Knowledge processing system employing confidence levels

• US 5,809,493 A
• Filed: 05/31/1996
• Issued: 09/15/1998
• Est. Priority Date: 12/14/1995
• Status: Expired due to Term

• Current Assignee: Alcatel-Lucent USA, Inc. (Nokia Corporation)
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Lawrence, Victor Bernard, Ahamed, Syed Vickar
• Primary Examiner(s): MacDonald, Allen R.
• Assistant Examiner(s): Smith, Jeffrey S.
• Application Number:
  US08/657,924
• Time in Patent Office: 837 Days
• Field of Search: 395/61, 395/75
• US Class Current: 706/52
• CPC Class Codes: G06N 5/025 Extracting rules from data

Abstract

A knowledge processing system includes a memory for storing knowledge in a plurality of knowledge modules in a knowledge bank, in which the knowledge includes data associated with pre-existing solutions to previous problems; and a processing unit, operating knowledge-based software, and including a knowledge processing unit for generating a plurality of artificial intelligence based modules, and for executing the plurality of artificial intelligence base modules to process input data, including problem related data, for generating a plurality of solution-related modules from the data representing the pre-existing solutions, with the solution-related modules corresponding to solutions to the current problem; a numerical processing unit for generating a confidence level associated with a respective solution to the current problem; and a database processing unit for controlling the storing of the solution-related modules in the memory. The knowledge processing unit responds to the confidence levels of the solutions to iteratively revise the solutions to generate an optimal solution to the current problem.

Claims

1. A knowledge processing system comprising:

a memory for storing knowledge in a plurality of knowledge modules in a knowledge bank, the knowledge including data associated with preexisting solutions to previous problems; and

a processing unit for receiving sensory signals from an external device, the sensory signals including problem-related data corresponding to a current problem, the processing unit operating knowledge-based software and including;

a knowledge processing unit for generating a plurality of artificial intelligence based modules by processing the current problem using a natural language processor to generate tentative code, and for executing the plurality of artificial intelligence based modules to process the input signals representing data, including problem related data corresponding to the current problem, for generating a plurality of solution-related modules from the data representing the pre-existing solutions, with the solution-related modules corresponding to solutions to the current problem;

a numerical processing unit for generating a confidence level associated with a respective solution to the current problem by applying the tentative code to the current problem and generating, the confidence level from a tentative solution corresponding to the tentative code; and

a database processing unit for controlling the storing of the solution-related modules in the memory;

wherein the knowledge processing unit, responsive to the confidence levels of the solutions, iteratively revises the solutions to generate an optimal solution to the current problem by determining a solution strategy using the confidence level corresponding to the tentative solution and adjusting the solution strategy to increase corresponding confidence levels the optimal solution being transmitted as data signals to at least one device.

2. The knowledge processing system of claim 1 wherein the knowledge processing unit iteratively revises the solutions until an optimal solution is generated having an associated confidence level above a predetermined value.

3. The knowledge processing system of claim 1 wherein the knowledge processing unit iteratively revises the solutions until an optimal solution is generated having an associated confidence level having a maximum value of the confidence levels of the solutions.

4. The knowledge processing system of claim 1 wherein the knowledge processing unit generates the plurality of artificial intelligence based modules using a set of predetermined knowledge operation mappings, including knowledge opcodes (kopcodes), between goals associated with the current problem.

5. The knowledge processing system of claim 4 wherein the knowledge processing unit operates the kopcodes on at least one goal as a knowledge operand, the at least one goal included in the solutions to the current problem.

6. The knowledge processing system of claim 5 wherein the knowledge processing unit operates at least one kopcode to perform a knowledge domain function on at least one goal.

7. The knowledge processing system of claim 5 wherein the knowledge processing unit operates at least one kopcode to retrieve attributes of goals.

8. The knowledge processing system of claim 5 wherein the knowledge processing unit operates at least one kopcode to retrieve forward or backward pointers for performing knowledge domain functions.

9. The knowledge processing system of claim 1 wherein the numerical processing unit tracks confidence levels associated with a set of knowledge microfunctions included in the plurality of solution-related modules to generate an overall confidence level associated with the solution of the current problem.

10. The knowledge processing system of claim 9 wherein the numerical processing unit includes a multiplier for evaluating a product of a plurality of confidence levels to determine the overall confidence level of the solution.

18. The knowledge processing system of claim 1 wherein said system is configured as an expert system, said knowledge including data associated with pre-existing solutions to previous problems being knowledge of a human expert.

11. A network for processing knowledge comprising:

a sensor for receiving sensory signals from an external device, the sensory signals including problem-related data corresponding to a current problem;

at least one output device for transmitting data signals to the external device;

a memory for storing knowledge in a plurality of knowledge modules in a knowledge bank, the knowledge including data associated with preexisting solutions to previous problems; and

a processing unit, operating knowledge-based software, the processing unit including;

a knowledge processing unit for generating a plurality of artificial intelligence based modules by processing the current problem using a natural language processor to generate tentative code, the knowledge processing unit executing the plurality of artificial intelligence based modules to process the sensory signals, including the problem-related data, for generating a plurality of solution-related modules from the data representing the pre-existing solutions, with the solution-related modules corresponding to solutions to the current problem;

a numerical processing unit for generating a confidence level associated with a respective solution to the current problem by applying the tentative code to the current problem and generating the confidence level from a tentative solution corresponding to the tentative code; and

a database processing unit for controlling the storing of the solution-related modules in the memory;

wherein the knowledge processing unit, responsive to the confidence levels of the solutions, iteratively revises the solutions to generate an optimal solution to the current problem by determining a solution strategy using the confidence level corresponding to the tentative solution and adjusting the solution strategy to increase corresponding confidence levels, the optimal solution being output as the data signals by the output device to the external device.

12. The network of claim 11 wherein the knowledge processing unit iteratively revises the solutions until an optimal solution is generated having an associated confidence level above a predetermined value;

and

the output device outputs the solution to the external for generating a feedback signal to the sensors.

13. The network of claim 11 wherein the knowledge processing unit iteratively revises the solutions until an optimal solution is generated having an associated confidence level having a maximum value of the confidence levels of the solutions for providing optimal control of the external device.

20. The network of claim 11 wherein the knowledge processing unit is adapted to read just a solution strategy in order to iteratively revise said solutions.

14. A method for processing knowledge comprising the steps of:

storing knowledge in a plurality of knowledge modules in a knowledge bank, the knowledge including data associated with pre-existing solutions to previous problems;

generating a plurality of artificial intelligence based modules in response to sensory signals transmitted by an external device, the sensory signals including problem-related data corresponding to a current problem and being received by a sensor, the plurality of artificial intelligence based modules being generated by processing the current problem using a natural language processor to generate tentative code;

executing the plurality of artificial intelligence base modules to process the input signals representing data, including problem related data corresponding to the current problem;

generating a plurality of solution-related modules from the data representing the pre-existing solutions, with the solution-related modules corresponding to solutions to the current problem;

generating a confidence level associated with a respective solution to the current problem using a numerical processing unit and applying the tentative code to the current problem and generating the confidence level from a tentative solution corresponding to the tentative code; and

iteratively revising the solutions using the confidence levels to generate an optimal solution to the current problem by determining a solution strategy using the confidence level corresponding to the tentative solution and adjusting the solution strategy to increase corresponding confidence levels, the optimal solution being transmitted as data signals to at least one device.

15. The method of claim 14 wherein the step of iteratively revising the solutions includes the step of:

iteratively revising until an optimal solution is generated having an associated confidence level above a predetermined value.

16. The method of claim 14 wherein the step of iteratively revising the solutions includes the step of:

iteratively revising until an optimal solution is generated having an associated confidence level having a maximum value of the confidence levels of the solutions.

17. The method of claim 14 wherein the step of processing the current problem using a natural language processor to generate tentative code includes the steps of:

determining goals and subgoals of the current problem; and

generating a set of predetermined knowledge operation mappings, including knowledge opcodes (kopcodes), between goals associated with the current problem.

19. The method of claim 14 wherein said method is operative in an expert system, said knowledge including data associated with pre-existing solutions to previous problems comprises knowledge of a human expert, and said method further comprises the step of communicating said optimal solution to a user of the expert system.