Language processing device, language processing method and language processing program

• US 7,664,631 B2
• Filed: 12/20/2005
• Issued: 02/16/2010
• Est. Priority Date: 03/09/2005
• Status: Expired due to Fees

• Current Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Original Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Inventors: Onda, Masanori, Masuichi, Hiroshi, Yoshimura, Hiroki, Sugihara, Daigo, Ashikaga, Hideaki, Kato, Masahiro, Satake, Masanori, Kimura, Shunichi, Itonori, Katsuhiko
• Primary Examiner(s): Hudspeth; David R
• Assistant Examiner(s): Rider; Justin W
• Application Number:
  US11/311,416
• Publication Number:
  US 20060206307A1
• Time in Patent Office: 1,519 Days
• Field of Search: 704/9
• US Class Current: 704/9
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation

Abstract

According to an aspect of the present invention, a language processing device has a text input section and an anaphora analysis section. The text input section acquires text data. The anaphora analysis section analyzes whether a correct anaphora relation is included in the text data acquired by the text input section. The correct anaphora relation has an anaphor and an antecedent corresponding to the anaphor.

Claims

1. A language processing device comprising:

i. a text input section that acquires text data;

ii. an anaphora analysis section that analyzes whether a correct anaphora relation is included in the text data acquired by the text input section;

iii. a determining section that determines whether the text data is fragmented based on corrections of the anaphora relation; and

iv. a document format information extracting section that extracts document format information from the text data acquired by the text input section, wherein the anaphora analysis section analyzes whether the correct anaphora relation is included in the acquired text data based on the document format information, when the anaphora analysis section analyzes the anaphora relation and the correct anaphora relation is not included in the text data, the anaphora analysis section adds an expression suitable to an antecedent or an anaphora or a user adds the expression, and the correct anaphora relation has an anaphor and an antecedent corresponding to the anaphor.

2. The language processing device according to claim 1, wherein the anaphora analysis section has a text part determination section, the text part determination section determines whether the text data acquired by the text input section is started from a beginning of a sentence.

3. The language processing device according to claim 2, wherein when the text part determination section determines that the acquired text data is not started from the beginning of a sentence, the anaphora analysis section analyzes that the correct anaphora relation is not included in the acquired text data.

4. The language processing device according to claim 1, wherein when the anaphora analysis analyzes that the correct anaphora relation is not included in the acquired text data, the anaphora analysis determines an antecedent corresponding to an anaphor from another part of text data.

5. The language processing device according to claim 1, wherein the anaphora analysis section has a text part determination section, the text part determination section determines whether the text data acquired by the text input section continues from a previous sentence.

6. The language processing device according to claim 1, further comprising:

an anaphor detection section that detects an anaphor from the text data acquired by the text input section,

wherein the anaphora analysis section analyzes whether the correct anaphora relation is included in the acquired text data based on the anaphor detected by the anaphor detection section.

7. The language processing device according to claim 1, further comprising:

a document format information extracting section that extracts document format information from the text data acquired by said text input section; and

an anaphor detection section that detects an anaphor from the text data acquired by the text input section;

wherein the anaphora analysis section analyzes the correct anaphora relation based on the document format information and the anaphor detected by the anaphor detection section.

8. The language processing device according to claim 6, further comprising:

a document type classification section that extracts a document type of the text data acquired by the text input section,

wherein the anaphor detection section detects an anaphor from the text data based on an anaphor detection rule corresponding to the type of the document.

9. The language processing device according to claim 1, further comprising:

a document type classification section that classifies a document type of the text data acquired by the text input section,

wherein the document format information extracting section extracts document format information from the text data based on a document format information extraction rule corresponding to the type of document.

10. The language processing device according to claim 1, wherein the anaphora analysis section generates a rule for analyzing an anaphora relation based on the type of document.

11. The language processing device according to claim 1, wherein the anaphora analysis section is a binary-value classifier which decides whether the text data discontinues.

12. The language processing device according to claim 1, wherein the anaphora analysis section analyzes the anaphora relation based on a score, each word in the text data having a score.

13. A language processing method comprising:

acquiring text data;

extracting document format information from the text data acquired by the text input section;

analyzing whether a correct anaphora relation is included in the acquired text data based on the document format information;

determining whether the acquired text data is fragmented based on the correctness of the anaphora relation; and

adding an expression suitable to an antecedent or an anaphora when the correct anaphora relation is not included in the text data,

wherein the correct anaphora relation has an anaphor and an antecedent corresponding to the anaphor, and

at least one of the above steps is preformed by a processor.

14. The language processing method according to claim 13, wherein the analyzing is performed based on whether the acquired text data is started from a beginning of a sentence.

15. A storage medium readable by a computer, the storage medium storing a program of instructions executable by the computer to perform a function for processing language, the function comprising:

acquiring text data;

extracting document format information from the text data acquired by the text input section;

analyzing whether a correct anaphora relation is included in the acquired text data based on the document format information;

determining whether the text data is fragmented based on the correctness of the anaphora relation; and

adding an expression suitable to an antecedent or an anaphora when the anaphora analysis section analyzes the anaphora relation and the correct anaphora relation is not included in the text data,

wherein the correct anaphora relation has an anaphor and an antecedent corresponding to the anaphor.