System for distinguishing names of organizations in Asian writing systems

• US 7,136,805 B2
• Filed: 06/11/2002
• Issued: 11/14/2006
• Est. Priority Date: 06/11/2002
• Status: Expired due to Fees

• Current Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Original Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Inventors: Wang, Xinkai
• Primary Examiner(s): McFadden, Susan
• Application Number:
  US10/166,377
• Publication Number:
  US 20030229487A1
• Time in Patent Office: 1,617 Days
• Field of Search: 704/4, 704 7- 8
• US Class Current: 704/9
• CPC Class Codes: G06F 40/53 Processing of non-Latin text kana-to-kanji conversion G06F40/129; vowelisation G06F40/232

Abstract

A system for distinguishing names of organizations in Chinese text, which includes a computer. The computer has at least an input, an output, a processor, and a memory and storage arrangement. Data is accessible by the processor that includes at least two groups of terms that frequently respectively form the first and last terms of the names of organizations in Chinese. The system includes software, which when performed by the computer causes computer processing including identifying terms in Chinese text that has been input to the computer for terms corresponding to those in the groups in the data; comparing the location in the Chinese text of each identified term from one of the groups to identified terms from the other group, and if predefined conditions are met, determining that the identified term from one group forms the name of an organization with an identified term from the other group.

Claims

1. A system for distinguishing a name of an organization in Chinese text, the system comprising:

a receiving part that receives Chinese text in electronic format;

an identification part that identifies terms in the received Chinese text that belong to one group of terms that are frequently used to form a first term of a name of an organization, and another group of terms that are frequently used to form a last term of a name of an organization in Chinese;

a comparing part that compares a location in the Chinese text of the identified term that belongs to the other group to the identified term that belongs to the one group, and if predetermined conditions are met, determines that the identified term that belongs to the other group forms a name of an organization with the identified term that belongs to the one group; and

an output part that outputs information of at least one of the identified term and the name of the organization.

2. The system of claim 1, wherein the one group of terms includes proper nouns, said identifying a term that belongs to the one group of terms includes identifying English text appearing in the Chinese text, and if predefined conditions are met, determining that the identified term that belongs to the another group forms the name of an organization including the English text.

3. The system of claim 1, wherein one of the predefined conditions is that the identified term that belongs to the other group must appear in the text after the identified term that belongs to the one group.

4. The system of claim 3, wherein the location of the identified term is compared to the other identified term only if both identified terms appear in the same phrase.

5. The system of claim 4, wherein the predefined conditions include that two terms that belong to the other group cannot together form a name of an organization.

6. The system of claim 1, wherein the predefined conditions include a plurality of rules applied in sequential order.

7. The system of claim 1, wherein the identification part defines an indicator that tends to indicate that a term immediately following the identified indicator in the text is the first term of a name of an organization and ending with the identified term that belongs to the other group.

8. A process for distinguishing a name of an organization appearing in a phrase or sentence in Chinese, the process comprising:

(a) establishing data including at least one group of terms comprising indicators that tend to indicate that a term immediately following an indicator is a first term of a name and of an organization, and another group of terms comprising terms frequently used to form a last term of a name of an organization in Chinese;

(b) identifying in the Chinese phrase or sentence terms corresponding to the data;

(c) comparing a location in the Chinese text of the identified term that belongs to the other group to the identified term that belongs to the one group, and if predefined conditions are met, determining that the identified term that belongs to the other group forms the name of an organization together with text immediately following, but not including, the identified term that belongs to the one group;

(d) outputting information of at least one of the identified term and the name of the organization.

9. The process of claim 8, further comprising identifying English text appearing in the Chinese phrase or sentence, comparing the location of the identified term that belongs to the other group to the identified English text, and if predefined conditions are met, determining that the identified term that belongs to the other group forms the name of an organization including English text.

10. The process of claim 8, wherein one of the predefined conditions is that the identified term that belongs to the other group must appear in the text after the identified term that belongs to the one group.

11. The process of claim 8, wherein the predefined conditions include that two terms that belong to the other group cannot together form the name of an organization.

12. The process of claim 8, further comprising identifying a proper noun in the text, comparing the location in the text of the identified term that belongs to the other group to the identified proper noun, and if predefined conditions are met, determining that the identified term that belongs to the other group forms a name of an organization with the identified proper noun.

13. The process of claim 12, wherein said identifying a proper noun in the text and comparing the location of the identified term that belongs to the other group to the identified proper noun is performed prior to said comparing the location of the identified term that belongs to the other group to the identified term that belongs to the one group.

14. The process of claim 8, wherein the indicators include a comma or a possessive or aspect indicator in Chinese.

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 distinguishing a name of an organization in a phrase or sentence in Chinese input to the computer, the function comprising:

receiving Chinese text in electronic format;

identifying terms in the received Chinese text that belong to one group of terms that are frequently used to form a first term of a name of an organization, and another group of terms that are frequently used to form a last term of a name of an organization in Chinese;

comparing a location in the Chinese text of the identified term that belongs to the other group to the identified term that belongs to the one group;

if predetermined conditions are met, determining that the identified term that belongs to the other group forms a name of an organization with the identified term that belongs to the one group; and

outputting information of at least one of the identified term and the name of the organization.

16. The storage medium of claim 15, wherein the function further comprises identifying English text appearing in the Chinese phrase or sentence, comparing a location of the identified term that belongs to the other group to the identified English text, and if predefined conditions are met, determining that the identified term that belongs to the other group forms a name of an organization including English text.

17. The storage medium of claim 15, wherein one of the predefined conditions is that the identified term that belongs to the other group must appear in the text after the identified term that belongs to the one group.

18. The storage medium of claim 15, wherein the predefined conditions include that two terms that belong to the other group cannot together form a name of an organization.

19. The storage medium of claim 15, wherein the function further comprises identifying an indicator that tends to indicate that a term immediately following an identified indicator in a phrase or sentence in Chinese is the first term of the name of an organization.

20. The storage medium of claim 15, wherein identifying an indicator is performed after said comparing, and said indicator includes any name of an organization determined in said comparing and appearing in the phrase or sentence prior to an identified term that belongs to the other group which has not been determined to form the name of an organization in said comparing.

21. A method for distinguishing a name of an organization in a phrase or sentence in Chinese input to a computer, the method comprising:

receiving Chinese text in electronic format;

identifying terms in the received Chinese text that belong to one group of terms that are frequently used to form a first term of a name of an organization, and another group of terms that are frequently used to form a last term of a name of an organization in Chinese;

if predefined conditions are met, determining that the identified term that belongs to the other group forms a name of an organization with the identified term that belongs to the one group; and

outputting information of at least one of the identification term and the name of the organization.