Methods for and applications of learning and inferring the periods of time until people are available or unavailable for different forms of communication, collaboration, and information access

• US 7,490,122 B2
• Filed: 01/31/2005
• Issued: 02/10/2009
• Est. Priority Date: 06/28/2001
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Horvitz, Eric J., Jacobs, Andrew W., Kadie, Carl M.
• Primary Examiner(s): Dalencourt; Yves
• Application Number:
  US11/047,527
• Publication Number:
  US 20050132006A1
• Time in Patent Office: 1,471 Days
• Field of Search: 709/200, 709/11, 709/45, 709/46
• US Class Current: 709/200
• CPC Class Codes: G06Q 10/109 Time management, e.g. calendars, reminders, meetings or time accounting

Abstract

A system and method are provided to learn and infer the time until a user will be available for communications, collaboration, or information access, given evidence about such observations as time of day, calendar, location, presence, and activity. The methods can be harnessed to coordinate communications between parties via particular modalities of interaction. The system includes a user state identifier that determines a user'"'"'s state from background knowledge, the flow of time, or one or more context information sources. A data log can be employed to store information about user state changes and observational evidence to accumulate statistics and build inferential models of the availability and unavailability of users for different kinds of communication, collaboration, and information access. A forecaster is constructed from the accumulated statistics and/or learned models to enable a determination of a user'"'"'s likely return, or, more generally, the probability distribution over a user'"'"'s likely return to particular states of availability. The forecaster can be employed to cache information for offline access, drive displays of availability and unavailability, to send messages that include availability forecasts, and to automatically perform scheduling or rescheduling of communications.

Claims

1. A communications system providing fine-grained inferences of a user, comprising:

a processor communicatively coupled to a memory having stored therein a first component that reasons about the amount of time until a user will be available for at least one of a type of interaction or a communications with a message sender, the first component reasons by learning statistics about at least a state of the user from one or more context information sources and building one or more prediction models from the statistics, wherein the one or more prediction models infer the user availability;

a second component that automatically responds to the message sender in relation to the reasoning of the first component, tracks the automatic responses communicated to the message sender and based on identity of the message sender provides updates with a recommendation about another channel of communication to the message sender if the reasoning of the first component fails;

wherein the first component determining when a user will be available to be interrupted with a particular class of alert or notification; and

wherein the determination is based on at least one of patterns of availability, inferences about a workload or associated cost of an interruption.

2. The system of claim 1, the first component reasons about the amount of time until a user will be available based on patterns of at least one of the user'"'"'s availability or context.

3. The system of claim 1, the first component determining where the user is and when the user will be available for at least one of a voice or video conference.

4. The system of claim 1, a special notation marks the user availability information communicated to the message sender indicating that the availability information was inferred by the one or more prediction models.

5. The system of claim 1, the one or more prediction models employ current information associated with user actions from the one or more context information sources to reason about the amount of time until the user is available.

6. The system of claim 1, further comprising a log of the automatic responses provided by the second component for the user to inspect and provide feedback by changing a profile associated with the user.

7. An automated communications system to send information, comprising:

a processor communicatively coupled to a memory having stored therein a forecaster that determines at least one of when a user will likely see a message, be available to see the message, or be available for one or more types of communications by employing a probability distribution over a time until the user returns and current contextual information regarding the user;

a communications component to track what communications have been sent to a contactor and based on the contactor to automatically send updates to the contactor regarding an alternate communication channel for contacting the user if predictions associated with the forecaster have changed;

wherein the forecaster is constructed from accumulated statistics; and

wherein the accumulated statistics enable a determination of the user'"'"'s at least one availability or unavailability.

8. The system of claim 7, the communications component replying to a collaborating party with the user is currently away, and it will probably be at least x hours before the user will see the message, x being an integer or fraction thereof.

9. The system of claim 7, the communications component sending an update if the user has not yet returned.

10. The system of claim 7, the communications component responding with a recommendation relating to at least one of another channel, or changing the nature of the message.

11. A computer readable storage medium containing computer executable instructions for performing a method of retrieving information to facilitate intelligent caching of information, the method comprising;

determining an expected time a user will be away from a system;

determining information access patterns of the user;

performing an approximation to extract a maximum value associated with caching information based on the information access patterns and the expected time away of the user;

determining if it is favorable to cache the information based at least in part on the value of the approximation crossing a predetermined threshold and caching the information where it is favorable and not caching the information where it is not favorable;

the method further comprising determining a cost associated with not having an item when needed given that the item is needed at some time t in the future; and

the cost determination is optimized by considering how long until the user has network availability.

12. The method of claim 11, the information access patterns comprising a determination of how long until various items are needed.

13. The method of claim 11, further comprising determining the size of the items.