BUILDING AND USING PREDICTIVE MODELS OF CURRENT AND FUTURE SURPRISES

• US 20110071964A1
• Filed: 04/03/2009
• Published: 03/24/2011
• Est. Priority Date: 11/16/2004
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Horvitz, Eric J.
• Granted Patent: US 8,706,651 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 18/2411 based on the proximity to a decision surface, e.g. support vector machines
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G08G 1/0104 Measuring and analyzing of parameters relative to traffic conditions

Abstract

Methods are described for identifying events that would be considered surprising by people and identifying how and when to transmit information to a user about situations that they would likely find surprising. Additionally, the methods of identifying surprising situations can be used to build a case library of surprising events, joined with a set of observations before the surprising events occurred. Statistical machine learning methods can be applied with data from the case library to build models that can predict when a user will likely be surprised at future times. One or more models of context-sensitive expectations of people, a view of the current world, and methods for recording streams or events before surprises occur, and for building predictive models from a case library of surprises and such historical observations can be employed. The models of current and future surprises can be coupled with display and alerting machinery.

Claims

1. A system that predicts and outputs events identified as being surprising to people, comprising:

an interface component that receives contextual data; and

a predictive model component that utilizes the contextual data to predict an event and outputs the prediction if the prediction corresponds to one or more definitions of surprise.

2. The system of claim 1, the predictive model component at least one of contemplates historical data when generating the prediction;

calculates a probability of correctness associated with the prediction and outputs the prediction if the probability of correctness is above a threshold; and

annotates the prediction with an indication of confidence.

3. The system of claim 1, the predictive model component outputs an occurring event if the occurring event corresponds to the one or more definitions of surprise.

4. The system of claim 1, the one or more definitions of surprise are a function of frequency of occurrence of events that can be predicted by the predictive model component.

5. The system of claim 1, further comprising:

a robust predictive model; and

a user expectancy model, the one or more definitions of surprise are based at least in part upon a comparison between predicted events output by the robust predictive model and predicted events output by the user expectancy model.

6. The system of claim 5, further comprising a difference analyzer that calculates a measure of difference between events output by the robust predictive model and events output by the user expectancy model.

7. The system of claim 1, further comprising a utility analyzer that calculates a measure of utility in connection with outputting the predicted event, the predicted event output if a measure of utility corresponding to the event is above a threshold.

8. The system of claim 1, further comprising an alerting component that utilizes alerting rules to determine a manner in which to alert the user of an event that corresponds to the one or more definitions of surprise.

9. The system of claim 1, further comprising:

a feedback component that receives user input relating to an event; and

an updating component that updates the one or more definitions of surprise based upon the received user input.

10. The system of claim 1, further comprising a suggestion component that outputs suggestions of actions that can be undertaken in response to the predicted event.

11. A method for recognizing surprising events and predicting future surprising events, comprising:

providing a predictive model;

associating definitions of atypical events with the predictive model;

providing the predictive model with contextual data;

predicting an event as a function of the contextual data; and

determining that the predicted event corresponds to the definitions of surprising events.

12. The method of claim 11, further comprising:

identifying when a user will be surprised by a current situation;

creating the definitions by way of the identified surprising situation; and

alerting the user.

13. The method of claim 11, further comprising:

identifying when a user will be surprised by a current situation;

building a case library based at least in part upon the identified surprising situation; and

building the predictive model as one or more statistical models that perform surprise forecasting from the case library based upon observations that occur before the surprising situation.

14. The method of claim 13, further comprising:

employing the predictive model in real-time by monitoring evidence and inferring and outputting a probability that there will be one of a surprise and a surprise of a particular kind in the future.

15. The method of claim 13, further comprising constructing the one or more statistical models by way of at least one machine learning method from a list comprising Bayesian networks, dynamic Bayesian networks, continuous time Bayesian networks, support vector machines, neural network models, Hidden Markov Models, Markov decision processes, and particle filtering methods.

16. The method of claim 11, further comprising:

recognizing an event based at least in part upon the received contextual data;

determining that the recognized event corresponds to at least one definition of a surprising event within the definitions of surprising events; and

outputting the recognized event as a surprising event.

17. The method of claim 11, further comprising:

calculating a measure of utility associated with the predicted event; and

outputting the predicted event if the measure of utility associated therewith is above a threshold.

18. The method of claim 11, further comprising:

receiving a first prediction of a first event from a user-expectancy model;

receiving a second prediction of a second event from a robust model; and

outputting the second prediction of the second event as a surprising event as a function of a comparison between the first prediction and the second prediction.

19. A predictive system, comprising:

means for predicting an event that occurs at a future point in time;

means for defining surprising events; and

means for correlating the predicted event with the defined surprising events to determine that the predicted event is a surprising event.

20. The predictive system of claim 19, further comprising:

means for storing surprises labeled by type together with observations that occur prior to each surprise; and

means for constructing models that reason abstractly about a probability of different types of surprises in the future.