Method of electronic knowledge extraction, transfer and use

Details Method of electronic knowledge extraction, transfer and use Method of electronic knowledge extraction, transfer and use

1999-06-22

2002-12-31

Black, Thomas (Department: 2121)

Data processing: artificial intelligence

Knowledge processing system

C706S046000, C706S047000, C706S048000

Reexamination Certificate

active

06502084

ABSTRACT:

BACKGROUND OF THE INVENTION
The invention relates generally to a method of processing stored data, e.g., text type data, and, more particularly, to a method of knowledge extraction, transfer and use of such stored data.
When generating documents and drawings repeatedly that are directed to a common subject, such as for example an engineering drawing. Certain types of errors commonly reoccur. Checklists have been used to detect reoccurring errors in many applications, such as in engineering drawings. These checklists are workable when they are kept to a reasonable length, whereby only the most common reoccurring errors are detected by use of such a checklist. However, many types of errors do not occur often enough to be included in the checklist, whereby such errors are often missed.
Such errors include, but are not limited to, the result of an improper entry into the computer system, i.e., human error. It would be advantageous to utilize the processing power of the computer to analyze and provide correction for such errors, particularly in such electronically stored documents and drawings that are of the type that are repeatedly generated.
BRIEF SUMMARY OF THE INVENTION
In an exemplary embodiment of the invention a method of knowledge extraction, transfer and use of stored data is provided. With electronic data files, e.g., documents and drawings, the items in the documents and on the drawings are associative and dependent. More specifically, there are groups of items that are always seen together, and these groups are dependent on the document or drawing type. Using known computer program techniques, such as are available with, e.g., “per15” programming language, it is possible to extract knowledge (rules) from groups of documents or drawings and apply the extracted rules to a new document or drawing without the need to reprogram. In an exemplary embodiment of the invention, a method (program) of checking a data file in a computer system comprises: providing a knowledge base (similar correct files) for a data file type, the data file (being checked) being of the data file type; applying the knowledge base to the data file; and; providing output data indicative of a result of the applying the knowledge base to the data file.
The program can learn what is correct, from groups of similar documents or drawings and apply that knowledge to new documents or drawings. The Program can also tailor or target, tests and outputs without a programmer. In addition to such rules extraction and use, the program can be made to be robust, portable, user friendly, versatile and may be used as a stand alone or embedded in another program.


REFERENCES:
patent: 4783752 (1988-11-01), Kaplan et al.
patent: 4891766 (1990-01-01), Derr et al.
patent: 4939668 (1990-07-01), Brown et al.
patent: 5402524 (1995-03-01), Bauman et al.
patent: 5412756 (1995-05-01), Bauman et al.
patent: 5418888 (1995-05-01), Alden
patent: 5450545 (1995-09-01), Martin et al.
patent: 5488714 (1996-01-01), Skidmore
patent: 5537590 (1996-07-01), Amado
patent: 5574828 (1996-11-01), Hayward et al.
patent: 5701400 (1997-12-01), Amado
patent: 5793933 (1998-08-01), Iwamasa
patent: 5899985 (1999-05-01), Tanaka
patent: 5940817 (1999-08-01), Kishi et al.
Learning to Extract Symbolic Knowledge from the World Wide3 Wed, Mark Craven; Dayne Freitag; Andrew McCallum; Tom Mitchell; Kamal Nigam; Choon Yang Quek; (1998) Proceeding of AAAI-98, Conf. of American Artificial Intelligence, pps. 1-17.
Learning Information Extraction Rules for Semi-structured and Free Text, Stephen Soderland; (1999) Machine Learning, pps. 1-44.
Relational Learning of Pattern-Match Rules for Information Extraction, Mary Elaine Califf and Raymond J. Mooney (1997) AAAI Spring Symposium on Applying Machine Learning to Discourse Preocessing, pps. 1-6.

Affiliated with

Also associated with

Rating

Method of electronic knowledge extraction, transfer and use does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Method of electronic knowledge extraction, transfer and use, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of electronic knowledge extraction, transfer and use will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2926310