Computer-implemented expert system-based method and system for document recognition and content understanding

• US 8,000,530 B2
• Filed: 10/26/2006
• Issued: 08/16/2011
• Est. Priority Date: 10/26/2006
• Status: Active Grant

• Current Assignee: Hubin Jiang
• Original Assignee: Hubin Jiang
• Inventors: Jiang, Hubin
• Primary Examiner(s): Le; Brian Q
• Assistant Examiner(s): Park; Edward
• Application Number:
  US11/553,375
• Publication Number:
  US 20080112620A1
• Time in Patent Office: 1,755 Days
• Field of Search: 382/181, 382/231, 382/135, 382/137, 382/138, 382/139, 382/140, 382/155, 382/161
• US Class Current: 382/181
• CPC Class Codes: G06F 16/353 into predefined classes

Abstract

A system and process for recognizing documents by type and understanding at least a portion of the contents thereof. The process includes the steps of providing a document in electronic form, determining a set of facts, data and information about the document, providing the set of facts, data and information to a reasoning management unit, receiving formatted knowledge from a rule verification, validation and management unit, using the formatted knowledge to prepare an expert system having an inference engine, matching the facts, data and information against formatted knowledge using the inference engine and determining a set of applicable rules and executing actions of applicable rules.

Claims

1. A computer-implemented, knowledge reasoning process for understanding, in natural language format, at least a portion of the contents of a document, comprising using a computer processor to perform the steps of:

(a) receiving a document in electronic form, the document stored on a computer readable medium;

(b) determining and extracting facts and goals, which consist of definitions, information, and data in natural language format, the facts and goals representative of local characteristics and context of the document;

(c) providing the facts and goals to a computer-implemented reasoning management unit;

(d) receiving formatted knowledge corresponding to human document understanding in natural language format from a computer-implemented rule verification, validation and management unit;

(e) using the formatted knowledge to prepare a computer-implemented expert system having an inference engine;

(f) reasoning the facts and goals against formatted knowledge using the inference engine of the computer-implemented expert system and determining a set of applicable rules regarding understanding of the content and context of the at least a portion of the document;

(g) providing the computer-implemented reasoning management unit with the applicable rules and information concerning the understanding of at least a portion of the document;

(h) sending a summary of abstract meaning concerning content understanding to a results and reports processing unit; and

(i) performing at least one function on the document in electronic form selected from document classification, document sorting, information redaction, information capturing and data extraction

wherein the rule verification, validation and management unit performs the following steps;

(1) determining whether conflicting rules exist;

(2) determining whether conflicting facts exist; and

(3) determining whether cyclic data locks, recursive loops and indefinite execution of rules are present.

2. The process of claim 1, wherein the facts and goals are generated using a facts generator using at least one tool selected from the group consisting of a pattern generator, an optical character recognition program, natural handwriting recognition program and an intelligent character recognition program.

3. The process of claim 1, wherein performing steps (a)-(f) does not result in the understanding of at least a portion of the document content and further comprising the step of sending the document to a training management unit.

4. The process of claim 3, further comprising the step of providing an expert with tools and mechanisms to enable the expert to provide information concerning the characteristics of the document type and goals of content understanding.

5. The process of claim 4, wherein the tools and mechanisms are provided through a logic language and graphics user interface.

6. The process of claim 4, further comprising the step of providing the expert with the document from the training management unit and having the expert provide information concerning the characteristics of the document type and goals of content understanding.

7. The process of claim 6, further comprising the step of providing a knowledge generator with the information concerning the characteristics of the document type and goals of content understanding and organizing the information so provided into rules and formatted knowledge.

8. The process of claim 7, further comprising the step of providing the rules and formatted knowledge to the rule verification, validation and management unit for verification and validation.

9. The process of claim 8, further comprising the step of saving the rules and formatted knowledge to a knowledge base.

10. The process of claim 1, wherein the rule verification, validation and management unit performs the following steps:

(iv) sorting rules into groups and managing the groups so arranged;

(v) managing the knowledge base; and

(vi) assembling the rules into a format useful for the expert system.

11. The process of claim 10, further comprising the step of providing the formatted rules to the reasoning management unit.

12. The process of claim 11, further comprising the step of providing the formatted rules to the expert system for execution.

13. A computer-implemented, knowledge reasoning system for understanding, in natural language format, at least a portion of the contents of a document, comprising:

(a) a computer processor for implementing the knowledge reasoning system;

(b) an electronic file management unit for providing a document in electronic form, the document stored on a computer readable medium;

(c) a facts generator for determining and extracting facts and goals, which consist of definitions, information, and data in natural language format, the facts and goals representative of local characteristics and context of the document;

(d) a reasoning management unit for receiving the facts and goals from said facts generator;

(e) a rule verification, validation and management unit for providing formatted knowledge;

(f) an expert system having an inference engine for reasoning the facts and goals against the formatted knowledge using said inference engine and determining a set of applicable rules regarding understanding of the content and context of the at least a portion of the document, said expert system implemented by said computer processor, and

(g) a report processing unit for receiving a summary of abstract meaning concerning content understanding

wherein said rule verification, validation and management unit determines whether conflicting rules exist;

determines whether conflicting facts exist; and

determines whether cyclic data locks, recursive loops and indefinite execution of rules are present.

14. The system of claim 13, wherein the set of facts, data and information is generating through the use of at least one tool selected from the group consisting of a pattern generator, an optical character recognition program, natural handwriting recognition program and an intelligent character recognition program.

15. The system of claim 13, further comprising a training management unit for receiving the document in a case where said reasoning management unit is not initially provided with information concerning the understanding of at least a portion of the document content.

16. The system of claim 15, further comprising tools and mechanisms to enable an expert to provide information concerning characteristics of the document type and goals of content understanding.

17. The system of claim 16, wherein said tools and mechanisms are provided through a logic language and graphics user interface.

18. The system of claim 16, wherein the expert is provided with the document from said training management unit and the expert provides information concerning the characteristics of the document type and goals of content understanding.

19. The system of claim 18, further comprising a knowledge generator for receiving information concerning the characteristics of the document type and goals of content understanding and organizing the information so provided into rules and formatted knowledge.

20. The system of claim 19, wherein said verification, validation and management unit verifies, validates and receives rules and formatted knowledge.

21. The system of claim 20, further comprising a knowledge base for saving the rules and formatted knowledge.

22. The system of claim 13, wherein said rule verification, validation and management unit sorts rules into groups and managing the groups so arranged;

manages said knowledge base; and

assembles the rules into a format useful for said expert system.

23. The system of claim 22, wherein the formatted rules are provided to said reasoning management unit.

24. The system of claim 23, wherein formatted rules are provided to said expert system for execution.