Computer-readable medium having sentence dividing program stored thereon, sentence dividing apparatus, and sentence dividing method

• US 9,009,023 B2
• Filed: 03/27/2008
• Issued: 04/14/2015
• Est. Priority Date: 06/29/2007
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Fuji, Masaru, Nagase, Tomoki, Okura, Seiji, Ushioda, Akira
• Primary Examiner(s): Shah, Paras D
• Application Number:
  US12/056,432
• Publication Number:
  US 20090006080A1
• Time in Patent Office: 2,574 Days
• Field of Search: 704/1, 704/2, 704/3, 704/8, 704/9, 704/10, 715256-264
• US Class Current: 704/9
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06F 40/53 Processing of non-Latin text kana-to-kanji conversion G06F40/129; vowelisation G06F40/232

Abstract

A typical sentence having a specific typical characteristic in the sentence is divided. A division target typical sentence is divided on the basis of a small clause definition. The sentence is divided where positions suitable for dividing the typical sentence based on the structure are expressed by a user. A small clause string including small clauses that serve as independent sentences is created after the division. The small clause string is compared to the structure patterns, and a structure pattern that is determined to match the small clause string is selected as a result of the typical sentence division.

Claims

1. A non-transitory computer-readable medium having a sentence dividing program stored thereon and a processor, the program causing the processor to execute:

receiving a typical sentence from a user via an input device;

dividing, using the processor, under a circumstance where positions suitable for dividing the typical sentence based on the structure are expressed by a user using property information of the sentence and are previously stored as a small clause definition in a first storage unit, a division target typical sentence on the basis of the small clause definition stored in the first storage unit, and of creating a small clause string including small clauses that serve as independent sentences after the division, the small clause string comprising two or more words;

analyzing, regarding the small clause string, dependency between the small clauses constituting the small clause string;

eliminating, according to a result of the analysis, a small clause that modifies a skeletal small clause that serves as a skeleton in the structure of the typical sentence to transform the small clause string into a skeletal small clause string including only skeletal small clauses;

comparing, using the processor, under a circumstance where structures suitable for the typical sentence are expressed by the user as combinations with each small clause definition and are previously stored as structure patterns in a second storage unit, the skeletal small clause string with the structure patterns stored in the second storage unit; and

selecting, using the processor, as a division result of the typical sentence, a structure pattern that is determined to match the skeletal small clause string,

wherein the comparing includes comparing the skeletal small clause string with the structure patterns,

wherein if a number of small clauses that link to a pointer-set small clause is one, a next small clause on the left is selected, and

wherein if the number of small clauses that link to the pointer-set small clause is not one, the next small clause on the left and small clauses that link to the small clause on the left is replaced with a trace.

2. The non-transitory computer-readable medium having the sentence dividing program stored thereon according to claim 1, further comprising:

evaluating, when a plurality of structure patterns are determined to match the small clause string, a degree of matching of the small clause string and each of the plurality of structure patterns; and

sorting, according to a result of the evaluation, the order of the plurality of structure patterns that are determined to match the small clause string.

3. A sentence dividing apparatus for dividing a typical sentence having a specific typical characteristic in a structure of the sentence, the apparatus comprising:

a memory; and

a processor coupled to the memory and configured to divide the typical sentence by;

receiving the typical sentence from a user via an input device;

previously storing a small clause definition in the memory, in which positions suitable for dividing the typical sentence based on the structure are expressed by a user using property information of the sentence;

previously storing structure patterns in the memory, in which structures suitable for the typical sentence are expressed as combinations with each small clause definition;

dividing a division target typical sentence on the basis of the small clause definition stored in the memory, and creating a small clause string including small clauses that serve as independent sentences after the division, the small clause string comprising two or more words;

analyzing, regarding the small clause string, dependency between the small clauses constituting the small clause string;

eliminating, according to a result of the analysis, a small clause that modifies a skeletal small clause that serves as a skeleton in the structure of the typical sentence to transform the small clause string into a skeletal small clause string including only skeletal small clauses;

comparing the skeletal small clause string with the structure patterns stored in the memory; and

selecting, as a division result of the typical sentence, a structure pattern that is determined to match the skeletal small clause string, wherein the comparing includes comparing the skeletal small clause string with the structure patterns,

wherein if a number of small clauses that link to a pointer-set small clause is one, a next small clause on the left is selected, and

wherein if the number of small clauses that link to the pointer-set small clause is not one, the next small clause on the left and small clauses that link to the small clause on the left is replaced with a trace.

4. A sentence dividing method for dividing a typical sentence having a specific typical characteristic in a structure of the sentence, the method comprising:

receiving the typical sentence from a user via an input device;

dividing, using a processor, under a circumstance where positions suitable for dividing the typical sentence based on the structure are expressed by a user using property information of the sentence and are previously stored as a small clause definition in a first storage unit, a division target typical sentence on the basis of the small clause definition stored in the first storage unit, and of creating a small clause string including small clauses that serve as independent sentences after the division, the small clause string comprising two or more words;

analyzing, regarding the small clause string, dependency between the small clauses constituting the small clause string;

eliminating, according to a result of the analysis, a small clause that modifies a skeletal small clause that serves as a skeleton in the structure of the typical sentence to transform the small clause string into a skeletal small clause string including only skeletal small clauses;

comparing, using the processor, under a circumstance where structures suitable for the typical sentence are expressed by the user as combinations with each small clause definition and are previously stored as structure patterns in a second storage unit, the skeletal small clause string with the structure patterns stored in the second storage unit; and

selecting, using the processor, as a division result of the typical sentence, a structure pattern that is determined to match the skeletal small clause string,

wherein the comparing includes comparing the skeletal small clause string with the structure patterns,

wherein if a number of small clauses that link to a pointer-set small clause is one, a next small clause on the left is selected, and

wherein if the number of small clauses that link to the pointer-set small clause is not one, the next small clause on the left and small clauses that link to the small clause on the left is replaced with a trace.