Adaptive dynamic personal modeling system and method

• US 8,326,792 B2
• Filed: 12/30/2008
• Issued: 12/04/2012
• Est. Priority Date: 04/30/2001
• Status: Expired due to Fees

• Current Assignee: Tanvir Goraya
• Original Assignee: The Theory Garden LLC
• Inventors: Goraya, Tanvir Y.
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): COUGHLAN, PETER D
• Application Number:
  US12/345,982
• Publication Number:
  US 20090228425A1
• Time in Patent Office: 1,435 Days
• Field of Search: None
• US Class Current: 706/59
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition

Abstract

The present invention provides a system and method for building and analyzing a quantitative model from qualitative tacit knowledge. In one aspect of the present invention, there is provided a system for creating and analyzing a model of a user'"'"'s semantic knowledge. The semantic knowledge model is based on cause and effect relationships as defined by the user. In another aspect of the present invention, there is provided a system for creating and analyzing a model of a user'"'"'s episodic knowledge. The episodic knowledge model is based on the user'"'"'s past experiences, including recalled stimuli and responses. The semantic and episodic models are used to describe the users internal mental model.

Claims

1. A computerized method performed on a computer, the method comprising:

providing output on a screen of the computer from a cause-and-effect model, to a user, wherein the providing includes displaying output effect values;

receiving in the computer from the user one or more desired effect values that are different from one or more corresponding of the output effect values; and

modifying, in the computer, the cause-and-effect model as a function of difference between the one or more desired effect values and the one or more corresponding output effect values;

wherein the cause-and-effect model includes a matrix of causes and effects, with numerical values each representing an influence the corresponding cause has on the corresponding effect; and

wherein the modifying includes changing numerical influence values as a function of the difference between the one or more desired effect values and the one or more corresponding output effect values; and

further comprising, prior to the providing the output, constructing the cause-and effect model, wherein the constructing includes;

receiving from the user the matrix of causes and effects, wherein the matrix of causes and effects is entered into the computer using a graphical user interface of the computer; and

receiving from the user the numerical influence values for the matrix.

2. The method of claim 1,

wherein the constructing also includes receiving from the user relative importance values associated with at least some of the causes;

and

wherein the modifying includes changing the influence values as a function of both the relative importance values and the difference between the one or more desired effect values and the one or more corresponding output effect values.

3. The method of claim 1, wherein the constructing also includes receiving from the user bias values associated with at least some of the effects.

4. The method of claim 1, wherein the constructing also includes receiving from the user faith values associated with at least some cause-and-effect pairs.

5. The method of claim 1, further comprising, after the constructing and before the providing the output, validating the model, wherein the validating includes:

receiving from the user cause values associated with the causes; and

calculating the output effect values from the causes and the influence values.

6. A computerized method performed on a computer, the method comprising:

providing output on a screen of the computer from a cause-and-effect model, to a user, wherein the providing includes displaying output effect values;

receiving in the computer from the user one or more desired effect values that are different from one or more corresponding of the output effect values; and

modifying, in the computer, the cause-and-effect model as a function of difference between the one or more desired effect values and the one or more corresponding output effect values;

wherein the receiving includes receiving input by the user manipulating the output display on the screen;

wherein the manipulating includes the user clicking and dragging on graphical representations of the output effect values;

wherein the providing includes providing a bar graph of output effect values as the graphical representations of the output effect values;

wherein the cause-and-effect model includes a matrix of causes and effects, with numerical values each representing an influence the corresponding cause has on the corresponding effect; and

wherein the modifying includes changing numerical influence values as a function of the difference between the one or more desired effect values and the one or more corresponding output effect values.

7. The method of claim 6,

wherein the receiving also includes receiving from the user one or more confidence levels associated with respective of the one or more desired effect values;

wherein the modifying includes changing the influence values as a function of both the one or more confidence levels and the difference between the one or more desired effect values and the one or more corresponding output effect values.

8. The method of claim 7,

wherein the model also includes relative importance values associated with at least some of the causes;

and

wherein the modifying includes changing the influence values as a function of the relative importance values, the one or more confidence levels, and the difference between the one or more desired effect values and the one or more corresponding output effect values.

9. The method of claim 8,

further comprising, prior to the providing the output, constructing the cause-and effect model, wherein the constructing includes:

receiving from the user the matrix of causes and effects, wherein the matrix of causes and effects is entered into the computer using a graphical user interface of the computer; and

receiving from the user the numerical influence values for the matrix.

10. The method of claim 9, wherein the constructing also includes receiving from the user bias values associated with at least some of the effects.

11. The method of claim 10, wherein the constructing also includes receiving from the user faith values associated with at least some cause-and-effect pairs.

12. The method of claim 9,

further comprising, after the constructing and before the providing the output, validating the model, wherein the validating includes:

receiving from the user cause values associated with the causes; and

calculating the output effect values from the causes and the influence values.