User Behavior Monitoring On A Computerized Device

• US 20140377727A1
• Filed: 06/20/2013
• Published: 12/25/2014
• Est. Priority Date: 06/20/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Yom-Tov, Elad, Horvitz, Eric J., De Choudhury, Munmun, White, Ryen William, Counts, Scott J
• Granted Patent: US 9,427,185 B2
• Time in Patent Office: Days
• US Class Current: 434/236
• CPC Class Codes: A61B 2562/0204 Acoustic sensors
• CPC Class Codes: A61B 5/165 Evaluating the state of mind, e.g. depression, anxiety
• CPC Class Codes: A61B 5/4803 Speech analysis specially adapted for diagnostic purposes
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G09B 19/00 Teaching not covered by other main groups of this subclass teaching or practice apparatus for gun-aiming or gun-laying F41G3/26
• CPC Class Codes: G09B 5/12 different stations being capable of presenting different information simultaneously G09B5/14 takes precedence
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

The subject disclosure is directed towards monitoring user behavior on a computerized device for any deviation from normal or acceptable behavior that is likely to affect the user'"'"'s mental state. A prediction model corresponding to features of one or more mental states may be compared with features based upon current user behavior. If the user'"'"'s current behavior presents a mental state indicative of an uncharacteristic deviation from the normal or acceptable behavior, descriptive data associated with that mental state may be presented to the user in addition to a trusted individual, such as a health care professional.

Claims

1. In a computing environment, a method performed at least in part on at least one processor, comprising, recognizing a behavior change likely to affect a user, including accessing a predictive model corresponding to features of at least one mental state, comparing the predictive model with user interaction data to identify a mental state of the at least one mental state, and in response to the deviation, generating description data for the mental state.

2. The method of claim 1 further comprising using the predictive model to monitor for symptoms of a mental illness.

3. The method of claim 1 further comprising communicating an alert corresponding to the mental state.

4. The method of claim 1 further comprising configuring the predictive model based upon user preferences associated with prediction error data.

5. The method of claim 1 further comprising processing the user interaction data comprising at least one of search engine queries and social media postings, and extracting feature information from the user interaction data.

6. The method of claim 1 further comprising identifying a deviation from a behavior baseline that is indicative of a current mental state or a future mental state.

7. The method of claim 1 further comprising selecting a future mental state from an expected set of features that corresponds to a current deviation from a behavior baseline.

8. In a computing environment, a computerized device or a configuration of coupled computerized devices, comprising, a monitoring mechanism configured to process a prediction model corresponding to at least one mental state, wherein the monitoring mechanism is further configured to extract features from current user behavior using the prediction model, and if a psychological event is evident from the features, generate a secure alert corresponding the psychological event.

9. The device of claim 8, wherein the monitoring mechanism is further configured to communicate encrypted descriptive data corresponding to the psychological event.

10. The device of claim 8, wherein the monitoring mechanism is further configured to process historical data corresponding to the user behavior and adapt the prediction model to the user behavior.

11. The device of claim 8, wherein the monitoring mechanism is further configured to communicate a message comprising a link to a protected web page for displaying descriptive data corresponding to the psychological event.

12. The device of claim 8, wherein the monitoring mechanism is further configured to access a classifier built upon the prediction model and customize the classifier using user preference data.

13. The device of claim 8, wherein the monitoring mechanism is further configured to communicate feedback to a server for updating the prediction model.

14. The device of claim 8, wherein at least part of the monitoring mechanism is further configured to operate on a server.

15. The device of claim 8, wherein the monitoring mechanism is further configured to update the prediction model with feedback related to previous instances of psychological events.

16. One or more computer-readable media having computer-executable instructions, which when executed perform steps, comprising:

generating at least one classifier built from at least one prediction model associated with at least one mental state;

configuring the at least one classifier to monitor for a symptom of a mental illness; and

processing feedback from the computerized device to update the at least one prediction model.

17. The one or more computer-readable media of claim 16, wherein configuring the at least one classifier further comprises configuring the at least one classifier on a plurality of computerized devices corresponding to a group of patients.

18. The one or more computer-readable media of claim 16 having further computer-executable instructions comprising:

communicating a secure alert according to alert preference data.

19. The one or more computer-readable media of claim 16 having further computer-executable instructions comprising:

accessing training data corresponding to user behavior of at least one similar user; and

using the training data as an initial prediction model for generating the at least one classifier.

20. The one or more computer-readable media of claim 16 having further computer-executable instructions comprising:

adjusting the prediction model based upon an expert assessment.