JP2002117028A - Device and method for dictionary generation and recording medium with recorded dictionary generating program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device and a method for dictionary generation which can accurately generate a dictionary of good quality between natural languages even without a translation corpus with sufficient capacity. SOLUTION: A 1st natural language-to-2nd natural language dictionary entry is taken out of a 1st natural language-to-2nd natural language dictionary storage part. A 3rd natural language dictionary entry corresponding to the 2nd natural language entry of the 1st natural language-to-2nd natural language dictionary entry is obtained from a 2nd natural language-to-3rd natural language dictionary storage part. The 2nd natural language entry of the 1st natural language-to-2nd natural language dictionary entry is replaced with this 3rd natural language entry in order to generate and store a translation dictionary of the 1st natural language to the 3rd natural language which has indexes and parts of speech of the 1st natural language and also indexes and parts of speech of the 3rd natural language as translations of 1st natural language entries in a 1st natural language-to-3rd natural language dictionary storage part.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dictionary creation device used in, for example, a natural language processing system, an information retrieval system, a machine translation system, etc., for creating a bilingual dictionary of different natural language pairs.

[0002]

2. Description of the Related Art In recent years, with the development of the Internet and the like, opportunities to come in contact with documents written in various natural languages have been increasing, and opportunities for users to come into contact with natural languages which are not always fluent have also increased. Along with this, in order to use documents and information written in natural languages that users are not necessarily proficient in, multilingual search and multilingual information processing, especially for digitized data,
The demand for multilingual translations and the like has increased, and dictionaries of various language pairs have been required to be used for them.
Therefore, it is required to create bilingual dictionaries of various language pairs with good quality.

In order to obtain a high-quality dictionary, it is common to create it manually (Shirai et al., "The number of combined pattern pairs necessary for Japanese-to-English machine translation and its collection method", Information Processing Society of Japan research report, Nature Language Processing Research Group, 95-NL-110-7, pp.43-50) Due to the enormous amount of work, creating a dictionary manually requires a great deal of cost, manpower, and time. In particular, in recent years, expansion to various language pairs has been required, and it is desired to create a high-quality dictionary automatically or semi-automatically without using any manual operation.

As a method for automatically creating a dictionary of natural language pairs, there is a method of using a bilingual corpus of natural language pairs to be created. A bilingual corpus is an electronic version of the same document written in different languages. To use the bilingual corpus, morphological analysis is performed on the bilingual corpus of the natural language pair with sentence correspondence, and the frequency of occurrence in the bilingual corpus for all word pairs of the language pair appearing in the bilingual sentence is calculated. , Word pairs with high frequency of appearance have high similarity, calculate word similarity, then parse the bilingual corpus to compare syntactic structures, and extract translation patterns from word similarity and syntactic structure comparisons, etc. (Kitamura et al., "Automatic acquisition of translation knowledge from bilingual bilingual corpora", Proc. Of the 8th Annual Conference of the Japanese Society for Artificial Intelligence, pp.645-648, 1994).

[0005] However, although a large number of corpora have recently become available on computers, the type and amount of the bilingual corpus are small, and a large-scale dictionary capable of practically using a bilingual corpus of various language pairs can be used. It is difficult to get enough to make (KWChurch and R.L.Mercer, "Introduction
to the special issue on computational linguistics
using large corpora ", Computational Linguistics Vo
l.19, PP1-24, 1993). In particular, the number of bilingual corpora with sentence correspondence is even smaller, and bilingual corpuses without sentence correspondence often do not always have accurate translations, such as not being one-to-one correspondence at the sentence level or being translated. .

[0006]

Of the above-mentioned prior arts, the manual method requires too much cost, manpower and time compared to the diversification of the required language pairs of the dictionary and the increase in the needs. There is a problem.

In the method using a bilingual corpus,
The bilingual corpus is small in type and quantity, and it has a problem that it is difficult to obtain a bilingual corpus sufficient for creating a large-scale dictionary that can withstand practical use for various language pairs.

[0008] The present invention has been made in view of the above,
It is an object of the present invention to provide a dictionary creating apparatus and method capable of accurately creating a dictionary between high-quality natural languages without a sufficient amount of a bilingual corpus, and a recording medium recording a dictionary creating program. It is in.

[0009]

According to one aspect of the present invention, there is provided a dictionary creating apparatus for creating a bilingual dictionary of a natural language pair, comprising: a first natural language heading and its part of speech. And a first natural language versus a second natural language dictionary storing a first natural language versus a second natural language dictionary having a bilingual expression of a second natural language and a part of speech of the first natural language entry. A second natural language-third natural language dictionary having a part, a heading of the second natural language and its part of speech, a bilingual expression of the third natural language which is a bilingual translation of the second natural language entry, and its part of speech. A second natural language-third natural language dictionary storage unit, and the first natural language-second
A dictionary entry reading unit for retrieving the first natural language pair to the second natural language dictionary entry from the natural language dictionary storage unit; and a third corresponding to the second natural language entry of the retrieved first natural language to second natural language dictionary entry. A dictionary creation that acquires a natural language entry from the second natural language to third natural language dictionary storage unit and replaces the second natural language entry of the first natural language to second natural language dictionary entry with the third natural language entry A management unit configured to replace the first natural language entry with the third natural language dictionary entry generated by replacing the second natural language entry of the first natural language entry with the second natural language dictionary entry with the third natural language entry; 1
A first natural language-third natural language bilingual dictionary having a natural language heading and its part of speech, a third natural language bilingual translation of the first natural language entry, and its part of speech, is created and stored. The gist of the present invention is to have one natural language-third natural language dictionary storage unit.

According to the first aspect of the present invention, the first natural language pair and the second natural language dictionary entry are fetched from the first natural language pair and the second natural language dictionary storage, and the fetched first natural language pair is extracted. A third natural language entry corresponding to the second natural language entry of the second natural language dictionary entry is obtained from the second natural language-to-third natural language dictionary storage unit, and the first natural language-to-second natural language entry is obtained by the third natural language entry. A first natural language having a heading of the first natural language and its part of speech by replacing the second natural language entry of the second natural language dictionary entry, and a third natural language bilingual translation and its part of speech of the first natural language entry; Create a bilingual dictionary of language versus third natural language
The first natural language versus the third natural language is stored in the natural language versus third natural language dictionary storage unit using the first natural language versus the second natural language dictionary and the second natural language versus the third natural language dictionary. Even if there is no dictionary or bilingual corpus of
A natural language versus third natural language dictionary can be created.

The present invention described in claim 2 is the same as the claim 1.
In the invention described in the above, further comprising a first natural language corpus information storage unit that stores statistical information of the first natural language corpus, wherein the dictionary creation management unit is configured to supply the first natural language corpus information supplied from the dictionary entry reading unit. An entry selection information providing unit that refers to the first natural language corpus information storage unit and uses the first natural language entry of the language-to-second natural language dictionary entry as a key to remove unnecessary first natural language-to-second natural language entries The gist is to have.

According to the present invention, the first natural language corpus information is stored using the first natural language entry of the first natural language versus second natural language dictionary entry supplied from the dictionary entry reading unit as a key. Section to remove unnecessary first natural language versus second natural language entries,
For example, when a dictionary limited to a field is created, unnecessary dictionary entries are not created.
The speed and efficiency of the creation and use of the natural language versus third natural language dictionary can be improved. In the case of creating a dictionary limited to a field, if the first natural language-to-second natural language dictionary and the second natural language-to-third natural language dictionary to be used can use the dictionary of the field, the reliability is higher. A field-specific first natural language versus third natural language dictionary can be created.

Further, the present invention according to claim 3 provides the invention according to claim 2.
The invention described in the above, further comprising a second natural language corpus information storage unit that stores statistical information of a second natural language corpus, wherein the entry selection information providing unit is configured to receive the first natural language corpus information from the dictionary entry reading unit. Using the second natural language entry of the natural language versus the second natural language dictionary entry as a key, referring to the second natural language corpus information storage unit and assigning translation word selection information / weighting information addition to assign the translation word selection information as weighting information The point is to have means.

According to the third aspect of the present invention, the second natural language corpus information is stored using the second natural language entry of the first natural language versus second natural language dictionary entry supplied from the dictionary entry reading unit as a key. For example, when there are a plurality of candidate translations from the first natural language to the second natural language in order to add translation word selection information as weighting information by referring to Is more appropriate, and among several possible translations,
It is possible to prioritize which translation candidates appear better. Further, even when there is one translation word candidate, the likelihood of the translation word candidate can be determined, and as a result,
A more reliable first natural language to third natural language dictionary can be created. In addition, since information for selecting a translated word is added at the stage of the second natural language, the same information can be used when only the third natural language is changed without changing the first natural language and the second natural language. When the third natural language is extended to various languages as a result, the first natural language-third natural language dictionary can be efficiently created.

According to a fourth aspect of the present invention, in the first, second or third aspect of the invention, the dictionary creation management unit stores the second natural language entry of the first natural language versus second natural language dictionary entry in the second natural language entry. When the third natural language entry cannot be replaced with the third natural language entry, the second natural language analysis unit analyzes the second natural language entry, and acquires the third natural language entry based on the analysis result of the second natural language analysis unit. And a replacement execution unit that replaces the second natural language entry with the obtained third natural language entry.

According to the present invention, when the second natural language entry of the first natural language-to-second natural language dictionary entry cannot be replaced with the third natural language entry, the second natural language entry cannot be replaced. In order to analyze the natural language entry and obtain a third natural language entry based on the analysis result and replace the second natural language entry with the obtained third natural language entry, for example, the second natural language entry is a compound word, Even if the notation is not registered in the second natural language-third natural language dictionary as it is, analysis is performed on the second natural language entry, such as division into sub-expression units constituting a compound word, and each sub-expression is analyzed. Is registered in the second natural language-third natural language dictionary, the second natural language entry of the first natural language-second natural language dictionary entry is replaced with the third natural language entry. Rukoto can, can be created as a result, as it is the third natural language entry also won the first natural language versus the third natural language dictionary for such difficult compound word of the replacement notation.

The present invention according to claim 5 provides the present invention according to claim 4.
The invention described in the above, further comprising a replacement rule storage unit storing a replacement rule for replacing the second natural language with the third natural language, wherein the dictionary creation management unit stores the analysis result of the second natural language analysis unit. A replacement rule search unit that searches the replacement rule storage unit based on the replacement rule, and obtains a replacement rule. The replacement execution unit uses the replacement rule extracted by the replacement rule search unit. Has a means for acquiring a third natural language entry from the second natural language / third natural language dictionary storage unit by using the plurality of replacement rules in combination.

According to the fifth aspect of the present invention, the replacement rule storage section is searched based on the analysis result of the second natural language analysis section, the replacement rule is extracted, and the extracted replacement rule is used. When there are a plurality of replacement rules, the plurality of replacement rules are used in combination to obtain a second natural language versus a third natural language.
In order to obtain the third natural language entry from the natural language dictionary storage unit, for example, if the second natural language entry is a compound word, a modified form, a combination thereof, or the like, the second natural language entry is the second
Even when not registered in the natural language-to-third natural language dictionary, a more complex result is obtained by using the analysis result of the second natural language entry, a replacement rule suitable for the analysis result, and the second natural language-to-third natural language dictionary. Of the first natural language to the second natural language dictionary entry more reliably for the second natural language entry.
The natural language entry can be replaced by a third natural language entry. Further, by using a plurality of replacement rules in combination, even if a replacement rule that can replace the analysis result of the second natural language entry in a lump is not stored in the replacement rule storage unit, the first natural language versus the second natural language entry can be used. The second natural language entry of the natural language dictionary entry is the third
The third natural language entry corresponding to the more complex second natural language entry, such as a compound word that can be replaced with a natural language entry, and as a result, is difficult to replace with the notation as it is, is obtained accurately, and the first natural language entry is replaced with the first natural language entry. 3. Create a natural language dictionary. Further, since the replacement rules can be used in combination, there is no need to hold all possible replacement rules in the replacement rule storage unit, and the first natural language-third natural language dictionary can be efficiently created. it can.

Further, the present invention described in claim 6 provides the present invention in claim 5.
In the invention described in the above, when the dictionary creation management unit uses a plurality of replacement rules extracted by the replacement rule search unit in the replacement execution unit in combination, the dictionary creation management unit counts the number of uses of the combined replacement rule. The gist of the present invention is to have a means and a replacement rule creating unit for storing the combined replacement rule in the replacement rule storage unit when the counted number of times of use of the replacement rule is equal to or more than a predetermined number.

According to the present invention, when a plurality of replacement rules are used in combination, the number of times of use of the combined replacement rule is counted, and when the number of uses becomes a predetermined number or more, In order to store the combined replacement rule in the replacement rule storage unit, a combination replacement rule having a high use frequency is acquired and stored in the replacement rule storage unit, so that the first natural language versus the second natural language dictionary entry is obtained. The speed of replacing the second natural language entry with the third natural language entry can be increased, and the first natural language-third natural language dictionary can be efficiently created.

The present invention according to claim 7 provides the invention according to claims 4 to 6
In the invention according to any one of the above, further comprising a second natural language dictionary storage unit for storing a second natural language dictionary storing the original form, part of speech, and utilization information of the second natural language, The gist is that the analysis unit has means for searching the second natural language dictionary storage unit and analyzing the second natural language entry.

According to the seventh aspect of the present invention, the second natural language stores the original form, part of speech, and utilization information.
Since the second natural language analysis unit has a natural language dictionary and searches the second natural language dictionary storage unit and analyzes the second natural language entry, for example, the second natural language entry includes a variant, In the notation of, even if it is not registered in the second natural language-third natural language dictionary, it is possible to acquire the original form of the variant included in the second natural language entry from the second natural language dictionary, and this original form is the second natural language entry. When registered in the language-to-third natural language dictionary, the second natural-language entry of the first natural language-to-second natural language dictionary entry can be replaced with the third natural language entry, and as a result, can be replaced with the literal notation It is also possible to acquire a third natural language entry for a second natural language entry such as a variant that is difficult to perform, and to create a first natural language-third natural language dictionary accurately.

The present invention described in claim 8 provides the present invention according to claim 4.
The invention according to any one of the first to seventh aspects, further comprising a third natural language dictionary storage unit that stores a third natural language dictionary that stores the original form, part of speech, and utilization information of the third natural language,
The replacement execution unit searches the third natural language dictionary storage unit based on the analysis result of the second natural language analysis unit or the replacement rule retrieved by the replacement rule search unit, and searches for the first natural language versus the second natural language. The gist of the present invention is to have means for replacing the second natural language entry of the language dictionary entry with the third natural language entry.

According to the eighth aspect of the present invention, the third natural language stores the original form, part of speech, and utilization information.
A third natural language dictionary storage unit that has a natural language dictionary and searches the third natural language dictionary storage unit based on the analysis result of the second natural language analysis unit or the replacement rule retrieved by the replacement rule search unit; In order to replace the second natural language entry of the dictionary entry with the third natural language entry, for example, the second natural language entry contains a variant, and is not registered in the second natural language-third natural language dictionary as it is. In this case, if the variant included in the second natural language entry is converted to the original form and the corresponding third natural language entry is obtained in the original form from the second natural language to third natural language dictionary, the third natural language dictionary To convert the original form of the third natural language entry into a notation of an appropriate variant, and as a result, corresponds to the second natural language entry such as a variant that is difficult to replace with the notation as it is 3 is converted into a natural language entry is also applicable exact representation can be created accurately first natural-language third natural language dictionary.

Further, the present invention described in claim 9 is based on claim 1.
In the invention according to any one of the first to eighth aspects, the dictionary creation management unit acquires a plurality of second natural language entries of the first natural language dictionary entry and the second natural language dictionary entry, and stores a third natural language entry for the second natural language entry. In a case where the natural language entries are acquired in duplicate, the gist of the present invention is to have a translation word selection information providing unit that determines that the third acquired natural language entry is more appropriate.

According to the ninth aspect of the present invention, a plurality of second natural language entries of the first natural language versus second natural language dictionary entry are obtained, and a third natural language entry corresponding to the second natural language entry is obtained. Is acquired twice,
In order to judge that the third acquired natural language entry is more appropriate, if there are a plurality of candidate translations at the stage of the second natural language entry, the third natural language entry is acquired from the plurality of second natural language entries. If the duplicated third natural language entries overlap, translation word selection information indicating that the duplicated third natural language entry is more appropriate than the third natural language entry that appeared alone can be added, and as a result, reliability A high first natural language to third natural language dictionary can be created.

[0027] According to a tenth aspect of the present invention, in the invention of any one of the first to ninth aspects, there is further provided a third natural language corpus information storage unit for storing statistical information of the third natural language corpus. And the dictionary creation management unit performs the third
The gist of the present invention is to include means for adding translation word selection information indicating whether the third natural language entry acquired by the dictionary creation management unit is appropriate, using information in the natural language corpus information storage unit.

In the present invention according to claim 10, the third aspect
A third natural language corpus information storage unit that stores the statistical information of the natural language corpus, and uses the information of the third natural language corpus information storage unit to acquire the third natural language acquired by the dictionary creation management unit From the first natural language entry, the second
Even if the number of translation candidates increases while tracing to the third natural language entry via the natural language entry, it is possible to know which translation is more appropriate, for example, a technical term or a field-specific word. In addition, a priority order of a word that is more frequently used among a plurality of translation word candidates can be given, and even when there is only one translation word candidate of the third natural language, the likelihood of the translation word candidate can be determined. Further, when the second natural language entry is a compound word and the third natural language entry cannot be obtained by the notation as it is, and the analysis is performed to obtain the third natural language entry, a word candidate in the third natural language entry is obtained. Even if there are a plurality of words, it is possible to select a more appropriate word combination from the co-occurrence information between words included in the acquired third natural language entry, and as a result, the first natural language versus the first 3. A natural language dictionary can be created.

According to an eleventh aspect of the present invention, there is provided a dictionary creation method for creating a bilingual dictionary of a natural language pair, comprising: a first natural language heading and its part of speech; and a bilingual translation of the first natural language entry. A first natural language versus a second natural language dictionary storing a first natural language versus a second natural language dictionary having a bilingual expression and a part of speech of the second natural language in a first natural language versus a second natural language dictionary storage unit A second natural language-third natural language dictionary having a part storing step, a heading of the second natural language and its part of speech, a bilingual expression of the third natural language which is a bilingual translation of the second natural language entry, and its part of speech. A second natural language pair to a third natural language dictionary storage unit stored in the second natural language pair to the third natural language dictionary storage unit; and a first natural language pair from the first natural language pair to the second natural language dictionary storage unit. Second
A dictionary entry reading step of extracting a natural language dictionary entry, and a third natural language entry corresponding to the second natural language entry of the extracted first natural language to second natural language dictionary entry.
A dictionary creation that acquires a natural language entry from the second natural language to third natural language dictionary storage unit and replaces the second natural language entry of the first natural language to second natural language dictionary entry with the third natural language entry A management step, based on the first natural language-third natural language dictionary entry generated by replacing the second natural language entry of the first natural language entry-second natural language dictionary entry with the third natural language entry; A first natural language pair having a first natural language headline and its part of speech, and a third natural language parallel translation and its part of speech that is a parallel translation of the first natural language entry;
A first natural language-third natural language dictionary storing step of creating and storing a natural language bilingual dictionary is essential.

In the present invention according to claim 11, the first
The first natural language / second natural language dictionary entry is retrieved from the natural language / second natural language dictionary storage unit, and the third natural language entry corresponding to the retrieved first natural language / second natural language dictionary entry is retrieved. A natural language entry is obtained from the second natural language to third natural language dictionary storage unit, and the third natural language entry replaces the second natural language entry of the first natural language to second natural language dictionary entry to obtain the first natural language entry. A first natural language-third natural language bilingual dictionary having a headline of a language and its part of speech, a third natural language bilingual translation of the first natural language entry and its part of speech, and creates a first natural language bilingual dictionary To store in the third natural language dictionary storage unit,
By using the first natural language-second natural language dictionary and the second natural language-third natural language dictionary, the first natural language-third natural language combination dictionary and the bilingual corpus are not required. A language-to-third natural language dictionary can be created.

According to a twelfth aspect of the present invention, in accordance with the eleventh aspect, the method further comprises the step of storing statistical information of the first natural language corpus in the first natural language corpus information storage unit, The creation management step refers to the first natural language corpus information storage unit by using the first natural language entry of the first natural language versus second natural language dictionary entry supplied from the dictionary entry reading step as a key, and The gist of the present invention is to include an entry selection information adding step of removing the first natural language pair and the second natural language entry.

According to the present invention, the first natural language corpus information is stored by using the first natural language entry of the first natural language versus second natural language dictionary entry supplied from the dictionary entry reading step as a key. To remove unnecessary first natural language vs. second natural language entries with reference to the section, unnecessary dictionary entries are not created when, for example, a dictionary limited to a field is created,
As a result, the speed and efficiency of creation and use of the first natural language versus third natural language dictionary can be improved. In the case of creating a dictionary limited to a field, if the first natural language-to-second natural language dictionary and the second natural language-to-third natural language dictionary to be used can use the dictionary of the field, the reliability is higher. A field-specific first natural language versus third natural language dictionary can be created.

Further, the present invention according to claim 13 is the invention according to claim 12, further comprising the step of storing statistical information of the second natural language corpus in a second natural language corpus information storage unit, The selecting information adding step refers to the second natural language corpus information storage unit using the second natural language entry of the first natural language versus second natural language dictionary entry supplied from the dictionary entry reading step as a key, and assigns a weight. The gist of the present invention is to provide a translation word selection information / weighting information provision step for providing translation word selection information as information.

According to the present invention, the second natural language corpus information is stored by using the second natural language entry of the first natural language versus the second natural language dictionary entry supplied from the dictionary entry reading step as a key. See section
To assign translation selection information, which is weighting information,
Even when there are multiple candidate translations from natural language to the second natural language, it is possible to find out which translation is more appropriate, for example, as a technical term or a field-specific word, and to appear more frequently among multiple translation candidates It is possible to prioritize which translation word candidates to perform. Furthermore, even when there is one translation word candidate, the likelihood of the translation word candidate can be determined, and as a result, a more reliable first natural language versus third natural language dictionary can be created. In addition, since information for selecting a translated word is added at the stage of the second natural language, the same information can be used when only the third natural language is changed without changing the first natural language and the second natural language. When the third natural language is extended to various languages as a result, the first natural language-third natural language dictionary can be efficiently created.

According to a fourteenth aspect of the present invention, in the invention of the eleventh aspect, the dictionary creation management step includes the first step.
If the second natural language entry of the natural language versus second natural language dictionary entry could not be replaced with the third natural language entry, a second analysis of the second natural language entry is performed.
A natural language analysis step; and a replacement execution step of acquiring a third natural language entry based on an analysis result of the second natural language analysis step and replacing the second natural language entry with the acquired third natural language entry. Is the gist.

In the present invention according to claim 14, the first aspect
If the second natural language entry of the natural language to second natural language dictionary entry cannot be replaced with the third natural language entry, the second natural language entry is analyzed, and the third natural language entry is analyzed based on the analysis result. In order to replace the second natural language entry with the obtained third natural language entry, for example, the second natural language entry is a compound word, and is registered in the second natural language-third natural language dictionary as it is. Even if it is not, if the second natural language entry is analyzed such as dividing it into sub-expression units constituting a compound word, and if each sub-expression is registered in the second natural language-third natural language dictionary, , The second natural language entry of the first natural language versus the second natural language dictionary entry can be replaced by the third natural language entry, and as a result, the notation can be replaced as it is. The third natural language entry for such difficult compound words can create the first natural language versus the third natural language dictionary won.

According to a fifteenth aspect of the present invention, there is provided a recording medium storing a dictionary creation program for creating a bilingual dictionary of a natural language pair, comprising: a first natural language heading and its part of speech; A first natural language versus a second natural language dictionary storing a first natural language versus a second natural language dictionary having a bilingual expression of the second natural language and its part of speech, A second natural language dictionary having a second natural language dictionary storage step, a heading of the second natural language and its part of speech, and a bilingual expression of the third natural language which is a bilingual translation of the second natural language entry and its part of speech. Storing the third natural language dictionary in the second natural language-third natural language dictionary storage unit; and storing the third natural language dictionary in the second natural language dictionary storage unit. 1st natural language A dictionary entry reading step of extracting a second natural language dictionary entry; and a third natural language entry corresponding to the second natural language entry of the first natural language pair and the second natural language dictionary entry thus extracted is stored in the second natural language pair. A dictionary creation management step of obtaining from the third natural language dictionary storage unit and replacing a second natural language entry of the first natural language pair second natural language dictionary entry with the third natural language entry; A first natural language heading and its part of speech based on the first natural language versus third natural language dictionary entry generated by replacing the second natural language entry of the second natural language dictionary entry with the third natural language entry; First natural language versus third natural language bilingual translation with third natural language bilingual translation of first natural language entry and its part of speech And gist dictionary creating program having a first natural-language third natural language dictionary storage step of storing create be recorded on the recording medium.

In the present invention according to claim 15, the first aspect
The first natural language / second natural language dictionary entry is retrieved from the natural language / second natural language dictionary storage unit, and the third natural language entry corresponding to the retrieved first natural language / second natural language dictionary entry is retrieved. A natural language entry is obtained from the second natural language to third natural language dictionary storage unit, and the third natural language entry replaces the second natural language entry of the first natural language to second natural language dictionary entry to obtain the first natural language entry. A first natural language-third natural language bilingual dictionary having a headline of a language and its part of speech, a third natural language bilingual translation of the first natural language entry and its part of speech, and creates a first natural language bilingual dictionary Since the dictionary creation program stored in the third natural language dictionary storage unit is recorded on the recording medium, the distribution can be improved by using the recording medium.

[0039]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram illustrating a configuration of a dictionary creation device according to an embodiment of the present invention. The dictionary creation device shown in FIG. 1 includes a first natural language-second natural language dictionary storage unit 10, a dictionary entry reading unit 11, a dictionary creation management unit 12, a first natural language-third natural language dictionary storage unit 13,
A first natural language corpus information storage unit 14, a second natural language corpus information storage unit 15, a replacement rule storage unit 16, a second natural language dictionary storage unit 17, a third natural language dictionary storage unit 18,
It comprises a second natural language-third natural language dictionary storage unit 19 and a third natural language corpus information storage unit 20.

First natural language to second natural language dictionary storage 1
0, one or more first natural language to second natural language dictionaries are stored. The stored dictionary records include at least a heading of the first natural language and its part of speech, and the first natural language. And a part-of-speech expression in the second natural language for the first natural language entry, which is a pair of the headline and the part of speech.

The dictionary entry reading unit 11 reads a first natural language entry from a dictionary stored in the first natural language-to-second natural language dictionary storage unit 10 and stores a second natural language entry including the bilingual expression thereof.
The natural language entry is taken out and supplied to the dictionary creation management unit 12.

The dictionary creation management unit 12 includes a first natural language corpus information storage unit 14, a second natural language corpus information storage unit 15, a replacement rule storage unit 16, a second natural language dictionary storage unit 17, and a third natural language dictionary. The first natural language versus the second natural language dictionary supplied from the dictionary entry reading unit 11 with reference to the storage unit 18, the second natural language to third natural language dictionary storage unit 19, and the third natural language corpus information storage unit 20 The second natural language entry of the entry is replaced with the third natural language entry, and stored in the first natural language-third natural language dictionary storage unit 13.

The dictionary creation management section 12 includes an entry selection information provision section 121, a second natural language analysis section 122, a substitution management section 123, and a translated word selection information provision section 124. Further, the replacement management unit 123 includes a replacement rule search unit 1231,
The replacement rule creation unit 1232 and the replacement execution unit 1233 are configured.

The entry selection information providing unit 121 of the dictionary creation management unit 12 uses the first natural language entry of the first natural language vs. second natural language dictionary entry supplied from the dictionary entry reading unit 11 as a key to generate a first natural language entry. Referring to the language corpus information storage unit 14, it is determined whether or not the first natural language-to-second natural language dictionary entry is necessary. If it is not necessary, the subsequent operation is not performed and the dictionary entry reading unit 1
The operation is shifted to the first natural language to second natural language dictionary entry supplied from 1.

The entry selection information providing section 121
With reference to the second natural language corpus information storage unit 15 using the second natural language entry of the first natural language versus second natural language dictionary entry supplied from the dictionary entry reading unit 11 as a key, the second natural language entry is The statistical information in the corpus is taken out and added to the second natural language entry, and the second natural language entry with the statistical information added is supplied to the replacement execution unit 1233 in the replacement management unit 123 of the dictionary creation management unit 12.

The replacement execution unit 1233 determines whether the second natural language
With reference to the natural language dictionary storage unit 19, the second natural language entry of the first natural language versus second natural language dictionary entry supplied from the dictionary entry reading unit 11 is replaced with a third natural language entry, and the first natural language entry And the third natural language entry pair to the translation word selection information providing unit 124.

However, if the third natural language entry for the second natural language entry of the first natural language versus second natural language dictionary entry cannot be obtained by the replacement execution unit 1233, the second natural language entry is managed by dictionary creation management. It is supplied to the second natural language analysis unit 122 of the unit 12.

The second natural language analysis unit 122 of the dictionary creation management unit 12 provides one analysis means for the second natural language entry.
The first natural language to second natural language dictionary entry in the replacement execution unit 1233 of the replacement management unit 123.
When the natural language entry cannot be replaced with the third natural language entry, the second natural language entry is analyzed with reference to the second natural language dictionary storage unit 17 as necessary, and the analysis result is replaced and managed. This is supplied to the replacement rule search unit 1231 of the unit 123.

Replacement management unit 123 of dictionary creation management unit 12
Obtains a third natural language entry for the second natural language dictionary entry of the first natural language to second natural language dictionary entry.

The replacement rule search unit 12 of the replacement management unit 123
31 is the first natural language analysis unit 122 supplied from the first
Using the analysis result of the second natural language entry of the natural language versus the second natural language dictionary entry as a key, a replacement rule in the replacement rule storage unit 16 is searched, and the second natural language entry of the first natural language versus the second natural language dictionary entry is searched. One or more replacement rules for replacing the second natural language entry with the third natural language entry are extracted using the analysis result of.

The replacement rule creation unit 12 of the replacement management unit 123
Reference numeral 32 denotes a combination of a plurality of existing replacement rules stored in the replacement rule storage unit 16 by the replacement rule search unit 1231, and a second natural language dictionary entry of the second natural language dictionary entry.
When the analysis result of the natural language entry is replaced and the number of appearances of the combination of the replacement rules (combination replacement rule) becomes a certain number or more, the combination replacement rule is stored in the replacement rule storage unit 16.

The replacement execution unit 1233 of the replacement management unit 123
Searches the second natural language-third natural language dictionary storage unit 19 using the second natural language entry of the first natural language-second natural language dictionary entry as a key, and searches the first natural language-second natural language dictionary. Replace the second natural language entry of the entry with a third natural language entry.

The replacement execution unit 1233 analyzes the second natural language entry of the first natural language dictionary entry against the second natural language dictionary by the second natural language analysis unit 122, or analyzes the second natural language entry. According to the result and one or more replacement rules retrieved by the replacement rule search unit 1231, the second natural language versus third natural language dictionary storage unit 19
Then, referring to the third natural language dictionary storage unit 18 as necessary, the second natural language entry of the first natural language versus second natural language dictionary entry is replaced with the third natural language entry.

If the third natural language entry corresponding to the second natural language entry of the first natural language dictionary entry to the second natural language dictionary entry cannot be obtained, the replacement execution unit 1233 To the second natural language analysis unit 1
22 and the process of replacing the second natural language entry of the first natural language versus second natural language dictionary entry with the third natural language entry is repeated in the same manner as described above.

However, the reason why the second natural language entry of the first natural language to second natural language dictionary entry is returned to the second natural language analysis unit 122 is that the analysis by the second natural language analysis unit 122 has not yet been performed. This is only when the analysis method has not been performed or when a plurality of analysis methods are stored in the second natural language analysis unit 122 and an analysis method that has not yet been performed on the second natural language entry remains. If not, a flag indicating that replacement has not been performed is given, and the second natural language entry of the first natural language versus second natural language dictionary entry is not replaced with the third natural language entry, or the second natural language entry is not replaced. The intermediate result of the replacement of the natural language entry with the third natural language entry is stored in the first natural language-third natural language dictionary storage unit 13.

The translation word selection information providing unit 124 refers to the third natural language corpus information storage unit 20 using the third natural language entry supplied from the replacement management unit 123 as a key, and Translation word selection information such as which translation word is given priority or selection is given.

First natural language to third natural language dictionary storage 1
3 includes at least the first natural language heading and its part of speech supplied from the dictionary creation management unit 12 and the first natural language entry corresponding to the first natural language heading and its part of speech, which is the first natural language entry. The bilingual expression of the third natural language and its part of speech and, if necessary, the translation word selection information of the third natural language can also be stored.

The first natural language corpus information storage unit 14
The statistical information extracted using the first natural language corpus, which is an electronic document, is stored.

The second natural language corpus information storage unit 15
Statistical information extracted using the second natural language corpus, which is an electronic document, is stored.

The replacement rule storage unit 16 stores the replacement management unit 12
3, a replacement rule used when the second natural language entry of the first natural language versus second natural language dictionary entry is replaced with the third natural language entry is stored.

The replacement rule storage unit 16 also stores the combination replacement rules that have appeared a fixed number of times or more among the combination replacement rules used by combining the existing replacement rules in the replacement rule search unit 1231.

The second natural language dictionary storage section 17 stores a monolingual dictionary of the second natural language, and stores at least the original form, the part of speech, and the utilization information as dictionary entries.

The third natural language dictionary storage section 18 stores a monolingual dictionary of the third natural language, and stores at least the original form, the part of speech, and the utilization information as dictionary entries.

Second natural language-third natural language dictionary storage unit 1
Reference numeral 9 denotes at least one second natural language-third natural language dictionary stored therein, and the stored second natural language-third natural language dictionary entry includes at least a heading of the second natural language and its part of speech. And a third corresponding to the second natural language entry
Contains natural language bilingual expressions and their parts of speech.

The third natural language corpus information storage unit 20
It stores statistical information extracted using a third natural language corpus, which is an electronic document.

Next, a specific example and operation of the dictionary creating apparatus configured as described above will be described with reference to the flowcharts shown in FIGS. In this description, the first natural language is Japanese, the second natural language is English, and the third natural language is French, but the combination of the first natural language, the second natural language, and the third natural language is the first natural language. As long as the first, second, and third natural languages are different languages, any language such as Thai, Chinese, and Spanish, or German, French, and Russian may be used.

FIG. 2 is a diagram showing the relationship between the first natural language and the second natural language shown in FIG.
3 shows an example of a first natural language versus second natural language dictionary entry stored in a natural language dictionary storage unit 10. The first natural language vs. second natural language dictionary is such that the first natural language is Japanese and the second is natural.
If the natural language is English, for example, the ALT-J / E Japanese-English Contrast Dictionary (NTT, “Japanese-English Machine Translation Technology”, NTTR &
Dvol.46, pp107-141,1997) and EDR Electronic Dictionary (Nihondenshikajishokenkyusho, http: //www.iijnet.or.jp/edr/J _- in
dex.html).

As shown in FIG. 2, at least a Japanese headline and its part of speech, and an English bilingual expression corresponding to the Japanese entry and its part of speech are registered as a first natural language versus second natural language dictionary entry. For example, in the first natural language versus second natural language dictionary entry whose Japanese heading is "bank", its part of speech "noun", its bilingual expression "bank", and its part of speech "noun" are registered. The first natural language versus second natural language dictionary entry whose word heading is "banker" has its part of speech "noun" and its bilingual expression "bank cle
rk "and its part of speech" noun "are registered, and the first natural language versus second natural language dictionary entry whose Japanese heading is" yagade "has its part of speech" noun "and bilingual expression"
burn "and its part-of-speech" noun ", and another bilingual expression" scald "and its part-of-speech" noun "are registered, and further, the first natural language vs. the second natural language whose Japanese heading is" four seasons " The part-of-speech “noun”, the bilingual expression “four seasons” and the part-of-speech “noun” are registered in the natural language dictionary entry, and the first natural language versus the second natural language whose Japanese heading is “Chiaki”. In the language dictionary entry, the part of speech “noun” and the translation “last performance of the seas”
on ”and its part of speech“ noun ”are registered.

In these examples, the Japanese heading is a noun, but the Japanese heading is not necessarily a noun, but may be a verb or an adjective. In these examples, the part of speech in English is also a noun, but this part of speech need not be a noun, but may be a verb or an adjective, and need not match the part of speech of the heading.

The first natural language / second natural language dictionary stored in the first natural language / second natural language dictionary storage unit 10 does not need to be one, and a plurality of dictionaries may be stored. For example, the ALT-J / E Japanese-English contrast dictionary or EDR
If the purpose is to store both computerized dictionaries and to create a dictionary limited to a field, for example, when it is desired to create a computer-related dictionary, it is possible to use a computer term dictionary.

Next, the function of the dictionary entry reading section 11 will be described. The dictionary entry reading unit 11 has a function of extracting a dictionary entry from one or a plurality of dictionaries stored in the first natural language versus second natural language dictionary storage unit 10 (Step S2 in FIG. 14). The first natural language versus second natural language dictionary entry read by the dictionary entry reading unit 11 is supplied to the dictionary creation management unit 12.

First natural language to second natural language dictionary storage unit 1
In the case where there is one dictionary stored in 0, for example, in the case of Japanese, the headings are sorted and taken out in a general word order in the dictionary of the first natural language, such as the order of the Japanese syllabary. The two natural language dictionary entries may be separated for each part of speech of the first natural language, and the headings may be sorted and extracted in the same part of speech.

First natural language to second natural language dictionary storage unit 1
When a plurality of dictionaries are stored in 0, for example, a fixed search order such as the order of the Japanese syllabary is determined, dictionary entries are extracted from each dictionary along the search order, and the first natural language entry is the same in a plurality of dictionaries. If there is a first natural language to second natural language dictionary entry, a method of extracting the first natural language to second natural language dictionary entry from all dictionaries can be considered.

In addition, a method is also conceivable in which a plurality of dictionaries are re-combined into one dictionary in advance using the first natural language entry as a key, and the dictionaries are taken out sequentially in the same manner as in the case of one dictionary.

In addition, the stored dictionary is 1
There is a way to pull one by one. In that case, the first dictionary is first sorted by the heading of the first natural language, and is searched in order, and it is checked whether the first natural language entry that has appeared is in another dictionary. All the first natural language-to-second natural language dictionary entries having entries are taken out, and the taken-out first natural language-to-second natural language dictionary entry is provided with a taken-out flag. Next, after searching the first dictionary to the end, the process proceeds to the next dictionary, and similarly extracts only the first natural language-second natural language dictionary entry to which the extracted flag is not added, and removes the first dictionary. If the first natural language entry and the second natural language dictionary entry having the first natural language entry are registered in other dictionaries, they are also taken out, and after being taken out, each dictionary is taken out while giving the taken-out flag. It is also conceivable to repeat until there is no remaining first natural language / second natural language dictionary entry.

In particular, when the dictionary of the first natural language to the third natural language to be created is a dictionary limited to the field, or when there is a difference in reliability between the dictionaries, the extracted first natural language to the third natural language is used. The two natural language dictionary entries are flagged as the first natural language versus the second natural language dictionary entry, and the first natural language versus the second natural language dictionary is determined based on the field and reliability of the first natural language versus the second natural language dictionary. It is also conceivable to weight the language versus second natural language dictionary entry.

Next, the dictionary creation management unit 12 will be described. The first natural language to second natural language dictionary entry supplied from the dictionary entry reading unit 11 is used by the dictionary creation management unit 12 to change the second natural language entry of the first natural language to second natural language dictionary entry to the third natural language. The first natural language entry and the third natural language dictionary are stored as a set of the first natural language entry and the third natural language entry.

In the dictionary creation management section 12, first, the entry selection information providing section 121 refers to the first natural language corpus information storage section 14 and reads the first natural language versus the second natural language supplied from the dictionary entry reading section 11. A determination is made as to whether the dictionary entry is useful for acquiring as the first natural language-third natural language dictionary entry, such as whether the dictionary entry is a word that frequently appears in the field of the first natural language-third natural language dictionary to be created. , And supplies it to the replacement execution unit 1233 if it is determined to be useful.

The statistical information of the first natural language corpus stored in the first natural language corpus information storage unit 14 referred to here has, for example, a format as shown in FIG. A set of words and parts of speech that appear in a certain Japanese corpus and statistical information such as the frequency of occurrence of the set in the corpus are shown.

The corpus which is the basis of the statistical information of the first natural language corpus stored in the first natural language corpus information storage unit 14 is, for example, a computerized Nihon Keizai Shimbun or Mainichi Shimbun in Japanese. Is raised.

The function of the entry selection information providing unit 121 will be specifically described with reference to FIGS. For example, assuming that it is desired to create an economic-related dictionary, the first natural language corpus information storage unit 14 stores statistical information obtained from the first economic-related natural language corpus. FIG. 3 is an example of statistical information obtained from a Japanese corpus of economic relations. Assuming that the first natural language vs. second natural language dictionary entry of “bank (noun)” or “bank employee (noun)” of the Japanese entry shown in FIG. As described above, since "bank (noun)" and "bank employee (noun)" appear in the Japanese corpus of economic relations, it is determined that the entry is useful if acquired, and is supplied to the replacement execution unit 1233. (Step S in FIG. 14)
3 (Yes)).

However, if the supplied Japanese entry of the first natural language vs. second natural language dictionary entry is “yaku (noun)”, “four seasons (noun)”, or “senshugaku (noun)”, FIG. No Japanese entry appears in the Japanese-language corpus of economic relations, so when creating a dictionary of economic relations, it is not considered useful to create dictionary entries corresponding to these Japanese entries. Then, the operation is moved to the next dictionary entry (step S3 (N
o)). However, if the dictionary to be created is not limited to an economic dictionary, "burns (noun)", "four seasons (noun)", and "Chiakiraku (noun)" are also supplied to the replacement execution unit.

The entry selection information providing section 121
With reference to the second natural language corpus information storage unit 15, the second natural language entry supplied from the dictionary entry reading unit 11 is weighted for selecting a translated word, that is, the translated word selection information is added and weighted. It also has a function of supplying the second natural language entry to the replacement execution unit 1233.

The statistical information of the second natural language corpus stored in the second natural language corpus information storage unit 15 referred to here has, for example, a format as shown in FIG. A set of words and parts of speech appearing in a certain English corpus and statistical information such as the frequency of occurrence of the set in the corpus are shown.

If the corpus serving as the source of the statistical information of the second natural language corpus stored in the second natural language corpus information storage unit 15 is English, for example, Financial Corpus
mes (http://www.jcp.grenet.fr/ELRA/cata/text _- det.ht
ml # mlcc) and Wall Street Journal (http: //morph.ldc.up
enn.edu/Catalog/LDC2000T43.html).

With reference to FIGS. 2 and 4, the weighting function of the entry selection information giving unit 121 will be specifically described. For example, the Japanese entry shown in FIG.
In the English entry of the first natural language versus second natural language dictionary entry, "burn (noun)" and "scale (noun)" are registered. However, it can be seen from FIG. 4 that the appearance frequency in the second natural language corpus is higher for “burn (noun)”. So, as an English translation of "yaga (noun)"
"burn (noun)" is more important than "scale (noun)"
Consequently, in order to emphasize the French words obtained using "burn (noun)" over the French words obtained using "scale (noun)", weighting for selecting a translation word is performed as shown in FIG. 5, for example ( Step S4 in FIG. 14).

If statistical information of the second natural language corpus obtained from a sufficiently large second natural language corpus or a reliable second natural language corpus can be used, the first natural language corpus can be used as in this example. It is also conceivable to weight the second natural language entry of the language-to-second natural language dictionary entry and supply the weighted second natural language entry to the replacement execution unit 1233, or to supply only the heavy weight of the second natural language entry to the replacement execution unit 1233. Can be

When another first natural language-to-second natural language dictionary entry shown in FIG. 2 is supplied to the entry selection information providing unit 121 having the above-mentioned weighting function, the English language corresponding to each Japanese entry is Since the entries are one by one, the weighting function of the entry selection information assigning unit 121 does not work, or information on the likelihood of the second natural language entry in the field is added when a dictionary limited to the field is created. It is also possible to do.

Next, the replacement execution unit 12 of the replacement management unit 123
33 will be described. In the replacement execution unit 1233, the second natural language dictionary entry of the supplied first natural language
With reference to the second natural language-third natural language dictionary storage unit 19 using the natural language entry as a key, a third natural language entry which is a translation of the second natural language entry is acquired, and the first natural language-second natural language entry is obtained. The second natural language entry of the language dictionary entry is replaced with the third natural language entry and supplied to the translated word selection information providing unit 124.

However, if the replacement execution unit 1233 cannot acquire the third natural language entry which is a translation of the second natural language entry of the supplied first natural language / second natural language dictionary entry, the second natural language entry is not obtained. The natural language entry is supplied to the second natural language analysis unit 122.

First, the second natural language-third natural language dictionary storage unit 19 will be specifically described with reference to FIG. FIGS. 6A and 6B show the second natural language versus the third natural language.
5 is an example of a second natural language to third natural language dictionary entry stored in the natural language dictionary storage unit 19;

Assuming that the second natural language is English and the third natural language is French, the second natural language to third natural language dictionary is, for example,
Lantmark Lexicon (LanTmark, http://www.icp.grenet.fr
/ ELRA / cata / text _- det.html # multilex) and Basic Multili
ngual Dictionary (http://www.icp.grenet.fr/ELRA/cat
a / text _- det.html # multilex).

As shown in FIGS. 6A and 6B, the second natural language-third natural language dictionary entry includes at least an English headline and its part of speech, and a bilingual expression of French and its part of speech corresponding to the English entry. Is registered.

For example, in FIG. 6A, the English heading is "b".
The second natural language versus third natural language dictionary entry "ank" has its part of speech "noun", its bilingual expression "banque" and its part of speech "female noun", and another bilingual expression "riv
e ”and its part of speech“ female noun ”are registered. Also, the second natural language-third natural language dictionary entry whose English heading is "burn" has its part of speech "noun" and its bilingual expression "
ruisseau ”and its part-of-speech“ masculine noun ”and another translation“ bru 対 lure ”and its part-of-speech“ feminine noun ”are registered, and the second natural language pair whose English heading is“ scald ” In the three natural language dictionary entries, the part of speech “noun”, the bilingual expression “bru お よ び lure”, and the part of speech “female noun” are registered.

In these examples, the English heading is a noun, but it is not necessarily a noun, but may be a verb or an adjective. Also, in these examples, the parts of speech in French were also masculine or feminine nouns, but these parts of speech need not be masculine or feminine nouns, but may be verbs or adjectives, and need to match the part of speech in the heading. Absent.

Further, the second natural language-third natural language dictionary stored in the second natural language-third natural language dictionary storage unit 19 need not be one, and a plurality of dictionaries may be stored. For example, the Lantmark Lexicon or Basic Multilingual
If the purpose is to store both dictionaries or to create a dictionary limited to a specific field, for example, when a computer-related dictionary is to be created, a dictionary of computer terms may be used.

Referring to FIG. 6A, replacement execution unit 1233
The function will be described specifically. Here, when "bank (noun)" which is an English entry for the Japanese entry "bank (noun)" shown in FIG. 2 is supplied to the replacement execution unit 1233, the second natural language shown in FIG. The language-to-third natural language dictionary storage unit 19 is searched, and the French entries “banque (female noun)” and “rive” which are translations of “bank (noun)”
(Female noun) "(step S5 (Y in FIG. 14)
es)), a pair of the Japanese entry “bank (noun)” and the French entries “banque (female noun)” and “rive (female noun)” is supplied to the translation word selection information provision unit 124 (step S6 in FIG. 14). ).

When the English entries “burn (noun)” and “scald (noun)” for the Japanese entry “yake (noun)” shown in FIG. The natural language-versus-third natural language dictionary storage unit 19 is searched, and from FIG. 6A, the French entry “ruisseau (male noun)” for “burn (noun)” is entered.

[Outside 1] And the French entry “bru ＾ lure” for “scald (noun)”
(Female noun) "(step S5 (Y in FIG. 14)
es)), Japanese entries "yaga (noun)" and French entries "ruisseau (male noun)" and "bru ＾ lure
(Female noun) ”and the information indicating that“ bru ＾ lure (female noun) ”appears redundantly in the set and supply them to the translation word selection information providing unit 124 (step in FIG. 14). S6).

Also, "last perform" which is an English entry for the Japanese entry "Chiakiraku (noun)" shown in FIG.
ance of the season (noun) ”is supplied to the replacement execution unit 1233, and even if the second natural language-third natural language dictionary storage unit 19 is searched, as shown in FIG. There is no second natural language-third natural language dictionary entry for, and no corresponding French entry can be obtained, so this English entry "last performance of the se
ason (noun) "is supplied to the second natural language analysis unit 122 (step S5 (No) in FIG. 14).

Similarly, the English entry “bank clerk (noun)” for “bank employee (noun)” and the English entry “four seasons (noun)” for “four seasons (noun)”
, The English entry is supplied to the second natural language analysis unit 122 (step S5 (No) in FIG. 14).

Next, the translated word selection information providing unit 124 will be described. The translated word selection information providing unit 124 verifies the third natural language corpus information storage unit 20 by using the supplied third natural language entry as a key, and determines whether the supplied third natural language entry is correct, or If so, weighting is performed for selecting a translated word such as which third natural language entry has priority, and stored in the first natural language versus third natural language dictionary storage unit 13.

The statistical information of the third natural language corpus stored in the third natural language corpus information storage unit 20 referred to here has, for example, a format as shown in FIG. A set of words and parts of speech appearing in a corpus of a natural language, French, and statistical information such as an appearance frequency and a co-occurrence frequency in the corpus of the set are shown.

If the corpus serving as the source of the statistical information of the third natural language corpus stored in the third natural language corpus information storage unit 20 is French, for example, Le Monde new
spaper (http://morph.ldc.upenn.edu/Catalog/LDC95T1
1.html) and Canadian Hansard (http: //www.ldc.upenn.
edu / ldc / news / release / hansard.html).

The function of the translation word selection information providing unit 124 will be specifically described with reference to FIGS. 3 and 7A. For example, assuming that it is desired to create an economic dictionary, the third natural language corpus information storage unit 20 stores statistical information obtained from the economic corpus. FIG. 7A shows an example of the appearance frequency acquired from a French corpus related to economics. As described in the description of the entry selection information providing unit 121, in the case where an economic dictionary is created, the statistical information of the first natural language corpus of economic relation is stored in the first natural language corpus information storage unit 14. And using the statistical information,
When the first natural language-to-second natural language dictionary entry is selected, the first natural language-to-second natural language dictionary entries of the Japanese entries "bank (noun)" and "bank employee (noun)" shown in FIG. Although only the English entry for the Japanese entry "bank employee (noun)" is supplied to the second natural language analysis unit 122, it is supplied to the translation word selection information adding unit 124 at this stage. There is only a Japanese entry "bank (noun)" and a French entry obtained from "bank (noun)" which is the only English entry for the Japanese entry.

Therefore, the Japanese entry "bank (noun)"
And French entries "banque (female noun)" and "rive"
Assuming that the set of “(female noun)” is supplied to the translation word selection information providing unit 124, “banque (female noun)” frequently appears in the economy-related French corpus from FIG.
rive (female noun) ”does not appear, and the French entry for“ bank (noun) ”is“ banque
(Female noun) "is more promising.
The English word “bank (noun)” has a meaning such as “shore” in addition to the meaning of “bank”, and “rive (female noun)” is a French entry that appeared in the latter meaning. In this way, the French entries “banque (female noun)” and “rive (female noun)” are weighted for selecting a translation, and the first natural language paired with the Japanese entry “bank (noun)” are selected. 3 is stored in the natural language dictionary storage unit 13 (step S7 in FIG. 14).

If statistical information of the third natural language corpus obtained from a sufficiently large third natural language corpus or a reliable third natural language corpus can be used, the weight of the third natural language entry is heavy. It is also conceivable to store only those in the first natural language versus third natural language dictionary storage unit 13.

However, in the above example, the first natural language corpus information storage unit 14 stores the statistical information of the first natural language corpus, and the first natural language versus the second natural language is used by using the statistical information. Although it is assumed that dictionary entries have already been selected, the first natural language corpus information storage unit 1
Similarly, if there is no first natural language corpus information storage unit 14 or if there is no selection of the first natural language versus second natural language dictionary entry, the third natural language corpus information storage unit The French entries can be selected or weighted using statistical information such as the frequency of appearance of the third natural language corpus stored in 20.

[0108] The translated word selection information providing section 124 is provided with the first
Even when there are a plurality of second natural language entries in the natural language versus second natural language dictionary entry, if the supplied third natural language entry appears repeatedly, the third occurrence of the third natural language entry appears.
It has a duplicate judgment function that judges that natural language is more appropriate.

The duplication determination function of the translation word selection information providing section 124 will be specifically described. As described above, the English entry for the Japanese entry "Yagade (noun)" shown in FIG.
From the scald (noun), the French entry “bru ＾ lure (female noun)” shown in FIG.
(Noun) ”from the French entry“ ruis shown in FIG.
seau (masculine noun) ”and“ bru 女性 lure (feminine noun) ”
Is obtained and supplied to the translated word selection information providing unit 124. In the supplied French entry, only "bru ＾ lure (female noun)" appears repeatedly. So, as a French entry for "burns (nouns)", "ruisseau
We weight that “bru ＾ lure (female noun)” is more promising than “masculine noun”. Actually, “burn (noun)” in English means “brook” in addition to “burn”. "Ruisseau (masculine noun)" is French for Ogawa. In this way, the French entries “bruurelure (female noun)” and “ruisseau (male noun)” are weighted for selecting the translation, and the first natural language is added together with the Japanese entry “yagage (noun)”. It is stored in the third natural language dictionary storage unit 13 (step S7 in FIG. 14). However, it is also conceivable that the third natural language entry is stored in the first natural language-third natural language dictionary storage unit 13 while leaving only those having a heavy weight.

Next, the second natural language analyzing unit 122 will be described. The second natural language analysis unit 122 is supplied when the replacement management unit 123 cannot replace the second natural language entry of the first natural language versus second natural language dictionary entry with the third natural language entry. Second
The natural language entry is analyzed, and the analysis result of the second natural language entry is supplied to the replacement management unit 123 again.

Examples of the analysis method of the second natural language analysis unit 122 include morphological analysis and syntax analysis. Brill Tagg is a morphological analyzer for English
er (Eric Brill, ftp: blaze.cs.jhu.edu: / pub / brill /)
Or Xelda (Xerox, ftp: //ftp.parc.xerox.com/pub/tagge
r), etc., and parsing for English is ApplePi
e Parser (New York University, http://cs.nyu.edu/cs/
projects / proteus / app /), Link Grammar (Carnegie M
ellon University, http://www.link.cs.cmu.edu/link
/).

First, the case where the second natural language analysis unit 122 performs morphological analysis and syntax analysis will be described with reference to FIGS.
The function of the second natural language analysis unit 122 will be specifically described with reference to FIG. FIGS. 8A to 8C show the second natural language analysis unit 12.
2 is an example of a result of analysis of a second natural language entry by No. 2.

The English entry “last performance of th” for the Japanese entry “Chiakiraku (noun)” shown in FIG.
e season (noun) ”is supplied to the second natural language analysis unit 122. FIG. 8A shows“ last performance of
8B shows the result of the morphological analysis by the second natural language analyzing unit 122 of the season ". FIG. 8B shows the result of the syntax analysis of the English entry by the second natural language analyzing unit 122. According to the morphological analysis results shown in (a), “last performance of the season”
(Adjective), "performance (noun)", "of (preposition)", "the (definite article)", "season (noun)", and from FIG. 8 (b), furthermore, "the (definite article)" ) "+" Season (noun) "becomes a noun phrase and" of
(Preposition) ”is added to form a prepositional phrase, and“ last
(Adjective) “+” performance (noun) ”is a noun phrase, and“ last (adjective) performance (noun) of (preposition) the (definite article) season (noun) ”is a noun phrase. The result is supplied to the replacement management unit 123 (Step S10 in FIG. 14).

As another analysis method of the second natural language analysis unit 122, for example, it is conceivable that only the morphological analysis is performed and the morphological analysis is passed to the replacement management unit 123. In this case, for example, the English entry “bank clerk (noun)” for the Japanese entry “bank clerk (noun)” shown in FIG.
Are “bank (noun)” and “cler” as shown in FIG.
The morphological analysis is performed on “k (noun)”, and the result is supplied to the replacement management unit 123 (step S10 in FIG. 14).

As another analysis method of the second natural language analysis unit 122, for example, in the case of a compound word, it is conceivable that the compound word is divided into words constituting the compound word and transferred to the replacement management unit 123. In that case, the English entry "bank clerk" for the Japanese entry "bank employee (noun)" shown in FIG.
(Noun) is divided into “bank” and “clerk” and replaced
It is supplied to the management unit 123 (step S1 in FIG. 14).
0).

As another analysis method of the second natural language analysis unit 122, for example, in the case of a compound word, it is considered that the compound word is divided into subsets and passed to the replacement management unit 123. In that case, the Japanese entry "Chiakiraku (noun)" shown in Fig. 2
English entry for “lastperformance of the seaso
n (noun) ”means“ last performance of the ”,“ la
st performance of ”,” last performance ”,” las
t ”,” performance of the season ”,” parformance
of the "," performance of "," performance "," o
f the season "," of the "," of "," the seaso
n "," the "," season "
23 (step S10 in FIG. 14).

Further, in FIG. 1 which is an embodiment of the present invention, the second natural language analyzing section 122 refers to the second natural language dictionary storing section 17 for storing the dictionary of the second natural language. It has a second natural language dictionary utilization analysis function that can analyze natural language entries.

The second natural language dictionary stored in the second natural language dictionary storage unit 17 stores at least the original form, part of speech, and utilization information of the second natural language entry. FIG.
Is an example of an English dictionary entry stored in the second natural language dictionary storage unit 17, which shows information on the English original form, the part of speech, and its variations.

Referring to FIG. 9, second natural language analyzing unit 122
The above-described second natural language dictionary use analysis function will be specifically described. When the English entry “four seasons (noun)” for the Japanese entry “four seasons (noun)” shown in FIG. 2 is supplied, the second natural language analysis unit 122 divides the supplied English entry into word units, From the second natural language dictionary storage unit 17 shown in FIG. 9, information that “four” is an adjective and is invariable, and “seasons” is a plural form of “season (noun)” is obtained. (Figure 1
4 Step S10).

Next, the replacement management unit 123 will be described. In FIG. 1, as already described in the description of the replacement execution unit 1233,
When the second natural language entry of the second natural language entry is supplied to the replacement management unit 123 without passing through the second natural language analysis unit 122, the second natural language entry is supplied to the replacement execution unit 1233 and replaced. The execution unit 1233 stores the second natural language versus the third natural language dictionary stored in the third natural language dictionary storage unit 19.
Searching a natural language to third natural language dictionary to obtain a third natural language entry corresponding to said second natural language entry;
The second natural language entry of the first natural language pair to the second natural language dictionary entry is replaced with a third natural language entry, and the set of the first natural language entry and the third natural language entry is supplied to the translation word selection information providing unit 124 I do.

When the analysis result of the second natural language entry is supplied to the replacement management unit 123, the analysis result is supplied to the replacement rule search unit 1231. The replacement rule search unit 1231 searches the replacement rule storage unit 16 using the supplied analysis result of the second natural language entry as a key, and obtains a replacement rule for obtaining a third natural language entry from the analysis result of the second natural language entry. Try to get one or more. When the replacement rule is obtained, the analysis result of the second natural language entry and the replacement rule are supplied to the replacement execution unit 1233, and the replacement execution unit 1233 checks the second natural language of the first natural language versus the second natural language dictionary entry. The entry is third
The first natural language entry and the third natural language entry are replaced with a natural language entry, and the translation word selection information providing unit 1
24.

Further, in the replacement rule creating section 1232, a plurality of replacement rules are obtained by the replacement rule searching section 1231,
When these replacement rules are used in combination, the number of appearances of the combination of the replacement rules is counted, and when the combination of the replacement rules (combination replacement rule) appears more than a certain number of times, it is stored in the replacement rule storage unit 16. .

The replacement rules stored in the replacement rule storage section 16 are manually created in advance, or machine learning using a decision tree or the like (for example, “Programs for Machine”).
Learning ”, Morgan Kaufman, 1992), or manually modified and stored by machine learning. As described above, the same combination replacement rule is replaced by the replacement rule search unit 1231. When the combination replacement rule appears more than a certain number of times, the combination replacement rule is also stored in the replacement rule storage unit 16. The combination replacement rule may be manually checked.

The function of the replacement management unit 123 will be specifically described with reference to FIGS. 8B, 10 and 6B. It is assumed that the English entry “last performance of the season (noun)” for “Chiaki (noun)” is supplied to the replacement management unit 123. FIG. 8B shows an analysis result supplied from the second natural language analysis unit 122, and FIG.
0 is an example of a replacement rule stored in the replacement rule storage unit 16, and FIG. 6B shows a dictionary entry of the English-French dictionary stored in the second natural language versus third natural language dictionary storage unit 19. It is an example.

In the replacement rule search unit 1231, for example,
A method of preferentially using a replacement rule that can be searched in the longest unit can be considered. In this case, “last” shown in FIG.
performance of the season ”
First, a replacement rule of “last (adjective) performance (noun) of (preposition) the (definite article) season (noun)” is searched. However, since the replacement rule is not stored in the replacement rule storage unit 16 as shown in FIG. The next longest unit is "of
(Preposition) the (definite article) season (noun) "is searched. In response to this, condition 1 of the replacement rules of rule numbers 5 and 6 is matched.
“Season (noun)” is searched from the second natural language-third natural language dictionary storage unit 19 shown in FIG. 6B, and the French entry “saison” corresponding to the English entry “season (noun)” is retrieved.
(Female noun) ". Since the part of speech of the French entry is" female noun ", the replacement rule 6 matches. So "of (preposition) the (definite article) season (noun)" is "d
e la saison (French prepositional phrase) ".

Next, similarly, the remaining "last (adjective) perf
ormance (noun) ". In response to this, condition 1 of the replacement rule of rule numbers 1 and 2 is met, and the second natural language versus third natural language dictionary shown in FIG. As the French entry corresponding to “performance (noun)” from the storage unit 19, “repre'sentation (female noun)” and “spectacle (male noun)” are acquired.
So “last (adjective) performance (noun)” is “d
ernie're repre'sentation (French noun phrase) "or"dernie'r spectacle (French noun phrase) ".

"De la saison (French preposition phrase)" obtained by the above substitution and "dernie're repre'sentation"
(French noun phrase) "or"dernie'r spectacle (French noun phrase) "is searched for a replacement rule, and"dernie'rerepre'sentat"
ion de la saison (noun phrase) ”or“ dernie'r
spectacle de la saison (noun phrase) "can be obtained (step S12 (Yes) in FIG. 14).

As described above, as a French entry for the Japanese entry "Chiakiraku (noun)", "dernie're repr
e'sentatione de la saison (noun phrase) "or" d
ernie'r spectacle de la saison (noun phrase) ".

However, if the second natural language-third natural language dictionary storage unit 19 stores a plurality of dictionaries, "pe
rformance (noun) "as a French entry
For example, "repre'sentation (female noun)" is described in any dictionary, but "spectacle (male noun)" is described only in some dictionaries.
Option (female noun) ”, and eventually“ dernie're ”
It is also conceivable to give weight to give priority to "repre'sentation dela saison".

Also, the "last performance of the se"
ason ", among the replacement rules shown in FIG. 10, rule number 6 and rule number 1 or 2,
And rule number 7 in combination,
The combination replacement rule shown in FIG. 11 can be created. If this combination replacement rule appears more than a predetermined number of times, for example (step S14 (Y
es)), and stored in the replacement rule storage unit 16 (FIG. 14).
Step S15).

If it is assumed that an economic dictionary has been created, the morphological analysis result of “bank clerk” shown in FIG.
In the same manner as in the case of “last performance of the season” (using the replacement rules of rule numbers 8 and 9 shown in FIG. 10). As a result, as a French entry for the Japanese entry “bank employee (noun)”, “ em
ploye ′ de banque ”and“ employe′e de banque ”
And “greffier de banque” and “employe′de riv
e ”and“ employe′e de rive ”and“ greffier de
rive ”can be obtained.

The replacement management unit 123 searches the second natural language-third natural language dictionary storage unit 19 using the analysis result of the supplied second natural language entry as a key, and further searches the third natural language dictionary storage unit 19 for the third natural language entry.
It has a third natural language dictionary search and replace function that searches the natural language dictionary storage unit 18 and replaces the second natural language entry with the third natural language entry.

The third natural language dictionary stored in the third natural language dictionary storage section 18 stores at least the original form, part of speech, and utilization information of the third natural language entry. FIG.
Reference numeral 2 denotes an example of a French dictionary entry stored in the third natural language dictionary storage unit 18, in which information on the original French form, part of speech, and variation is stored.

With reference to FIG. 12, the replacement management section 123 having the above-described functions will be specifically described. The English entry “fo” for the Japanese entry “four seasons (noun)” shown in FIG.
ur seasons (noun) ”is the analysis result
Consider the case where the data is supplied to the H.23. As described in the description of the second natural language analysis unit 122 having the second natural language dictionary use analysis function, the analysis result is “four seasons
(Noun) ”is divided into“ four (adjective) ”and“ seasons (noun) ”, and“ four ”is an adjective unchanged from the second natural language dictionary storage unit 17 shown in FIG.
It is assumed that information that "s" is a plural form of "season (noun)" is supplied (step S10 in FIG. 14).

First, the second natural language-third natural language dictionary storage unit 19 shown in FIG. 6B is searched, and the English entry “fo” is searched.
Obtain the French entries “quatre (adjective)” and “saison (female noun)” for “ur (adjective)” and “season (noun)”, where “season (noun)” is originally “se
"saison" (female noun)
Search for plurals in FIG. 12 and get “saisons”
e saisons ”.

However, in the above example, the replacement management section 123 having the third natural language dictionary search / replacement function includes the second natural language analysis section 122 in which the second natural language dictionary storage section 17 stores the second natural language dictionary storage section 17. Although the replacement is performed using the analysis result of the second natural language entry performed using the natural language dictionary, the analysis result of the second natural language entry supplied to the replacement management unit 123 is not necessarily the second natural language entry. The analysis result need not be the analysis result using the dictionary storage unit 17, but may be an analysis result by another analysis method of the second natural language analysis unit 122, such as a result of morphological analysis.

In addition, using only the heading of the second natural language entry as a key, searching the second natural language / third natural language dictionary storage unit 19, and extracting the third natural language entry that matches only with the face, etc. Depending on the analysis method in the second natural language analysis unit 122, various replacement methods can be considered.

When the second natural language entry of the first natural language to second natural language dictionary entry can be replaced with the third natural language entry, the first natural language entry and the third natural language entry are replaced. The set is supplied to the translated word selection information providing unit 124.

It should be noted that the replacement execution unit 1233
Even if the second natural language entry of the natural language versus second natural language dictionary entry could not be replaced with the third natural language entry, the analysis method that has not been performed on the second natural language entry is the second natural language analysis. If it remains in the unit 122, the second natural language entry is returned to the second natural language analysis unit 122, and the same processing is performed again (step S9 (Yes) in FIG. 14).

Next, the translated word selection information providing unit 124 will be additionally described. As described above, the translation word selection information providing unit 124 searches the third natural language corpus information storage unit 20 using the supplied third natural language entry as a key, and stores the stored statistical information of the third natural language corpus. Is used to determine the likelihood of the acquired third natural language entry, or when a plurality of third natural language entries are supplied, weighting is performed for selecting a translated word. And the first set of third natural language entries
It is stored in the natural language versus third natural language dictionary storage unit 13.

As described above, in the example of the Japanese entry “bank (noun)” shown in FIG. 2, the French entry obtained from the English entry “bank (noun)” corresponding to the Japanese entry is stored in the third natural language. The case where weighting is performed using the appearance frequency in the French corpus as shown in FIG. 7A stored in the language corpus information storage unit 20 has been described.

Here, an example in which the co-occurrence frequency in the third natural language corpus as shown in FIG. 7B is used as the statistical information of the third natural language corpus will be described. For example, assuming that it is desired to create an economic dictionary, the third natural language corpus information storage unit 20 stores the statistical information obtained from the third economic natural language corpus. FIG. 7B shows an example of the co-occurrence frequency acquired from the French corpus of economic relations, in which at least French 1 and its part of speech 1, French 2 and its part of speech 2, and their co-occurrence frequencies are stored. .

With reference to FIG. 7B, the function of the translated word selection information providing unit 124 will be specifically described. Suppose you want to create an economic dictionary. Here, "employe 'de banque" and "employe'e de banque" and "employe'e de banque"
que ”and“ greffier de banque ”and“ employe ′ d
e rive ”and“ employe′e de rive ”and“ greffi
When the third natural language corpus information storage unit 20 is searched, as shown in FIG. 7B, "employe '(male noun)" and "banque" (Female noun) "and" employ
e′e (female noun) ”and“ hanque ”(feminine noun) co-occur, but otherwise do not, so the“ employe ′ de banque ”and“ e ”in the supplied French entries
mploye'e de banque "is promising (step S7 in FIG. 14).

As for “rive (female noun)”,
As described in the example of the "bank (noun)", FIG.
FIG. 14A also shows that a French entry using "banque (female noun)" is more promising (step S7 in FIG. 14).

The weighting is similarly applied to the example of “Chiakiraku (noun)”. However, the French corpus used in this case is not limited to economic relations. In the case of the “four seasons (noun)” example, only one French entry has been acquired, but in order to verify the likelihood of the acquired French entry, weighting may be performed with reference to a French corpus. (Step S7 in FIG. 14).

These results are stored in the first natural language versus third natural language dictionary storage unit 13.

Finally, the first natural language-third natural language dictionary storage unit 13 will be described. The first natural language-third natural language dictionary storage unit 13 stores at least a heading of the first natural language and its part of speech, a bilingual expression of the third natural language corresponding to the first natural language entry and its part of speech, and If so, the translation word selection information of the third natural language can also be stored.

FIGS. 13A and 13B are examples of the first natural language-third natural language dictionary entries stored in the first natural language-third natural language dictionary storage unit 13. FIG. FIG.
(A) shows a Japanese headline of the first natural language and its part of speech,
This is an example in which a bilingual expression for the Japanese entry in French, which is a third natural language, and its part of speech are stored. FIG.
In (a), only promising French entries are stored in the first natural language versus third natural language dictionary storage unit 13 (step S8 in FIG. 14). FIG. 13B shows an example in which weight information for selecting a translated word is added, and unpromising French entries are also stored. However, even when leaving only promising French entries or leaving non-promising French entries, a certain threshold may be determined in advance, and French entries whose weight for selecting a translation is less than the threshold may be discarded. Conceivable. Also in FIG. 13B, “greffier de banque” and “emplo” acquired as French entries for the Japanese entry “bank employee (noun)”
y′e de rive ”and“ employe′e de rive ”and“
"greffier de rive" is discarded because the weight is less than or equal to the threshold. The created first natural language-third natural language dictionary may be manually checked.

The above-described processing is performed until there is no more dictionary entry stored in the first natural language-to-second natural language dictionary storage unit 10, and when there are no more entries, the processing is terminated (step S1).

The processing procedure of the dictionary creation method of the above embodiment is recorded as a program on a recording medium, and this recording medium is incorporated in a computer system, and the program recorded on the recording medium is downloaded or installed in the computer system. By operating the computer system with the program, it is of course possible to function as a dictionary creation device that implements the dictionary creation method, and by using such a recording medium, it is possible to enhance its distribution. You can do it.

[0151]

As described above, according to the present invention,
The first natural language / second natural language dictionary entry is fetched from the first natural language / second natural language dictionary storage unit, and corresponds to the second natural language entry of the fetched first natural language / second natural language dictionary entry. Third natural language entry to second
The first natural language entry obtained from the natural language-third natural language dictionary storage unit, and the third natural language entry replaces the second natural language entry of the first natural language-second natural language dictionary entry. And a first natural language-third natural language bilingual dictionary having a third natural language bilingual translation and its part of speech, which is a bilingual translation of the first natural language entry, to produce a first natural language-third natural language dictionary Since the first natural language and the third natural language dictionary are stored in the storage unit, the first natural language and the third natural language dictionary are used to store the first natural language and the third natural language.
Even without a dictionary of a combination of natural languages or a bilingual corpus, a first natural language-third natural language dictionary can be created.

Further, according to the present invention, the first natural language corpus information storage is referred to by using the first natural language entry of the first natural language vs. second natural language dictionary entry supplied from the dictionary entry reading unit as a key. Since unnecessary first natural language versus second natural language entries are removed, unnecessary dictionary entries are not created when, for example, a dictionary limited to a field is created, and as a result, the first natural language is not created. The speed and efficiency of creation and use of the third natural language dictionary can be improved. In the case of creating a dictionary limited to a field, if the first natural language-to-second natural language dictionary and the second natural language-to-third natural language dictionary to be used can use the dictionary of the field, the reliability is higher. Field-limited first
A natural language versus third natural language dictionary can be created.

Further, according to the present invention, the second natural language corpus information storage unit is referred to by using the second natural language entry of the first natural language versus second natural language dictionary entry supplied from the dictionary entry reading step as a key. Since the translated word selection information, which is the weighting information, is added, even when there are a plurality of translated word candidates from the first natural language to the second natural language, for example, as a technical term or a field-specific word, which translated word is more appropriate. It is possible to know which word candidate exists, and to assign a priority to which of the plurality of candidate word candidates appears more frequently. Furthermore, even when there is one translation word candidate, the likelihood of the translation word candidate can be determined, and as a result, a more reliable first natural language versus third natural language dictionary can be created. Also, since information for selecting a translated word is added at the stage of the second natural language, the same information can be used when only the third natural language is changed without changing the first natural language and the second natural language. When the third natural language is extended to various languages as a result, the first natural language-third natural language dictionary can be efficiently created.

According to the present invention, if the second natural language entry of the first natural language to second natural language dictionary entry cannot be replaced with the third natural language entry, the second natural language entry is analyzed. Since the third natural language entry is obtained based on the analysis result and the second natural language entry is replaced with the obtained third natural language entry,
For example, even if the second natural language entry is a compound word and the notation is not registered in the second natural language-third natural language dictionary, the second natural language entry may be used as a sub-expression unit forming a compound word for the second natural language entry. Analysis such as division is performed, and if each sub-expression is registered in the second natural language dictionary to the third natural language dictionary, the second natural language entry of the first natural language versus the second natural language dictionary entry is changed to the third natural language entry. It can be replaced with a language entry, and consequently, a third natural language entry for a compound word or the like that is difficult to replace with the notation as it is can also be obtained to create a first natural language-third natural language dictionary.

Further, according to the present invention, the replacement rule storage section is searched based on the analysis result of the second natural language analysis section, and the replacement rule is extracted. In the case of, the third natural language entry is acquired from the second natural language-third natural language dictionary storage unit by using the plurality of replacement rules in combination, so that, for example, the second natural language entry is a compound word or a variant form. Alternatively, even if the notation is not registered in the second natural language versus third natural language dictionary as it is in a combination thereof, the analysis result of the second natural language entry, the replacement rule suitable for the analysis result, and the second natural language Using the third natural language dictionary to ensure that the first even for more complex second natural language entries
The second natural language entry of the natural language to second natural language dictionary entry may be replaced with a third natural language entry. Further, by using a plurality of replacement rules in combination, even if a replacement rule that can replace the analysis result of the second natural language entry in a lump is not stored in the replacement rule storage unit, the first natural language versus the second natural language entry can be used. A third natural language entry for a more complex second natural language entry, such as a compound word that is difficult to replace in its native form, such as a compound word that can replace the second natural language entry of the natural language dictionary entry with a third natural language entry And the first natural language dictionary can be created. Furthermore, because you can use a combination of replacement rules,
There is no need to store all possible replacement rules in the replacement rule storage unit, and the first natural language-third natural language dictionary can be created efficiently.

Further, according to the present invention, when a plurality of replacement rules are used in combination, the number of usages of the combined replacement rule is counted, and when the number of usages exceeds a predetermined number, the combined replacement rule is used. Since the rules are stored in the replacement rule storage unit, a combination replacement rule that is frequently used is acquired and stored in the replacement rule storage unit, so that the second natural language of the first natural language versus second natural language dictionary entry is obtained. It is possible to increase the speed of replacing the entry with the third natural language entry.
A natural language dictionary can be created efficiently.

According to the present invention, there is provided the second natural language dictionary storing the original form, part of speech, and utilization information of the second natural language, and the second natural language analysis unit searches the second natural language dictionary storage unit. Then, since the second natural language entry is analyzed, even if, for example, the second natural language entry includes a variant and is not registered in the second natural language-third natural language dictionary as it is, the second natural language entry is analyzed. The original form of the variant included in the second natural language entry can be obtained from the natural language dictionary, and if this original form is registered in the second natural language versus third natural language dictionary, the first natural language versus the second natural language The second natural language entry of the dictionary entry can be replaced with the third natural language entry, and as a result, the third natural language entry for the second natural language entry such as a variant that is difficult to replace with the literal notation is also obtained. Can, it is possible to create precisely the first natural language versus the third natural language dictionary.

Further, according to the present invention, there is provided a third natural language dictionary storing the original form, part of speech, and utilization information of the third natural language, and the analysis result of the second natural language analysis unit or the replacement rule search is provided. The third natural language dictionary storage unit is searched based on the replacement rule extracted by the unit, and the second natural language entry of the first natural language versus second natural language dictionary entry is replaced with the third natural language entry. If the natural language entry includes a variant and the notation is not registered in the second natural language versus third natural language dictionary, the variant included in the second natural language entry is converted to the original form,
When the corresponding third natural language entry is obtained in the original form from the second natural language pair to the third natural language dictionary, the third natural language dictionary is searched and the original form of the third natural language entry is converted into an appropriate variant notation. The third natural language entry corresponding to the second natural language entry, such as a variant that is difficult to replace with the notation as it is, can be converted into an accurate expression, and the first natural language-third natural language dictionary is converted. It can be created accurately.

Further, according to the present invention, a plurality of second natural language entries of the first natural language versus second natural language dictionary entry are obtained, and the third natural language entry corresponding to the second natural language entry is duplicated. If acquired, the third natural language entry acquired in duplicate is determined to be more appropriate. Therefore, if there are a plurality of candidate translations at the stage of the second natural language entry, a plurality of second natural language entries are obtained. If the third natural language entries obtained from the natural language entries overlap, translation word selection information indicating that the duplicated third natural language entry is more appropriate than the third natural language entry that has appeared alone can be given, As a result, a highly reliable first natural language to third natural language dictionary can be created.

According to the present invention, there is provided a third natural language corpus information storage unit for storing the statistical information of the third natural language corpus, and the information of the third natural language corpus information storage unit is used. Since translation word selection information indicating whether the third natural language entry acquired by the dictionary creation management unit is appropriate is added, while tracing from the first natural language entry to the third natural language entry via the second natural language entry, Even if the number of translation candidates increases, it is possible to know which translation is more appropriate, such as a technical term or a field-specific word. In addition, a priority order of a word that is more frequently used among a plurality of translation word candidates can be given, and even when there is only one translation word candidate of the third natural language, the likelihood of the translation word candidate can be determined. Further, when the second natural language entry is a compound word and the third natural language entry cannot be obtained by the notation as it is, and the analysis is performed to obtain the third natural language entry, a word candidate in the third natural language entry is obtained. Even if there are a plurality of words, it is possible to select a more appropriate word combination from the co-occurrence information between words included in the acquired third natural language entry, and as a result, the first natural language versus the first 3. A natural language dictionary can be created.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a dictionary creation device according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating a first natural language-second natural language dictionary entry stored in a first natural language-second natural language dictionary storage unit used in the dictionary creation device of FIG. 1;

FIG. 3 is a diagram illustrating first natural language corpus information stored in a first natural language corpus information storage unit used in the dictionary creation device in FIG. 1;

FIG. 4 is a diagram showing second natural language corpus information stored in a second natural language corpus information storage unit used in the dictionary creation device of FIG. 1;

FIG. 5 is a diagram illustrating an execution result of an entry selection information providing unit used in the dictionary creation device of FIG. 1;

FIG. 6 is a diagram showing a second natural language-third natural language dictionary entry stored in a second natural language-third natural language dictionary storage unit used in the dictionary creation device of FIG. 1;

FIG. 7 is a diagram showing third natural language corpus information stored in a third natural language corpus information storage unit used in the dictionary creation device of FIG. 1;

FIG. 8 is a diagram showing an analysis result of a second natural language entry by a second natural language analysis unit used in the dictionary creation device of FIG. 1;

9 is a diagram showing a second natural language dictionary entry stored in a second natural language dictionary used in the dictionary creation device of FIG. 1;

FIG. 10 is a diagram showing a replacement rule stored in a replacement rule storage unit used in the dictionary creation device of FIG. 1;

FIG. 11 is a diagram showing a replacement rule used in a replacement rule search unit used in the dictionary creation device of FIG. 1;

FIG. 12 is a diagram illustrating a third natural language dictionary entry stored in a third natural language dictionary storage unit used in the dictionary creation device in FIG. 1;

13 is a diagram showing a first natural language-third natural language dictionary entry stored in a first natural language-third natural language dictionary storage unit used in the dictionary creation device of FIG. 1;

FIG. 14 is a flowchart showing the operation of the dictionary creation device of FIG. 1;

[Explanation of symbols]

 10 first natural language versus second natural language dictionary storage unit 11 dictionary entry reading unit 12 dictionary creation management unit 13 first natural language versus third natural language dictionary storage unit 14 first natural language corpus information storage unit 15 second natural language Corpus information storage unit 16 Replacement rule storage unit 17 Second natural language dictionary storage unit 18 Third natural language dictionary storage unit 19 Second natural language versus third natural language dictionary storage unit 20 Third natural language corpus information storage unit 121 Selection of entry Information addition unit 122 Second natural language analysis unit 123 Replacement management unit 124 Translated word selection information addition unit 1231 Replacement rule search unit 1232 Replacement rule creation unit 1233 Replacement execution unit

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Hiromi Nakaiwa 2-3-1, Otemachi, Chiyoda-ku, Tokyo F-term in Nippon Telegraph and Telephone Corporation (reference) 5B091 AA01 BA02 CC16 EA07 EA17

Claims (15)

[Claims] 1. A dictionary creating apparatus for creating a bilingual dictionary of a natural language pair, comprising: a heading of a first natural language and its part of speech; a bilingual expression of a second natural language which is a bilingual translation of the first natural language entry; A first natural language versus second natural language dictionary storage unit for storing a first natural language versus second natural language dictionary having the part of speech, a second natural language heading and its part of speech, and the second natural language entry A second natural language pair and a third natural language dictionary storage unit for storing a second natural language pair and a third natural language dictionary having a bilingual expression of a third natural language and its part of speech, which is a bilingual translation of the first natural language pair A dictionary entry reading unit for retrieving the first natural language pair to the second natural language dictionary entry from the second natural language dictionary storage unit; and a dictionary entry reading unit corresponding to the retrieved first natural language to second natural language dictionary entry. A third natural language entry is obtained from the second natural language to third natural language dictionary storage, and a second natural language entry of the first natural language to second natural language dictionary entry is replaced with the third natural language entry. A dictionary creation management unit, wherein a first natural language entry to a third natural language dictionary entry generated by replacing the second natural language entry of the first natural language entry to the second natural language dictionary entry with a third natural language entry; A first natural language-third natural language bilingual dictionary having a first natural language heading and its part of speech, a third natural language bilingual translation of the first natural language entry and its part of speech, and A dictionary creation device comprising: a first natural language-third natural language dictionary storage unit for storing. A first natural language corpus information storage unit that stores statistical information of the first natural language corpus, wherein the dictionary creation management unit is configured to output the first natural language corpus information supplied from the dictionary entry reading unit. An entry selection information providing unit that refers to the first natural language corpus information storage unit and uses the first natural language entry of the language-to-second natural language dictionary entry as a key to remove unnecessary first natural language-to-second natural language entries 2. The dictionary creation device according to claim 1, comprising: And a second natural language corpus information storage unit for storing statistical information of a second natural language corpus, wherein the entry selection information providing unit includes a first natural language corpus information storage unit. Using the second natural language entry of the natural language versus second natural language dictionary entry as a key
3. The dictionary creating apparatus according to claim 2, further comprising translation word selection information / weighting information provision means for referring to the natural language corpus information storage unit and providing translation word selection information as weighting information. 4. The dictionary creation management unit, when the second natural language entry of the first natural language-to-second natural language dictionary entry cannot be replaced with the third natural language entry, A second natural language analysis unit for analyzing the third language, and a third natural language analysis unit based on the analysis result of the second natural language analysis unit.
4. The dictionary creating apparatus according to claim 1, further comprising: a replacement execution unit that acquires a natural language entry and replaces the second natural language entry with the acquired third natural language entry. 5. The apparatus according to claim 1, further comprising: a replacement rule storage unit storing a replacement rule for replacing the second natural language with a third natural language, wherein the dictionary creation management unit stores the analysis result of the second natural language analysis unit. A replacement rule search unit that searches the replacement rule storage unit based on the replacement rule, and obtains a replacement rule. The replacement execution unit uses the replacement rule extracted by the replacement rule search unit. 5. The dictionary creation device according to claim 4, further comprising means for acquiring a third natural language entry from said second natural language to third natural language dictionary storage unit by using said plurality of replacement rules in combination. apparatus. 6. When the replacement execution unit uses a plurality of replacement rules extracted by the replacement rule search unit in combination, the dictionary creation management unit counts the number of times the combined replacement rules are used. Means,
6. The dictionary according to claim 5, further comprising: a replacement rule creating unit that stores the combined replacement rule in the replacement rule storage unit when the counted usage number of the replacement rule becomes equal to or more than a predetermined number. Creating device. 7. A second natural language dictionary storage unit for storing a second natural language dictionary storing an original form, part of speech, and utilization information of the second natural language, wherein the second natural language analysis unit comprises: 7. The dictionary creation device according to claim 4, further comprising means for searching a second natural language dictionary storage unit and analyzing the second natural language entry. 8. A third natural language dictionary storage unit for storing a third natural language dictionary storing an original form, a part of speech, and inflection information of the third natural language, wherein the replacement execution unit includes the second natural language dictionary. The third natural language dictionary storage unit is searched based on the analysis result of the natural language analysis unit or the replacement rule retrieved by the replacement rule search unit, and the second natural language entry of the first natural language versus second natural language dictionary entry is searched. The dictionary creating apparatus according to any one of claims 4 to 7, further comprising: means for replacing with a third natural language entry. 9. The dictionary creation management unit obtains a plurality of second natural language entries of the first natural language versus second natural language dictionary entry, and generates a third natural language entry for the second natural language entry.
4. The method according to claim 1, further comprising a translation word selection information providing unit that determines that the third natural language entry acquired in duplicate is more appropriate when the natural language entry is acquired in duplicate. 8. The dictionary creation device according to any one of 8. 10. A third natural language corpus information storage unit storing statistical information of a third natural language corpus, wherein the dictionary creation management unit stores information of the third natural language corpus information storage unit. 2. The apparatus according to claim 1, further comprising means for providing translation word selection information indicating whether the third natural language entry acquired by the dictionary creation management unit is appropriate.
The dictionary creation device according to any one of claims 9 to 9. 11. A dictionary creation method for creating a bilingual dictionary of a natural language pair, comprising: a heading and a part of speech of a first natural language; a bilingual expression of a second natural language that is a bilingual translation of the first natural language entry; A first natural language-second natural language dictionary storage unit storing a first natural language-second natural language dictionary having the part of speech in a first natural language-second natural language dictionary storage unit; A second natural language-third natural language dictionary having a natural language heading and its part of speech, a third natural language bilingual expression that is a bilingual translation of the second natural language entry, and its part of speech is stored in the second natural language-third language. Second stored in natural language dictionary storage
A natural language-to-third natural language dictionary storage unit storing step; a dictionary entry reading step of obtaining the first natural language-to-second natural language dictionary entry from the first natural language-to-second natural language dictionary storage unit; Obtaining a third natural language entry corresponding to the second natural language entry of the first natural language to second natural language dictionary entry from the second natural language to third natural language dictionary storage unit; A dictionary creation management step of replacing a second natural language entry of the second natural language dictionary entry with the third natural language entry; and replacing the second natural language entry of the first natural language versus second natural language dictionary entry with a third natural language entry And a part of speech of the first natural language based on the first natural language-third natural language dictionary entry generated by A first natural language-third natural language dictionary storing step of creating and storing a first natural language-third natural language bilingual dictionary having a third natural language bilingual translation of a natural language entry and its part of speech; A dictionary creation method comprising: 12. The method according to claim 12, further comprising the step of storing statistical information of the first natural language corpus in a first natural language corpus information storage unit, wherein the dictionary creation management step includes the step of storing the first natural language corpus information supplied from the dictionary entry reading step. An entry selection information providing step of referring to the first natural language corpus information storage unit using the first natural language entry of the language-to-second natural language dictionary entry as a key and removing unnecessary first natural language-to-second natural language entries 12. The dictionary creation method according to claim 11, comprising: 13. The method according to claim 13, further comprising the step of storing statistical information of the second natural language corpus in a second natural language corpus information storage unit, wherein the step of providing entry selection information includes the step of: Natural language versus second
A translation word selection information / weighting information providing step of referring to the second natural language corpus information storage unit using the second natural language entry of the natural language dictionary entry as a key and providing translation word selection information as weighting information. 13. The dictionary creation method according to claim 12, wherein: 14. The dictionary creation management step, if the second natural language entry of the first natural language-to-second natural language dictionary entry cannot be replaced with the third natural language entry, the second natural language entry A second natural language analysis step of analyzing the second natural language entry, and obtaining a third natural language entry based on the analysis result of the second natural language analysis step, and replacing the second natural language entry with the obtained third natural language entry. 14. The dictionary creation method according to claim 11, further comprising an execution step. 15. A recording medium storing a dictionary creation program for creating a bilingual dictionary of a natural language pair, comprising: a heading of a first natural language and its part of speech; and a second natural language which is a bilingual translation of the first natural language entry. A first natural language versus a second natural language dictionary storage for storing a first natural language versus a second natural language dictionary having a bilingual language expression and its part of speech in a first natural language versus a second natural language dictionary storage A second natural language-third natural language dictionary having a step, a heading of the second natural language and its part of speech, a bilingual expression of the third natural language which is a bilingual translation of the second natural language entry and its part of speech, The second language stored in the natural language-third natural language dictionary storage unit
A natural language-to-third natural language dictionary storage unit storing step; a dictionary entry reading step of obtaining the first natural language-to-second natural language dictionary entry from the first natural language-to-second natural language dictionary storage unit; Obtaining a third natural language entry corresponding to the second natural language entry of the first natural language to second natural language dictionary entry from the second natural language to third natural language dictionary storage unit; A dictionary creation management step of replacing a second natural language entry of the second natural language dictionary entry with the third natural language entry; and replacing the second natural language entry of the first natural language versus second natural language dictionary entry with a third natural language entry And a part of speech of the first natural language based on the first natural language-third natural language dictionary entry generated by A first natural language-third natural language dictionary storing step of creating and storing a first natural language-third natural language bilingual dictionary having a third natural language bilingual translation of a natural language entry and its part of speech; A storage medium storing a dictionary creation program, characterized by comprising: