/20/2012
• Est. Priority Date: 06/20/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: CHELLAPILLA, KUMAR, BOTTOU, LEON, SIMARD, PATRICE Y., CHICKERING, DAVID MAX
• Granted Patent: US 9,104,960 B2
• Time in Patent Office: Days
• US Class Current: 705/14.45
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements

Abstract

Methods, systems, and computer-storage media having computer-usable instructions embodied thereon for calculating event probabilities are provided. The event may be a click probability. Event probabilities are calculated using a system optimized for runtime model accuracy with an operable learning algorithm. Bin counting techniques are used to calculate event probabilities based on a count of event occurrences and non-event occurrences. Linear parameters, such and counts of clicks and non-clicks, may also be used in the system to allow for runtime adjustments.

Claims

1. One or more computer storage media storing computer-useable instructions that, when used by one or more computing devices, cause the one or more computing devices to perform a method, the method comprising:

identifying a request to calculate an event probability, wherein the event probability indicates an expected fraction of times the event will occur;

associating information from the request with one or more feature groups, wherein a feature group is a classification of the information from the request, and wherein each of the one or more feature groups includes a plurality of bins;

associating a bin of the plurality of bins with the information from the request;

identifying counter information from at least one of an event counter or a non- event counter associated with the bin, wherein the event counter counts a number of event occurrences for the request and the non-event counter counts a number of non- event occurrences for the request;

training the event counter and the non-event counter using a linear-training algorithm; and

calculating, utilizing the counter information, the event probability.

2. The one or more computer storage media of claim 1, wherein the event is a click probability.

3. The one or more computer storage media of claim 2, wherein the information from the request includes one or more of a query identifier identifying the query, an advertisement identifier identifying the advertisement, an internet protocol address a match type, wherein the match type indicates the query:

ad pair is one of an exact match, a phrase match, a broad match, or a smart match.

4. The one or more computer storage media of claim 1, wherein the one or more feature groups is one of a simple feature, a dictionary feature, a semantic feature, or a cross feature.

5. The one or more computer storage media of claim 4, wherein the one or more feature groups is a function feature, wherein the function feature does not include linear trainable parameters.

6. The one or more computer storage media of claim 1, wherein associating the bin with the information from the request includes:

determining whether the information from the request has been previously identified, wherein the information from the request has been previously identified if it is associated with an existing bin;

based on a determination that the information from the request has been previously identified, associating the information from the request with the existing bin.

7. The one or more computer storage media of claim 1, wherein associating the bin with the information from the request includes:

determining whether the information from the request has been previously identified, wherein the information from the request has been previously identified if it is associated with an existing bin; and

based on a determination that the information from the request has not been previously identified, associating the information from the request with one of a new bin or a garbage bin.

8. The one or more computer storage media of claim 1, further comprising assigning a weight to the counter information.

9. The one or more computer storage media of claim 1, wherein the event counter is a click counter that counts a number of clicks of an advertisement and wherein the non-event counter is a non-click counts that counts a number of times the advertisement is presented that does not result in a click of the advertisement.

10. The one or more computer storage media of claim 1, wherein the counter information used to calculate the event probability is one of raw counts indicating event occurrences and non-event occurrences or regularized counter information, wherein regularized counter information includes a log of each of the event occurrence counts and non-event occurrence counts.

11. A calculating system for predicting event probabilities, comprising:

a computing device associated with one or more processors and one or more computer-readable storage media;

a data store coupled with the computing device; and

a predicting engine that

identifies a request to calculate a probability for an event;

associates information from the request with a bin that includes one or more counters, wherein the one or more counters count a number of event occurrences and a number of non-event occurrences;

identifies counter information from the one or more counters; and

calculates, using the counter information, the probability for the event.

12. The system of claim 11, wherein the event is a click probability for an advertisement.

13. The system of claim 12, wherein the counter information used to calculate the click probability is one of raw counts indicating the number of clicks and the number of non- clicks or regularized counter information, wherein regularized counter information includes a log of each of the click counts and non-click counts.

14. The system of claim 11, wherein the event predicting engine further trains the one or more counters using a linear-training algorithm.

15. The system of claim 11, wherein the predicting engine determines whether the information from the request has been previously identified, wherein the information from the request has been previously identified if it is associated with an existing bin, and wherein the information from the request that has been previously identified is associated with the existing bin and information from the request that has not been previously identified is associated with one of a new bin or a garbage bin.

16. The system of claim 11, wherein the event is a social network activity including adding a friend in a social networking environment.

17. One or more computer storage media storing computer-useable instructions that, when used by one or more computing devices, cause the one or more computing devices to perform a method, the method comprising:

identifying a request to calculate a click probability for an advertisement, wherein the click probability indicates an expected number of times the advertisement will be selected by a user with presented in combination with a query;

associating information from the request with a bin, wherein the bin corresponds with the information from the request, and wherein the bin includes both a click counter and a non-click counter;

identifying counter information from at least one of the click counter or the non- click counter, wherein the click counter counts a number of clicks of the advertisement and the non-click counter counts a number of impressions of the advertisement that are not clicked;

calculating, utilizing the counter information, the click probability for the advertisement;

identifying a traffic attribute for the click probability, wherein the traffic attribute is a specified time interval;

updating the counter information based on the traffic attribute such that the updated counter information includes only clicks and non-clicks for the specified time interval; and

calculating, utilizing the updated counter information, an updated click probability for the advertisement based on the specified time interval.

18. The one or more computer storage media of claim 17, wherein the time interval includes a day of a week, a week, a month, a day, minutes, hours, or a combination thereof.

19. The one or more computer storage media of claim 17, wherein the traffic attribute further includes a user characteristic, wherein the user characteristic includes a gender of the user, a location of the user, a user profile associated with the user, or a combination thereof.

20. The one or more computer storage media of claim 17, wherein the information from the request includes one or more of a query identifier identifying a query associated with the advertisement, an advertisement identifier identifying the advertisement, or an internet protocol address.