Contextual responses based on automated learning techniques

• US 8,020,104 B2
• Filed: 01/11/2005
• Issued: 09/13/2011
• Est. Priority Date: 12/18/1998
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Matteson, Eric L., Robarts, James O.
• Primary Examiner(s): HAILU, TADESSE
• Application Number:
  US11/033,974
• Publication Number:
  US 20060004680A1
• Time in Patent Office: 2,436 Days
• Field of Search: 715736-748, 715764-769, 715/751, 715/759, 709/224, 709/225, 709/226, 706/45, 706/60, 706/61, 706/51, 706/43
• US Class Current: 715/744
• CPC Class Codes: G06F 1/163 Wearable computers, e.g. on a belt
• CPC Class Codes: G06F 16/40 of multimedia data, e.g. slideshows comprising image and additional audio data retrieval of still image data G06F16/50; retrieval of audio data G06F16/60; retrieval of video data G06F16/70
• CPC Class Codes: G06F 21/00 Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
• CPC Class Codes: G06F 21/606 by securing the transmission between two devices or processes
• CPC Class Codes: G06F 21/6245 Protecting personal data, e.g. for financial or medical purposes
• CPC Class Codes: G06F 2221/2149 Restricted operating environment
• CPC Class Codes: G06F 3/011 Arrangements for interaction with the human body, e.g. for user immersion in virtual reality blind teaching G09B21/00
• CPC Class Codes: G06F 3/016 Input arrangements with force or tactile feedback as computer generated output to the user
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/004 Artificial life, i.e. computing arrangements simulating life

Abstract

Techniques are disclosed for using a combination of explicit and implicit user context modeling techniques to identify and provide appropriate computer actions based on a current context, and to continuously improve the providing of such computer actions. The appropriate computer actions include presentation of appropriate content and functionality. Feedback paths can be used to assist automated machine learning in detecting patterns and generating inferred rules, and improvements from the generated rules can be implemented with or without direct user control. The techniques can be used to enhance software and device functionality, including self-customizing of a model of the user'"'"'s current context or situation, customizing received themes, predicting appropriate content for presentation or retrieval, self-customizing of software user interfaces, simplifying repetitive tasks or situations, and mentoring of the user to promote desired change.

Claims

1. A method in a computing device for improving responses to context information based on automated learning techniques, multiple responses each being associated with a group of context information, the method comprising:

employing a processor executing computer executable instructions to perform the following acts;

receiving indications of past user actions by at least one user related to each of multiple automatically generated responses and of context information related to each of the multiple responses, the context information includes at least one attribute provided by a sensing device, the at least one attribute includes an attribute value and a timestamp indicating the age of the attribute value;

automatically detecting a relationship between a group of context information and one of the responses based on the received indications and at least one defined theme, wherein the at least one defined theme includes a related set of attributes that reflect a context of a user; and

creating an association between the group of context information and the one response, so that the one response can be provided in the future in response to a context that matches the group of context information.

2. The method of claim 1 wherein the computing device performs the method on behalf of a user of a thin client computing device.

3. The method of claim 1, wherein the theme relates to at least one of a hobby, travel destination, health condition, occupation, cognitive load, entertainment, or personal relationship of the user.

4. The method of claim 1, wherein the attributes related to user state, user physical environment, and user telecommunication environment are processed into a higher level indication.

5. The method of claim 1, wherein duplicate values for an attribute provided by multiple sources are mediated to determine which value will be employed.

6. The method of claim 5, wherein the determination is based upon at least one of the newness of each of the duplicate attribute values, and average of the duplicate attribute values or the accuracy of each of the duplicate attribute values.

7. The method of claim 1, wherein the theme defines a theme layout for information presentation.

8. The method of claim 1, wherein artificial intelligence is employed to automatically detect a relationship between a group of context information and one of the responses.

9. The method of claim 1, wherein a user has a plurality of themes and the user switches themes as their context changes.

10. The method of claim 1, wherein each attribute of the related set of attributes within a theme has an indicator of whether the attribute is required for the theme.

11. The method of claim 1, wherein each theme has an indicator of priority that is employed to determine which theme is used in automatically detecting a relationship between a group of context information and one of the responses.

12. The method of claim 1, wherein the at least one attribute includes an uncertainty value of the attribute value.

13. The method of claim 1, wherein the uncertainty value indicates a range of values around the attribute value that the attribute value will likely be within.

14. The method of claim 12, wherein the at least one attribute includes an indication of measurement units in which the uncertainty value is expressed.

15. A system for improving responses to context information based on automated learning techniques, multiple responses each being associated with a group of context information, comprising:

at least one processor configured to execute;

a processor;

a memory communicatively coupled to the processor, the memory having stored therein computer-executable instructions to implement the system, including;

a learning component that receives indications of past user actions by at least one user related to each of multiple automatically generated responses and of context information related to each of the multiple responses, the context information includes at least one attribute, the at least one attribute includes an attribute value and an uncertainty value of the attribute value, wherein the uncertainty value indicates a range of values around the attribute value that the attribute value will likely be within, and the learning component automatically detects a relationship between a group of context information and one of the responses based on the received indications and at least one defined theme, wherein the theme includes a related set of attributes that reflect a context related to a user; and

a response modifier component that creates an association between the group of context information and the one response.

16. The system of claim 15 wherein the system is a wearable computer.

17. A system for improving responses to context information based on automated learning techniques, multiple responses each being associated with a group of context information, comprising:

a processor;

a memory communicatively coupled to the processor, the memory having stored therein computer-executable instructions to implement the system, including;

means for receiving indications of past user actions by at least one user related to each of multiple automatically generated responses and of context information related to each of the multiple responses;

at least one processor configured to automatically detect a relationship between a group of context information and one of the responses based on the received indications and an automatically selected one of a plurality of defined themes, wherein a t-he theme includes a related set of attributes that reflect a context of a user;

where each of the plurality of themes is assigned a priority by the at least one user, the automatic selection of one of the plurality of themes is based at least upon priority; and

means for creating an association between the group of context information and the one response.

18. A computer-readable storage medium whose contents cause a computing device to improve responses to context information based on automated learning techniques, multiple responses each being associated with a group of context information, by:

employing a processor executing computer executable instructions embodied on the computer readable storage medium to perform the following acts;

receiving indications of past user actions by at least one user related to each of multiple automatically generated responses and of context information related to each of the multiple responses, the context information includes a plurality of attributes;

automatically detecting a relationship between a group of context information and one of the responses based on the received indications and at least one automatically selected theme from a plurality of defined themes, wherein a defined theme includes a related set of attributes that reflect a context of a user, wherein each attribute of the related set of attributes within the defined theme has an indicator of whether the attribute is required for the defined theme, the selected theme is selected based upon a degree of match between required attributes of the defined theme versus the presence of the required attribute amongst the plurality of attributes included in the context information, the selected theme has the highest degree of match; and

creating an association between the group of context information and the one response.

19. The computer-readable storage medium of claim 18 wherein the computer-readable storage medium is a memory of the computing device.