Apparatus for and method of analyzing chinese

• US 7,983,899 B2
• Filed: 12/10/2004
• Issued: 07/19/2011
• Est. Priority Date: 12/10/2003
• Status: Expired due to Fees

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation), Toshiba Digital Solutions Corporation (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Izuha, Tatsuya
• Primary Examiner(s): Wozniak; James S
• Assistant Examiner(s): Shah; Paras
• Application Number:
  US11/008,253
• Publication Number:
  US 20050154579A1
• Time in Patent Office: 2,412 Days
• Field of Search: 704/9, 704/8, 704/1, 704/2, 704/7
• US Class Current: 704/9
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/53 Processing of non-Latin text kana-to-kanji conversion G06F40/129; vowelisation G06F40/232
• CPC Class Codes: G06F 40/55 Rule-based translation

Abstract

An apparatus for analyzing Chinese according to one aspect of the present invention includes a dependency structure analyzer analyzing a dependency relationship between words by extracting a parent word being a dependency destination of each of the words and a child word being a dependent from each of the words. The words are obtained by dividing a Chinese sentence. The apparatus also includes a lihe-word processor referring to lihe-word information that includes a first word and capable of being a part of a lihe-word and a second word forming the lihe-word with the first word. The lihe-word processor detects the first word and the second word from the words analyzed, and then changes a dependency destination of a word depending on both the first word and the second word to the lihe-word formed by combining the first word with the second word.

Claims

1. An apparatus for analyzing Chinese language, comprising:

an input unit inputting a Chinese sentence;

a morpheme analyzer dividing the Chinese sentence into words;

a dependency relationship analyzer analyzing a dependency relationship between a parent word being a dependency destination of each of the words and a child word being a dependent from each of the words, the words being obtained by dividing the input Chinese sentence;

a memory unit that stores lihe-word information that registers a first word being a Chinese morpheme and capable of being a part of a lihe-word with a plurality of second words each forming a lihe-word with the first word, does not include words inserted between the first word and the second word, and lihe-words formed by the first word and the second words in the lihe-word information are not grouped based on similarity with other lihe-words;

a lihe-word processor detecting the first word and a second word from the dependency relationship, based on the lihe-word information without using the similarity of the lihe-word with other lihe-words, and the lihe-word processor changing a dependency destination of a word depending on both the first word and the second word to the lihe-word formed by combining the first word with the second word in the dependency relationship;

a generating unit generating a sentence based on the lihe-word formed by combining the first word with the second word and the word depending on the lihe-word by changing the dependency destination; and

an output unit outputting the sentence.

2. The apparatus according to claim 1, wherein:

the memory unit further stores dependency analysis information in which the parent word is associated with the child word;

the dependency relationship analyzer registers the parent word and the child word of each divided word in the dependency analysis information; and

the lihe-word processor detects the first word and the second word from the dependency relationship based on the lihe-word information, and changes the dependency destination of the word depending on both the first word and the second word to the lihe-word in the dependency analysis information.

3. The apparatus according to claim 2,

wherein the memory unit further stores Chinese dictionary information in which a word being a morpheme of Chinese sentences, a part of speech of the word, and the lihe-word information are associated with each other,

the dependency analysis information includes the word and the part of speech associated with the word, and

the lihe-word processor detects, based on the lihe-word information, the first word and the second word from the words analyzed, retrieves the Chinese dictionary information for the lihe-word formed by combining the first word and the second word, registers the lihe-word retrieved and the part of speech thereof in the dependency analysis information, and changes the dependency destination of the word depending on both the first word and the second word to the lihe-word in the dependency analysis information.

4. The apparatus according to claim 3,

wherein the dependency analysis information includes modification information associated with the word, the modification information indicating a type of modification for the parent word, and

the lihe-word processor, when changing the dependency destination of the word depending on both the first word and the second word to the lihe-word, changes the modification information of the word based on the part of speech of the lihe-word.

5. The apparatus according to claim 4, wherein the lihe-word processor, when changing the dependency destination of a word having a pronominal modifier as the modification information to the lihe-word, changes the modification information to a predicate modifier.

6. The apparatus according to claim 3, wherein the generator unit comprises a meaning analyzer generating the sentence by analyzing a meaning and contents of the Chinese sentence, using the Chinese dictionary information and the dependency analysis information.

7. The apparatus according to claim 6, wherein the generator unit comprises a translator generating the sentence by translating the dependency relationship of the Chinese sentence into a dependency relationship of an equivalent sentence in a language other than Chinese.

8. The apparatus according to claim 7, wherein the language other than Chinese is English.

9. The apparatus according to claim 7, wherein the generator unit comprises a translated sentence generator generating an equivalent sentence in the language other than Chinese, based on the dependency relationship of the equivalent sentence translated by the translator.

10. The apparatus according to claim 9, wherein the language other than Chinese is English.

11. The apparatus according to claim 9, further comprising a display unit displaying the equivalent sentence generated by the translated sentence generator.

12. The apparatus according to claim 9, wherein the output unit comprises a printer printing the equivalent sentence generated by the translated sentence generator.

13. The apparatus of claim 1, further comprising:

a syntax analyzer that includes the dependency relationship analyzer and the lihe-word processor.

14. A method of analyzing Chinese language implemented by a computer processor executing instructions, comprising:

inputting, by a computer receiving input from a user interface, a Chinese sentence;

dividing, by the computer, the sentence into words;

analyzing, by the computer, a dependency relationship between a parent word being a dependency destination of each of the words and a child word being a dependent from each of the words;

detecting, by the computer, a first word being a Chinese morpheme and capable of being a part of a lihe-word and a second word forming the lihe-word with the first word, from the dependency relationship, based on a lihe-word information in a memory unit, without using the similarity of the lihe-word with other lihe-words, wherein the lihe-word information registers the first word with a plurality of second words each forming a lihe-word with the first word, but does not include words inserted between the first word and the second word, and the lihe-word formed by the first word and the second words in the lihe-word information not being grouped based on similarity with other lihe-words;

changing, by the computer, a dependency destination of a word depending on both the first word and the second word to the lihe-word formed by combining the first word with the second word in the dependency relationship;

generating, by the computer, a sentence based on the lihe-word formed by combining the first word with the second word and the word depending on the lihe-word by changing the dependency destination; and

outputting the sentence.

15. A computer program product having a non-transitory computer readable medium including programmed instructions, wherein the instructions, when executed by a computer, cause the computer to perform a method which comprises:

inputting a Chinese sentence;

dividing the sentence into words;

analyzing a dependency relationship between a parent word being a dependency destination of each of the words and a child word being a dependent from each of the words;

detecting, by the computer, a first word being a Chinese morpheme and capable of being a part of a lihe-word and a second word forming the lihe-word with the first word, from the dependency relationship, based on a lihe-word information in a memory unit without using the similarity of the lihe-word with other lihe-words, wherein the lihe-word information registers the first word with a plurality of second words each forming a lihe-word with the first word, but does not include words inserted between the first word and the second word, and the lihe-word formed by the first word and a second word in the lihe-word information being not grouped based on similarity with other lihe-words;

changing a dependency destination of a word depending on both the first word and the second word to the lihe-word formed by combining the first word with the second word in the dependency relationship;

generating a sentence based on the lihe-word formed by combining the first word with the second word and the word depending on the lihe-word by changing the dependency destination; and

outputting the sentence.

16. The computer program product of claim 15, the method further comprising:

combining the first word with the second word in the dependency relationship to form the lihe-word.