Action prediction and identification temporal user behavior

• US 8,412,665 B2
• Filed: 11/17/2010
• Issued: 04/02/2013
• Est. Priority Date: 11/17/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Wang, Gang, Chen, Weizhu, Chen, Zheng, Fan, Zhikai
• Primary Examiner(s): Holmes, Michael B
• Application Number:
  US12/947,871
• Publication Number:
  US 20120123993A1
• Time in Patent Office: 867 Days
• Field of Search: 706/52
• US Class Current: 706/52
• CPC Class Codes: G06F 16/3349 Reuse of stored results of previous queries
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/043 Distributed expert systems; Blackboards
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

User behavior modeling can include determining temporal- or time-based actions performed by various users. From the mined temporal-based user actions, future actions can be predicted. Certain implementations include providing information and/or services based on the predicted future actions. Some implementations, include providing relevant information, services, and/or goods regarding the predicted future action.

Claims

1. A method performed by one or more computing systems comprising:

determining a population of users performing online actions, wherein determining the population includes indexing actions of users of the population over a particular time period;

selecting from the population of users, a group of users that have performed particular time-based user actions that relate to a user behavior; and

predicting a subsequent user action based at least in part upon the user behavior.

2. The method of claim 1, wherein the determining the population is based at least in part on gathering an initial number of users based on a seed pattern.

3. The method of claim 1, wherein the determining the population of users further includes determining offline actions based on online searches and website visits.

4. The method of claim 1, wherein the selecting from the population is based at least in part on an ensemble of models that include models for particular user behavior actions.

5. The method of claim 1, wherein the particular time-based user actions are characterized by a set of websites that the user has visited.

6. The method of claim 5, wherein the actions are correlated with the website.

7. The method of claim 1 further comprising updating the population of users.

8. The method of claim 1 further comprising removing, from the group of users, a user that has performed the subsequent action.

9. The method of claim 1 further comprising providing an identification of the group of users to parties that provide information associated with the subsequent action.

10. A computing device comprising:

one or more processors;

memory coupled to the processor and comprising an action-prediction module that is configured to cause the one or more processors to;

select, from a population of users, a group of users that have performed particular user actions during a particular time period, wherein the user actions are associated with a behavior model; and

determine a subsequent action based at least in part upon the particular user actions.

11. The computing device of claim 10, wherein the population of users is taken from a user log.

12. The computing device of claim 10, wherein the behavior model is part of an ensemble of behavior models.

13. The computing device of claim 10, wherein the action-prediction module is configured to select the population of users based at least in part on online actions performed by the users of the population of users.

14. The computing device of claim 10, wherein the action-prediction module is further configured to selectively provide information regarding the group of uses to parties providing information regarding the subsequent action.

15. The computing device of claim 10, wherein the action-prediction module is further configured to determine which users of the group of users have performed the subsequent action and remove, from the group, the users that have performed the subsequent action.

16. The computing device of claim 10, wherein the action-prediction module is further configured to collect actions based at least in part on data provided by respective users on a browser toolbar.

17. A computer readable storage medium having computer readable instructions for performing steps comprising:

determining a population of online users performing particular time-based online and offline actions;

indexing the time-based actions of the online users;

grouping a set of users from the population of online users based at least in part on a behavior pattern of the time-based actions; and

predicting a subsequent action based at least in part on the behavior pattern.

18. The computer readable storage medium of claim 17, wherein the determining the population of online users includes tracking users search queries and web site visits.

19. The computer readable storage medium of claim 17, wherein the determining includes checking for a new user and predicting if the new user will perform the subsequent action.

20. The computer readable storage medium of claim 17, further comprising removing users that perform the subsequent action from the set of users.