Contextually-Appropriate Task Reminders

• US 20120242482A1
• Filed: 03/25/2011
• Published: 09/27/2012
• Est. Priority Date: 03/25/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Elumalai, Arulkumar, Bouchard, Thomas, Chang, Steve, Shtiegman, Eran, Tinling, Aaron
• Granted Patent: US 8,766,793 B2
• Time in Patent Office: Days
• US Class Current: 340/539.13
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/00 Computing arrangements based on specific mathematical models
• CPC Class Codes: G06Q 10/109 Time management, e.g. calendars, reminders, meetings or time accounting

Abstract

A computing device learns over time how to identify task reminders that are most likely to be helpful to a user in different contexts. The task reminders can remind the user of activities that the user needs to do. The computing device displays a graphical user interface (GUI) that contains the task reminders identified as being most likely to be helpful to the user in the user'"'"'s current context. The computing device updates the task reminders in the GUI as the user'"'"'s context changes. In this way, the computing device can present task reminders that are likely to be currently helpful to the user while suppressing task reminders that are less likely to be helpful to the user at the current time.

Claims

1. A method for providing contextually-appropriate task reminders, the method comprising:

learning, by a computing device, over time how to identify contextually-appropriate task reminders in different contexts;

displaying a graphical user interface (GUI) on a display unit, the graphical user interface containing a reminder area; and

as a current context of the user changes, updating the reminder area such that the reminder area contains task reminders that are contextually-appropriate in the current context of the user.

2. The method of claim 1,

wherein the method further comprises:

receiving context data that provides information about the current context of the user; and

using the context data to generate data representative of a likelihood that the user would find a given task reminder helpful in the current context; and

wherein displaying the GUI comprises displaying the given task reminder in the reminder area after determining that the likelihood is sufficiently great.

3. The method of claim 2,

wherein the method further comprises:

calculating scores for a plurality of task reminders based at least in part of the context data, the given task reminder being among the plurality of task reminders, the score for the given task reminder being the data representative of the likelihood that the user would find the given task reminder helpful; and

ranking the plurality of task reminders based on the scores for the task reminders;

wherein displaying the GUI comprises;

displaying ones of the task reminders having highest ranks either without displaying ones of the task reminders having lower ranks or displaying the task reminders having lower ranks less prominently than the task reminders having the highest ranks.

4. The method of claim 3, wherein calculating the scores for the plurality of task reminders comprises:

providing the context data to input layer artificial neurons in an artificial neural network; and

receiving a given score for a given task reminder from one or more output layer artificial neurons in the artificial neural network, the given task reminder being among the plurality of task reminders.

5. The method of claim 2, wherein learning over time how to identify the contextually-appropriate task reminders comprises:

receiving feedback from the user; and

modifying how the computing device identifies the contextually-appropriate task reminders in response to the feedback.

6. The method of claim 4, wherein receiving the feedback comprises receiving input from the user to reorder task reminders in the reminder area.

7. The method of claim 2, wherein receiving the context data comprises receiving at least some of the context data from another computing device.

8. The method of claim 7, wherein receiving the context data comprises receiving at least some of the context data from a source within the computing device.

9. The method of claim 1, wherein the method further comprises generating a plurality of task reminders, the contextually-appropriate task reminders being among the plurality of task reminders, each task reminder in the plurality of task reminders corresponding to a task to be performed by the user.

10. The method of claim 9, wherein generating the plurality of task reminders comprises generating the plurality of task reminders based on task items in a task list of the user.

11. The method of claim 9,

wherein the method further comprises:

displaying a document in the GUI; and

receiving input from the user to flag the document; and

wherein generating the plurality of task reminders comprises generating a given task reminder in response to receiving the input from the user to flag the document.

12. The method of claim 9,

wherein the method further comprises:

displaying a document in the GUI;

identifying a set of task source elements within the document;

receiving input to drag a given task source element to the reminder area, the given task source element being in the set of task source elements; and

wherein generating the plurality of task reminders comprises generating a given task reminder in response to receiving the input.

13. The method of claim 1, further comprising:

determining that a given task reminder that was determined not to be contextually-appropriate in one context to be appropriate in a current context of the user; and

performing a given notification action in response to determining that the given task reminder is now contextually appropriate.

14. The method of claim 13, further comprising:

learning, by the computing device, over time how to select appropriate notification actions given the different contexts; and

determining that the given notification action is more likely than a plurality of other available notification actions to be a most appropriate notification action given the current context of the user.

15. A computing device comprising:

one or more computer storage media that store instructions; and

one or more processing units that access and execute the instructions, execution of the instructions by the one or more processing units configuring the computing device such that the computing device;

learns over time how to identify contextually-appropriate task reminders for different contexts of a user; and

causes a display unit to display a graphical user interface (GUI), the GUI containing a reminder area that is updated as a current context of the user changes such that the reminder area contains task reminders that are contextually-appropriate in the current context of the user.

16. The computing device of claim 15, wherein execution of the instructions by the one or more processing units configures the computing device such that the computing device:

receives context data that provides information about a current geographical location of the computing device; and

learns over time that it is contextually-appropriate to display task reminders corresponding to tasks involving a given geographical location when the current geographical location of the computing device is within a given distance of the given geographical location.

17. The computing device of claim 15, wherein execution of the instructions by the one or more processing units configures the computing device such that the computing device:

receives input from the user to drag a contact to the reminder area, the contact comprising contact information for another user;

generates a given task reminder in response to receiving the input;

receives first context data indicating that the other user is not available;

determines, after receiving the first context data, that the given task reminder is not among the contextually-appropriate task reminders for a current context of the user;

receives second context data indicating that the other user is available; and

determines, after receiving the second context data, that the given task reminder is among the contextually-appropriate task reminders for a later context of the user.

18. The computing device of claim 15, wherein execution of the instructions by the one or more processing units configures the computing device such that the computing device:

generates a plurality of task reminders, each of the task reminders corresponding to a task to be performed by the user; and

scrolling the reminder area in response to input from the user, wherein prior to scrolling the reminder area, the reminder area contains ones of the task reminders that the computing device determines to be most likely to be helpful to the user given a current context of the user, wherein after scrolling the reminder area, the reminder area contains ones of the task reminders that the computing device determines to be less likely to be helpful to the user given the current context of the user.

19. The computing device of claim 15,

wherein the GUI includes an activity area, the reminder area being a bar adjacent to one edge of the activity area;

and

wherein execution of the instructions by one or more processing units configures the computing device such that the computing device;

receives selection input from the user, the selection input indicating a given task reminder in the reminder area; and

displays in the activity area additional information about a task corresponding to the given task reminder.

20. A computer storage medium that stores instructions, execution of the instructions by one or more processing units of a computing device configuring the computing device such that the computing device:

displays a graphical user interface (GUI) on a display unit, the graphical user interface containing a reminder area and an activity area, the reminder area containing a set of top-ranked task reminders, the set of top-ranked task reminders determined by the computing device to have highest likelihoods of being helpful a user in an initial context;

receives feedback from the user;

applies the feedback to a machine learning algorithm;

receives context data after applying the feedback to the machine learning algorithm, the context data representing a changed context;

uses the machine learning algorithm to calculate scores for a plurality of task reminders, the scores being numerical values indicative of likelihoods of the task reminders being helpful to the user in the changed context;

ranks the task reminders based on the scores for the task reminders; and

updates the reminder area to display ones of the task reminders having the greatest scores.