Prediction of user response actions to received data

• US 9,245,225 B2
• Filed: 08/13/2014
• Issued: 01/26/2016
• Est. Priority Date: 12/15/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Winn, John, Garcia, Fernando, Sundelin, Tore, Edelen, James
• Primary Examiner(s): Gaffin, Jeffrey A
• Assistant Examiner(s): BROWN JR, NATHAN H
• Application Number:
  US14/459,224
• Publication Number:
  US 20140351189A1
• Time in Patent Office: 531 Days
• Field of Search: 706/12, 706/16, 706/21
• US Class Current: 1/1
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• CPC Class Codes: G06F 2216/03 Data mining
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: G06Q 10/10 Office automation; Time management

Abstract

A system is provided for automatically predicting actions a user is likely to take in response to receiving data. The system may be configured to monitor and observe a user'"'"'s interactions with incoming data and to identify patterns of actions the user may take in response to the incoming data. The system may enable a trainer component and a classifier component to determine the probability a user may take a particular action and to make predictions of likely user actions based on the observations of the user and the identified pattern of the user'"'"'s actions. The system may also be configured to continuously observe the user'"'"'s actions to fine-tune and adjust the identified patterns of user actions and to update the probabilities of likely user actions in order increase the accuracy of the predicted user action in response to incoming data.

Claims

1. A method executed in a computing device to predict user response actions, the method comprising:

observing a user receiving incoming data from a data source;

processing the incoming data to extract contextual data associated with the incoming data, wherein the contextual data includes messages previously received in the incoming data from the user;

monitoring the incoming data;

tracking a first set of user actions to the incoming data;

analyzing the incoming data to identify a pattern to the first set; and

storing the pattern.

2. The method of claim 1, further comprising:

detecting new incoming data from another data source; and

predicting a second set of user actions to the new incoming data based on the pattern, a context of the new incoming data, and a content of the new incoming data.

3. The method of claim 1, further comprising:

detecting new incoming data from another data source; and

predicting a second set of user actions to the new incoming data based on a seed model.

4. The method of claim 3, further comprising:

utilizing the seed model to make predictions for the user, wherein the seed model is a pre-generated global knowledge model of a statistically significant number of sample users.

5. The method of claim 1, further comprising:

producing a list of previous user response actions gathered from the incoming data.

6. A system to predict user response actions, the system comprising:

a feature extractor component configured to detect incoming data for a user from a data source;

sensor components configured to detect user actions to the incoming data and report the user actions; and

an actions processor that is coupled to the feature extractor component, sensor components, a trainer component, a classifier component, and an accuracy calculator component, wherein the actions processor is configured to;

receive the incoming data and the users actions;

process the incoming data to extract contextual data associated with the incoming data, wherein the contextual data includes messages previously received in the incoming data from the user;

determine a pattern to the user actions through a process that includes one or more from a set of;

a combination operation, storage operation, and organization operation of the incoming data;

produce a user behavior database that includes features of the user actions; and

provide the user behavior database to the trainer component.

7. The system of claim 6, wherein the feature extractor component is further configured to:

process the incoming data to extract information associated with the incoming data, wherein the information includes a subject of the data and a recipient name.

8. The system of claim 6, wherein the actions processor is further configured to:

produce a list of previous user response actions gathered from the incoming data.

9. The system of claim 6, wherein the actions processor is further configured to:

include in the features one or more from a set of;

a content of the incoming data, a time the incoming data was sent, the user who sent the incoming data, and a location of the user when the incoming data was received.

10. The system of claim 6, wherein the trainer component is configured to:

receive the user behavior database from the actions processor; and

gather the incoming data from the user behavior database during a training period.

11. The system of claim 10, wherein the trainer component is further configured to:

synthesize the incoming data to determine an accurate pattern.

12. The system of claim 6, wherein the accuracy calculator is configured to:

compare predictions of the classifier component to the user actions of the user behavior database, wherein the classifier component utilizes a statistical model to make the predictions for the user actions based on the pattern.

13. The system of claim 12, wherein the accuracy calculator is further configured to:

measure an accuracy of the predications, wherein the accuracy calculator interacts with the trainer component to analyze the accuracy of the classifier component.

14. The system of claim 6, wherein the incoming data includes one or more of a subject of the incoming data, a recipient name of the incoming data, an urgency of the incoming data, a date of the incoming data, time information of the incoming data, a location of the user when the incoming data was received, and a language based content of the incoming data.

15. The system of claim 14, wherein the language based content of the incoming data includes one or more from a set of:

keywords, topics, and a number of times the user has replied to a message.

16. A non-transitory computer-readable storage device with instructions stored thereon to predict user response actions, the instructions comprising:

observing a user receiving incoming data from a data source;

processing the incoming data to extract contextual data associated with the incoming data, wherein the contextual data includes messages previously received in the incoming data from the user;

monitoring the incoming data;

tracking a first set of user actions to the incoming data;

analyzing the incoming data to identify a pattern to the first set;

storing the pattern;

detecting new incoming data from another data source; and

predicting a second set of user actions to the new incoming data based on the pattern, a context of the new incoming data, and a content of the new incoming data.

17. The non-transitory computer-readable storage device of claim 16, wherein the instructions further comprise:

detecting new incoming data from another data source; and

predicting a second set of user actions to the new incoming data based on a seed model.

18. The non-transitory computer-readable storage device of claim 17, wherein the instructions further comprise:

utilizing the seed model to make predictions for the user, wherein the seed model is a pre-generated global knowledge model of a statistically significant number of sample users.

19. The non-transitory computer-readable storage device of claim 16, wherein the instructions further comprise:

producing a list of previous user response actions gathered from the incoming data.