COMPUTER-BASED SYSTEM AND METHOD FOR GENERATING, CLASSIFYING, SEARCHING, AND ANALYZING STANDARDIZED TEXT TEMPLATES AND DEVIATIONS FROM STANDARDIZED TEXT TEMPLATES

• US 20130080886A1
• Filed: 09/27/2012
• Published: 03/28/2013
• Est. Priority Date: 08/14/2002
• Status: Active Grant

• Current Assignee: Bureau of National Affairs Incorporated (Bloomberg LP)
• Original Assignee: Document Analytic Technologies LLC
• Inventors: Anderson, Robert
• Granted Patent: US 9,195,639 B2
• Time in Patent Office: Days
• US Class Current: 715/256
• CPC Class Codes: G06F 40/186 Templates
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/56 Natural language generation
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A method for generating, classifying, searching, and analyzing standardized text templates drawn from a plurality of text documents and for identifying standardized text deviations from standardized text templates. Semi-standardized documents may be represented as standardized templates and deviations from standardized templates, with such templates themselves automatically generated by a computer-implemented method from a plurality of similar text documents. The method enables enhanced analysis of semi-standardized documents and automatic extraction of information from standardized text templates.

Claims

1. A computer system configured to automatically analyze text documents by performing the following steps:

comparing text from a subject text to text of a plurality of given text templates, each text template containing at least one paragraph of text;

determining which given text template or text templates has text that matches the text from the subject text document to a given degree of correspondence; and

generating a report of the differences between the text from the subject text document and the text of the matching text template or text templates.

2. The computer system of claim 1 also configured to perform the steps of:

comparing a family of specimen text documents;

identifying one paragraph of text within one of the family of specimen text documents that most closely matches a paragraph of text in all of the other specimen text documents, as compared to all of the other paragraphs in the one specimen text document; and

generating one of the text templates containing at least the one identified paragraph of text.

3. The computer system of claim 2 wherein the identifying the paragraph that most closely matches step uses an edit distance algorithm.

4. The computer system of claim 2 wherein the identifying the paragraph that most closely matches step uses a longest common subsequence algorithm.

5. The computer system of claim 1 wherein the determining matches to a degree of correspondence step uses an edit distance algorithm.

6. The computer system of claim 1 wherein the determining matches to a degree of correspondence step uses a longest common subsequence algorithm.

7. The computer system of claim 1 wherein the differences in the report are computed according to an edit distance algorithm.

8. Non-transitory, tangible, computer readable media containing computer programming instructions that, when loaded in a computer system having a memory, cause the computer system, in response to user commands, to automatically analyze text documents by performing the following steps:

comparing text from a subject text document to text of a plurality of given text templates, each text template containing at least one paragraph of text;

determining which given text template or text templates has text that matches the text from the subject text document to a given degree of correspondence; and

generating a report of the differences between the text from the subject text document and the text of the matching text template or text templates.

9. The media of claim 8 wherein the programming instructions, when loaded in the computer system, also cause the computer system, in response to user commands, to perform the following steps:

comparing a family of specimen text documents;

identifying one paragraph of text within one of the family of specimen text documents that most closely matches a paragraph of text in all of the other specimen text documents, as compared to all of the other paragraphs in the one specimen text document; and

generating one of the text templates containing at least the one identified paragraph of text.

10. The media of claim 9 wherein the identifying the paragraph that most closely matches step uses an edit distance algorithm.

11. The media of claim 9 wherein the identifying the paragraph that most closely matches step uses a longest common subsequence algorithm.

12. The media of claim 8 wherein the determining matches to a degree of Correspondence step uses an edit distance algorithm.

13. The media of claim 8 wherein the determining matches to a degree of correspondence step uses a longest common subsequence algorithm.

14. The media of claim 8 wherein the differences in the report are computed according to an edit distance algorithm.

15. A computer-implemented method of automatically analyzing text documents in which a computer performs the following steps:

comparing text from a subject text to text of a plurality of given text templates, each text template containing at least one paragraph of text;

determining which given text template or text templates has text that matches the text from the subject text document to a given degree of correspondence; and

generating a report of the differences between the text from the subject text document and the text of the matching text template or text templates.

16. The computer-implemented method of claim 15 further comprising the steps of:

comparing a family of specimen text documents;

identifying one paragraph of text within one of the family of specimen text documents that most closely matches a paragraph of text in all of the other specimen text documents, as compared to all of the other paragraphs in the one specimen text document; and

generating one of the text templates containing at least the one identified paragraph of text.

17. The computer-implemented method of claim 16 wherein the identifying the paragraph that most closely matches step uses an edit distance algorithm.

18. The computer-implemented method of claim 16 wherein the identifying the paragraph that most closely matches step uses a longest common subsequence algorithm.

19. The computer-implemented method of claim 15 wherein the determining matches to a degree of correspondence step uses an edit distance algorithm.

20. The computer-implemented method of claim 15 wherein the determining matches to a degree of correspondence step uses a longest common subsequence algorithm.

21. The computer-implemented method of claim 15 wherein the differences in the report are computed according to an edit distance algorithm.