Method and system for document translation and extraction

• US 5,629,846 A
• Filed: 09/28/1994
• Issued: 05/13/1997
• Est. Priority Date: 09/28/1994
• Status: Expired due to Term

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Crapo, Andrew W.
• Primary Examiner(s): Hayes, Gail O.
• Assistant Examiner(s): Poinvil, Frantzy
• Application Number:
  US08/313,961
• Time in Patent Office: 958 Days
• Field of Search: 364/419.1, 395/600, 395/145, 395/146, 395/148, 395/500
• US Class Current: 708/705
• CPC Class Codes: G06F 40/103 Formatting, i.e. changing of presentation of documents automatic justification G06F40/189; automatic line break hyphenation G06F40/191

Abstract

A method and system for translating an electronic document from one format to an electronic document in a second format. Selected portions from a source document are extracted and transformed into the format of a target document. A translation rule set is then deduced from the extracted portions and the transformed portions. The translation rule set is then applied to the source document, producing a first draft. If the translation rule set is unable to translate a portion from the source document, then the user is notified of the untranslatable portion. The user then provides examples of how the untranslatable portion should be translated into the format of the target document. The translation rule set is then modified in accordance with the examples. Next, the modified translation rule set is applied to the source document, producing a second draft. The above steps are repeated until the source document has been completely translated into the format of the target document or until the user is satisfied with the translation.

Claims

1. A method for translating an electronic source document having a first format into an electronic target document having a second format, the method comprising the steps of:

selecting portions from the source document having various constructs and formats;

extracting the selected portions from the source document;

transforming the format of the extracted portions into the second format of the electronic target document;

deducing a translation rule set from the extracted portions and the transformed portions;

applying the translation rule set to the electronic source document;

producing a first draft of the electronic target document as the translation rule set is applied to the electronic source document; and

identifying portions from the electronic source document which were unable to be translated into the target document.

2. The method according to claim 1, further comprising modifying the translation rule set to account for the identified untranslatable portions.

3. The method according to claim 2, wherein the step of modifying includes interactively entering example sets of various portions and segments in the second format that correspond to the identified untranslatable portions.

4. The method according to claim 3, wherein the step of modifying comprises generating a new rule set containing rules for translating the identified untranslatable portions.

5. The method according to claim 4, further comprising applying the new rule set to the electronic source document.

6. The method according to claim 5, further comprising producing a second draft of the electronic target document as the new translation rule set is applied to the electronic source document.

7. The method according to claim 6, further comprising repeating the steps of identifying, modifying, generating, applying, and producing, until the electronic source document has been translated into the format of the electronic target document.

8. The method according to claim 1, wherein the step of transforming is performed by using an editor.

9. The method according to claim 1, wherein the step of identifying comprises providing an error log identifying the portions of the electronic source document that were untranslatable.

10. A method for translating an electronic source document having a first format into an electronic target document having a second format, the method comprising the steps of:

selecting portions from the source document having various constructs and formats;

extracting the selected portions from the source document;

transforming the format of the extracted portions into the second format of the electronic target document;

deducing a translation rule set from the extracted portions and the transformed portions;

applying the translation rule set to the electronic source document;

producing a first draft of the electronic target document as the translation rule set is applied to the electronic source document;

identifying portions from the electronic source document which were unable to be translated into the electronic target document;

modifying the translation rule set to account for the identified untranslatable portions, the modified translation rule set being a new rule set;

applying the new rule set to the electronic source document; and

repeating the steps of producing, identifying, modifying, and applying, until the target document is in a desired format.

11. The method according to claim 10, wherein the step of identifying comprises providing an error log identifying the portions of the electronic source document that were untranslatable.

12. The method according to claim 10, wherein the step of modifying includes interactively entering example sets of various portions and segments in the second format that correspond to the identified untranslatable portions.

13. A system for translating an electronic source document having a first format into an electronic target document having a second format, the system comprising:

means for selecting portions from the source document having various constructs and formats;

means for extracting the selected portions from the source document;

means for transforming the format of the extracted portions into the second format of the electronic target document;

means for deducing a translation rule set from the extracted portions and the transformed portions;

first means for applying the translation rule set to the electronic source document;

first means for producing a first draft of the electronic target document as the translation rule set is applied to the electronic source document; and

first means for identifying portions from the electronic source document which were unable to be translated into the target document.

14. The system according to claim 13, further comprising means for modifying the translation rule set to account for the identified untranslatable portions.

15. The system according to claim 14, wherein the modifying means includes means for interactively entering example sets of various portions and segments in the second format that correspond to the identified untranslatable portions.

16. The system according to claim 13, wherein the transforming means is an editor.

17. The system according to claim 14, wherein the modifying means comprises means for generating a new rule set containing rules for translating the identified untranslatable portions.

18. The system according to claim 17, further comprising second means for applying the new rule set to the electronic source document.

19. The system according to claim 18, further comprising second means for producing a second draft of the electronic target document as the new translation rule set is applied to the electronic source document.

20. The system according to claim 19, further comprising second means for identifying more portions from the electronic source document which were unable to be translated in the target document.

21. The system according to claim 19, wherein the first and second identifying means provides an error log identifying the portions of the electronic source document that were untranslatable.

22. A system for translating an electronic source document having a first format into an electronic target document having a second format, the system comprising:

means for selecting portions from the source document having various constructs and formats;

means for extracting the selected portions from the source document;

means for transforming the format of the extracted portions into the second format of the electronic target document;

means for deducing a translation rule set from the transformed portions;

first means for applying the translation rule set to the electronic source document;

means for producing a first draft of the electronic target document as the translation rule set is applied to the electronic source document;

means for identifying portions from the electronic source document which were unable to be translated into the electronic target document;

means for modifying the translation rule set to account for the identified untranslatable portions, the modified translation rule set being a new rule set;

second means for applying the new rule set to the electronic source document; and

means for repeating the steps of producing, identifying, modifying, and applying, until the target document is in a desired format.

23. The system according to claim 22, wherein the identifying means provides an error log identifying the portions of the electronic source document that were untranslatable.

24. The system according to claim 22, wherein the modifying means includes interactively entering example sets of various portions and segments in the second format that correspond to the identified untranslatable portions.