Method and system for extracting information from documents by document segregation

• US 20060177118A1
• Filed: 02/09/2005
• Published: 08/10/2006
• Est. Priority Date: 02/09/2005
• Status: Active Grant

• Current Assignee: International Business Machines Corporation, JP Morgan Chase Bank, N.A. (JP Morgan Chase & Co)
• Original Assignee: International Business Machines Corporation, JP Morgan Chase Bank, N.A. (JP Morgan Chase & Co)
• Inventors: Nowicki, Tomasz J., Reilly, Michael J., Walach, Eugene, Ibikunle, Tayo, Tresser, Charles P., Sachar, Howard E., Karnin, Ehud, Liberis, William A., Weeshoff, David A.
• Granted Patent: US 7,983,468 B2
• Time in Patent Office: Days
• US Class Current: 382/137
• CPC Class Codes: G06Q 20/042 characterized in that the payment protocol involves at least one cheque
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/262 using context analysis, e.g. lexical, syntactic or semantic context

Abstract

A method (and system) for extracting information from a document, includes segregating a set of documents from a plurality of documents based on a likelihood that at least one document in the set of documents carries an instance of a preset information.

Claims

1. A method for extracting information from a document, comprising:

segregating a set of documents from a plurality of documents based on a likelihood that at least one document in said set of documents carries an instance of a preset information.

2. The method according to claim 1, further comprising obtaining information from said plurality of documents.

3. The method according to claim 2, wherein said information is obtained from at least one of a front of said documents and a back of said plurality of documents.

4. The method according to claim 2, wherein said information is obtained from at least one field on said plurality of documents.

5. The method according to claim 2, wherein said information is obtained from a plurality of fields on said plurality of documents.

6. The method according to claim 4, wherein said at least one field comprises a field comprising payee information.

7. The method according to claim 1, wherein said plurality of documents each comprise a check.

8. The method according to claim 2, further comprising:

combining previously known information with information from said obtaining information.

9. The method according to claim 1, wherein said set of documents comprises a known number of documents.

10. The method according to claim 1, wherein said set of documents comprises an approximately known number of documents.

11. The method according to claim 1, wherein said plurality of documents comprises at least one batch of documents having a likelihood of carrying an instance of said preset information.

12. The method according to claim 11, wherein said at least one batch of documents appears sequentially within said plurality of documents.

13. The method according to claim 11, wherein said at least one batch of documents comprises at least one subset of documents in which each document in said subset of documents comprises said preset information.

14. The method according to claim 2, wherein said obtaining information comprises at least one of a plurality of information obtaining techniques.

15. The method according to claim 14, wherein said plurality of information obtaining techniques comprises preprocessing data associated with an image of said plurality of documents.

16. The method according to claim 14, wherein said plurality of information obtaining techniques comprises segmentation.

17. The method according to claim 14, wherein said plurality of information obtaining techniques comprises information extraction.

18. The method according to claim 14, wherein said plurality of information obtaining techniques comprises classification.

19. The method according to claim 14, wherein said plurality of information obtaining techniques comprises interpretation.

20. The method according to claim 1, wherein said preset information comprises at least one of a single instance of preset information and a plurality of instances of preset information.

21. The method according to claim 1, further comprising:

determining whether said preset information is present on documents of said plurality of documents by indicia recognition.

22. The method according to claim 21, wherein said indicia recognition comprises at least one of handwritten character recognition, printed text recognition and stamped text recognition.

23. The method according to claim 2, further comprising:

weighting said information obtained from said plurality of documents.

24. The method according to claim 1, further comprising:

introducing a marker document into said plurality of documents.

25. A system of extracting information from a document, comprising:

a segregation unit that segregates a set of documents from a plurality of documents based on a likelihood that at least one document in said set of documents carries an instance of a preset information.

26. A system of extracting information from a document, comprising:

means for recognizing indicia in a plurality of documents; and

means, coupled to said recognizing means, for segregating a set of documents from said plurality of documents based on a likelihood that at least one document in said set of documents carries an instance of a preset information in said indicia.

27. A signal-bearing medium tangibly embodying a program of machine readable instructions executable by a digital processing apparatus to perform a method for extracting information from a document, said method comprising:

segregating a set of documents from a plurality of documents based on a likelihood that at least one document in said set of documents carries an instance of a preset information.

28. A method for deploying computing infrastructure, comprising integrating computer-readable code into a computing system, wherein the computer readable code in combination with the computing system is capable of performing a method for extracting information from a document, said method for extracting information from a document, comprising:

segregating a set of documents from a plurality of documents based on a likelihood that at least one document in said set of documents carries an instance of a preset information.

29. A system of extracting information from a document, comprising:

a recognition unit that recognizes indicia including handwritten characters on a document; and

a segregating unit that segregates a batch of documents from a plurality of documents based on a likelihood that at least one document in said batch of documents carries an instance of a preset information in said indicia,

wherein said recognition unit recognizes said handwritten text on a front side and a rear side of said document.

30. The system according to claim 29, wherein said document comprises a check.

31. The system according to claim 30, wherein said recognition unit recognizes a payment amount handwritten on said check to determine if said payment amount is in a preset payment amount range.