JPH113336A - Machine translation apparatus and computer-readable recording medium recording machine translation processing program - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To allow a user to register a function word by eliminating side effects caused by the registration of the function word, thereby obtaining a high-accuracy translation result sentence, and reducing the load of post-editing. A computer-readable recording medium that records a machine translation device and a machine translation processing program. A headword and a part of speech are input (S51).
0: Y), if it is determined that the entry is not in the dictionary (S520: N), dictionary information is newly created (S53).
0). And when it is registered in the dictionary (S550:
Y), it is determined whether or not the part of speech is a particle (S560). If the part of speech is a particle (S560: Y), a morphological analysis is performed with a dummy noun attached (S570), and each obtained element is identified as an element headword. Is registered in the dictionary together with the dictionary information and the division flag (S580 to S590).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a machine translation device for translating an original sentence of a first language into a second language, and a computer-readable recording medium on which a machine translation processing program is recorded.

[0002]

2. Description of the Related Art Conventionally, in this type of machine translation apparatus, a plurality of types of dictionaries are generally prepared. In a translation process, a dictionary most suitable for a text to be translated is used. Is configured.
Here, as the type of the prepared dictionary, for example,
A wide range of words and phrases are registered in general translations and grammar, and a basic dictionary stored in advance by a machine translation device and a specialized dictionary in which words and translations used only in a specific field are registered And a user dictionary that the user can independently edit.

[0003] Not only machine translation processing but also manual translation processing requires the use of various types of dictionaries, general learning dictionaries with abundant examples, and specialized fields in which bilingual technical terms are arranged. Dictionaries, specialized books, magazines, papers, etc. Further, in the actual translation processing, these commercially available ones are not enough, and the knowledge of the person who performs the translation processing, a glossary of original terms and the like are also used. This situation is the same in translation processing by a machine.

[0004] Since actual translations contain a considerable proportion of terms and phrases used by individual users, the basic dictionary and specialized dictionary prepared in advance by the system are not sufficient. Yes, there are many unregistered words. In addition, even if a phrase is registered in the basic dictionary or specialized dictionary, its translation may not always be the most suitable for the user, and may be translated literally into a fixed phrase, unique expression, or idiom used by a specific user. Therefore, a translated sentence that does not make sense or an unnatural translated sentence may be generated. The more such translations, the more extensive post-editing must be performed by the user. Therefore, a user dictionary that the user can freely edit is indispensable for obtaining a natural translation result.

However, since the user is not familiar with the inside of the machine translator, there is a problem that an appropriate user dictionary cannot always be created. That is, the user tends to register the headword only by considering the translation of the focused part, but the translation of the part other than the focused part becomes inappropriate because the registered headword is registered. Side effects may occur.

Therefore, in the conventional machine translator, the part of speech that can be registered in the user dictionary is limited to only nouns or only independent words such as verbs and adjectives so as not to induce such adverse effects.

Japanese Patent Application Laid-Open No. 7-85059 discloses a dictionary inspection apparatus which issues a warning for dictionary registration that adversely affects translation processing.

[0008]

However, as described above, providing only a limit to dictionary registration, as well as a novice user, is very difficult for users who use the machine translator freely. I have. That is, not only the editing of independent words, but also the use of collocations with a high degree of freedom among particles and auxiliary verbs called functional words, even though the sentence becomes more natural, such parts of speech cannot be analyzed by morphological analysis. It was not open to the user because it involved the danger of proper results.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and enables a function word to be registered in a user dictionary, suppresses side effects in translation processing due to the registration, and achieves high accuracy. An object of the present invention is to provide a machine translation device capable of obtaining a translation result and a computer-readable recording medium recording a machine translation processing program.

[0010]

In order to achieve this object, a machine translation apparatus according to a first aspect of the present invention provides a machine translation apparatus comprising: A dictionary that stores bilingual translations and various dictionary registration information used for translation processing such as connection information necessary for morphological analysis processing, and adds information such as headwords of the first language to the dictionary. Dictionary editing means for correcting, deleting, etc., morphological analysis means for performing a morphological analysis by searching for an entry word registered in the dictionary for the input original text of the first language, and morphological analysis by the morphological analyzing means A parsing means for parsing a result, wherein the original text of the first language is translated into the second language. To be A specific part-of-speech determining means for determining whether or not the putative word is a specific part-of-speech, and when the specific part-of-speech determining means determines that the specific part of speech, the morphological analysis means performs morphological analysis of the headword. Headword division means for performing division into a plurality of elements, and registering each of the plurality of elements divided by the headword division means as the headword in the dictionary, A registration unit that adds element headword information indicating that the element is a headword element as dictionary registration information of each element.

Therefore, the specific part-of-speech determining means determines whether the headword to be registered in the dictionary is a specific part-of-speech, and the headword dividing means determines that the specific part-of-speech determining means is the specific part of speech. In this case, a morphological analysis of the headword is performed by the morphological analysis unit to divide the headword into a plurality of elements, and the registration unit registers each of the plurality of elements divided by the headword division unit as a headword in the dictionary. At the same time, element headword information indicating that the element is an element of the original headword before division by the headword division means is added as dictionary registration information of each element.

Further, the machine translation device according to claim 2 is
The specific part-of-speech determining means is configured to determine whether the headword to be registered is a particle, and the headword dividing means converts a temporary noun immediately before the headword to be registered. It is configured to perform morphological analysis by connecting.

Therefore, the specific part-of-speech determining means determines whether or not the headword to be registered is a particle. If the specific part-of-speech determining means determines that the headword is a particle,
The headword division unit connects the provisional noun immediately before the headword to be registered and performs morphological analysis to divide the headword into a plurality of elements.

Further, the machine translation device according to claim 3 is
The syntax analysis means includes: an element headword determination means for determining whether a headword having the element headword information is present in a morphological analysis result of the input original text by the morphological analysis means; A part-of-speech determining means for determining whether a plurality of headwords having the element headword information are used as the specific part-of-speech in the original text when the headword is determined to be present by the word determining means; Replacement means for replacing a plurality of headwords having the element headword information with the original headword when the part-of-speech determination means determines that the headword is used as the specific part of speech.

Therefore, the element headword determining means determines whether or not a headword having the element headword information exists in the morphological analysis result of the input original sentence by the morphological analysis means. Determines whether or not a plurality of headwords having the element headword information are used as the specific part of speech in the original sentence when the element headword determination means determines that the headword exists. The replacement means, when it is determined by the part of speech determination means is used as the specific part of speech,
A plurality of headwords having the element headword information are replaced with the original headword.

According to a fourth aspect of the present invention, there is provided a computer-readable recording medium storing a machine translation processing program, comprising: a headword of a first language; a part of speech; and a translation of a second language corresponding to each headword. A dictionary editing program for adding, modifying, deleting, etc., information such as headwords of the first language with respect to a dictionary storing various dictionary registration information used for translation processing such as connection information necessary for morphological analysis processing A morphological analysis program for performing a morphological analysis by searching for an entry word registered in the dictionary for the input original text of the first language, and a syntax for performing a syntax analysis on a morphological analysis result by the morphological analysis program In particular, the dictionary editing program is designed to store an analysis program and a headword to be registered in the dictionary with a specific part of speech. A specific part-of-speech determination program that determines whether or not the headword is divided into a plurality of elements by performing a morphological analysis of the headword with the morphological analysis program when the specific part-of-speech determination program determines that the specific part of speech is A headword division program, and each of the plurality of elements divided by the headword division program are registered in the dictionary as headwords, and are elements of the original headword before being divided by the headword division program. And a registration program for adding element headword information indicating the above as dictionary registration information of each element.

Therefore, by executing a program using this recording medium, it is determined whether or not the headword to be registered in the dictionary is a specific part of speech. If the headword is determined to be the specific part of speech, The morphological analysis of the headword is performed by the morphological analysis program, the headword is divided into a plurality of elements, and each of the plurality of divided elements is registered in the dictionary as a headword, and the original headword before the division is obtained. Element headword information indicating that the element is a word element is added as dictionary registration information of each element.

According to a fifth aspect of the present invention, there is provided a computer-readable recording medium storing the machine translation processing program according to the fifth aspect, wherein the specific part-of-speech determination is performed so as to determine whether the headword to be registered is a particle. A program and a headword division program for performing morphological analysis by connecting a temporary noun immediately before the headword to be registered are stored.

Accordingly, by executing a program using this recording medium, it is determined whether or not the headword to be registered is a particle, and if it is determined that the headword is a particle, the registration is attempted. A tentative noun can be connected immediately before the headword to be morphologically analyzed and divided into a plurality of elements.

According to a sixth aspect of the present invention, there is provided a computer-readable recording medium storing the machine translation processing program according to the first aspect of the present invention, wherein a heading having the element headword information is included in a morphological analysis result of the input original sentence by the morphological analysis program. An element headword determination program for determining whether or not a word exists; and, when the element headword determination program determines that the headword is present, a plurality of headwords having the element headword information are included in the original text. A part-of-speech determination program that determines whether or not the part-of-speech is used as the specific part-of-speech; and a plurality of parts having the element headword information when the part-of-speech determination program determines that the part-of-speech is used. And a replacement program for replacing the headword with the original headword. To.

Therefore, by executing a program using this recording medium, it is possible to determine whether or not a headword having the element headword information exists in the morphological analysis result of the input original text by the morphological analysis means. Determining, when it is determined that the headword is present, determining whether or not a plurality of headwords having the element headword information are used as the specific part of speech in the original sentence; Thus, when it is determined that the headword is used as the specific part of speech, a plurality of headwords having the element headword information can be replaced with the original headword.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment embodying a machine translation apparatus according to the present invention will be described below with reference to the drawings.
In the present embodiment, an example will be described in which a user can register particles in a user dictionary in a machine translator using Japanese as a first language and English as a second language.

First, FIG. 1 shows a block diagram of a schematic configuration of a machine translation apparatus according to the present embodiment.

As shown in FIG. 1, the machine translation apparatus of this embodiment has an input unit 2 composed of a keyboard or the like for inputting a Japanese original to be translated, various operation instructions, and the like, and is used for translation processing. An external storage device 4 including a hard disk, an optical disk, or the like, in which various dictionaries are stored, a display unit 6 including a CRT or the like for displaying an input Japanese original and an English translation as a translation result, and the like. A display control unit 7 for controlling data to be displayed on the display unit 6, a printing unit 8 such as a laser printer for printing and outputting English translations and the like as translation results, the input unit 2 and external translation The control unit 10 performs a translation process based on the data input from the device 4, displays the translation result on the display unit 6, and outputs the translation result from the printing unit 8.

The input unit 2 includes a character key for inputting characters, an input sentence editing key for activating various processes,
Various function keys such as a translation execution key, a dictionary editing key, an output sentence editing key, an end key, and a print execution key; cursor keys used for editing an input sentence or an output sentence; operating a screen displayed on the display unit 6; An insert key, a delete key, a decision key, a next candidate selection key, and the like are provided.

The external storage device 4 stores a basic dictionary 12 and a user dictionary 14 used for translation processing. This user dictionary 14 constitutes the dictionary of the present invention.

The dictionaries used by the machine translation device include a basic dictionary held in advance by the machine translation device, a specialized dictionary edited for each specialized field, and a user dictionary constructed by the user independently. The specialized dictionary and the user dictionary may or may not be used. If a specialized dictionary is used, it is stored in the external storage device 4. Further, the specialized dictionary may be configured such that a plurality of dictionaries can exist simultaneously for each specialized field, and the user dictionary can be simultaneously present for each user or each usage field, for example. In the present embodiment, their use is not specifically stated.

FIG. 2 shows an example of the data structure of the above-mentioned basic dictionary. Each dictionary registration information is composed of a headword 30, a part of speech 32 of each headword, a translated word 34 for each headword, and various dictionary information 36 including connection information and semantic information before and after.

FIG. 3 shows an example of the data structure of the above-mentioned user dictionary. Each dictionary registration information is composed of a headword 40, a part of speech 42 of each headword, a translated word 44 for each headword, and various dictionary information 46 including connection information and semantic information before and after. .

Next, the control unit 10 includes a CPU 18 and an R
It comprises a well-known microcomputer having an OM 20, a RAM 22, an input / output port 24, and a bus line 26 connecting these components. In the control unit 10, various processes are executed by a program stored in the ROM 20. For example, a translation and editing process for translating a Japanese original sentence input from the input unit 2 into English, editing the original sentence and the translated sentence, and a dictionary editing for constructing a user dictionary used at the time of translation. Processing and the like are executed.

The ROM 20 stores various types of information used in a translation process described later.

The RAM 22 stores the Japanese original sentence input from the input unit, the English translation translated and edited by the translation and editing processing, and the dictionary registration information of the user dictionary 14.

Next, the translation / editing process executed by the control unit 10 will be described with reference to the flowcharts shown in FIGS.

As shown in FIG. 4, when the translation / editing process is executed, first, step 110 (hereinafter, step is referred to as S
Abbreviated. In the same manner as in the other steps), the user waits until the user operates a character key or the like of the input unit 2 to input a Japanese original text to be translated. Here, the input of the original text may be configured to read the file of the original text that has already been created. Then, when a Japanese original sentence (hereinafter, input sentence) is input, the process proceeds to S120, and the input sentence is displayed in a predetermined area of the display unit 6.

Next, in S130, it is determined whether or not the input sentence edit key provided in the input unit 2 has been operated. If not (S130: N), the process proceeds to S150. If the input sentence edit key is operated (S13
0: Y), and proceeds to S140 to execute an input sentence editing process. When the input sentence editing process is executed, the user uses the cursor keys, the delete key, the insert key, the character keys, and the like provided on the input unit 2 to change the input sentence displayed in a predetermined area of the display unit 6. Can be directly modified. S1
At 50, it is determined whether or not the translation execution key provided on the input unit 2 has been operated, and if it has not been operated (S15).
0: N), the process proceeds to S190, and if the translation execution key is operated (S150: Y), the process proceeds to S160,
A translation process for translating a Japanese input sentence into an English translated sentence (hereinafter referred to as an output sentence) is executed, and in S170, the output sentence obtained by the translation process is paired with the input sentence in the RAM.
In step S180, the input sentence and the output sentence are displayed in predetermined areas provided in the display unit 6, respectively.

The translation process may be performed for all input sentences at once, or for a range of sentences, and may be performed only for the specified range of sentences.

In S190, it is determined whether or not the dictionary editing key provided on the input unit 2 has been operated. If not (S190: N), the process proceeds to S210. Also,
If the dictionary edit key is operated (S190: Y), S
The process proceeds to 200, where a dictionary editing process is executed. When the dictionary editing process is executed, the user can correct or delete the content of the information registered in the user dictionary 14 or register a new necessary phrase. Also, S2
The processing of 00 functions as a dictionary editing means or a dictionary editing program of the present invention.

In S210, it is determined whether or not the output sentence editing key provided on the input unit 2 has been operated. If not (S210: N), the flow proceeds to S230. If the output sentence edit key is operated (S210:
Y), and proceeds to S220 to execute an output sentence editing process.
When the output sentence editing process is executed, the user enters S1
When the translation process is executed at 60, the display unit 6
The output sentence (English translation) displayed in the specified area of
As in the input sentence editing process in S140, it is possible to directly make corrections using a cursor key, a delete key, an insert key, a character key, and the like provided on the input unit 2.

In S230, it is determined whether or not the end key provided on the input unit 2 has been operated. If the end key has not been operated (S230: N), the process proceeds to S240, for example, a translated output sentence is printed. It detects other key operations such as a print execution key that is output to the unit 8 and prints, and executes a process corresponding to the operated key.

Then, S140, S180, S20
After the processing of 0, S220, and S240, the process returns to S130, and the same processing is repeatedly executed.

If it is determined in S230 that the end key has been operated (S230: Y), the process is terminated.

As described above, in the translation / editing process, after the Japanese original sentence is input, the input sentence editing process, translation process, dictionary editing process, and the like are performed in accordance with the operation of various keys provided on the input unit 2. Various processes such as an output sentence editing process are executed. Then, if the translation process is executed by operating the translation execution key, if the obtained output sentence (English translation) is incorrect or unnatural, the input sentence edit key is operated. Execute the input sentence editing process to re-edit the Japanese original sentence that is the input sentence, or edit the user dictionary by operating the user dictionary edit key, and then operate the translation execution key again to re-translate Alternatively, an output sentence editing key is operated to execute an output sentence editing process, and an English translation sentence, which is an output sentence, is directly edited.

If the translation result desired by the user can be obtained by repeating such processing, the print key is operated to output the translation result from the printing unit 8 or to the RAM 22 or the external storage device 4 as text data. And then operate the end key to end the process.

Next, S2 in the flowchart shown in FIG.
00 will be described in detail with reference to the flowchart shown in FIG.

First, in S510, it is determined whether or not the user has input a headword to be edited and its part of speech. If it has been input (S510: Y), the process proceeds to S520. In S520, it is determined whether or not the input headword and part-of-speech pair is already registered in the user dictionary 14. If already registered (S520: Y), S54
Move to 0. In S540, the data in the user dictionary is re-edited, and the process proceeds to S550. If it has not been registered yet (S520: N), the process proceeds to S530. S53
In the case of 0, the dictionary information is newly edited and the process proceeds to S550. In S550, it is determined whether or not the edited dictionary information is registered. If registered (S550: Y),
The process moves to S560. In S560, it is determined whether or not the part of speech of the headword input by the user is a particle. Note that the processing of S560 functions as the specific part-of-speech determining means or the specific part-of-speech determining program of the present invention.

If it is a particle (S560: Y), the flow shifts to headword division processing S570. Headword division processing S57
In the case of 0, a dummy noun (temporary noun) is added before the input headword, and morphological analysis processing S250 is performed.
Move to 80. This is because, although there is a difference depending on the particle, the particle basically follows the noun, and the morphological analysis result is obtained on the assumption of the particle. The processing of S570 functions as a headword division unit or a headword division program of the present invention.

In S580, a headword division process S570
Of each divided element (hereinafter, referred to as an element headword) retrieved as a result of the above, and connection information such that only the next element headword is searched immediately after each element. . The backward connection of the last element headword is the connection information that the word originally has. A division flag is set for the dictionary information of all the element headwords, and the original headword and the part of speech before the division are stored as dictionary information.
Move to 0. Note that the processing in S580 functions as the registration unit or the registration program of the present invention, and the division flag and the original headword correspond to the element headword information.

Here, FIG. 3 shows an example of a headword divided into individual elements when the particle “related” is registered in the user dictionary.

When the part of speech of the input headword is not a particle (S560: N), the flow shifts to S590.

In step S590, the edited information is stored in the user dictionary 14, and the flow advances to step S510 to prepare for editing the next headword.

If it is not registered in S550 (S55
0: N), the dictionary information is discarded, and the routine goes to S510.

In S510, when the headword and the part of speech have not been input (S510: N), this processing ends.

Next, S1 in the flowchart shown in FIG.
FIG. 6 shows a translation process executed in step S60 for translating an input sentence (Japanese original sentence) into an output sentence (English translated sentence).
This will be described according to the flowchart shown in FIG.

First, in S250, predetermined judgment rules in the morphological analysis processing, the basic dictionary 12, and the user dictionary 1
Morphological analysis that searches each dictionary for words that make up the input sentence using the connection information in 4 and adds search information such as translations and parts of speech to the morphological analysis results for those for which a headword could be searched. Execute the process. Here, the user dictionary 14 has priority over the same character string. If a specialized dictionary is used, it is given priority next to the user dictionary.
When not found in the user dictionary 14, the basic dictionary 12 is searched. At this time, the particle in the user dictionary is searched only when all the elements are continuously searched according to the connection information. The process of S250 functions as a morphological analysis unit or a morphological analysis program of the present invention.

In S260, the result of the morphological analysis process in S250 is received, and an analysis process for analyzing a Japanese structure is executed using a predetermined analysis grammar stored in the ROM 20. At this time, the processing shifts to the determination processing of S400, and the Japanese morphological analysis is performed while processing the part where the division flag indicating the headline divided by the headword division means of the dictionary editing processing S200 is added to the morphological analysis result. Execute the process. Incidentally, the processing of S260 is performed as a syntax analysis means or a syntax analysis program of the present invention.
The process of S400 functions as an element headword determination unit or an element headword determination program.

In the following S270, using a predetermined conversion grammar stored in the ROM 20, the Japanese structure analyzed by the analysis process is converted into the English structure, and the individual words are converted into English. Execute the process. At S280, using a predetermined generation grammar stored in the ROM 20, an English word order is determined from the English structure converted by the conversion processing, an article specific to English is generated, and an input sentence is determined. Even if it does not exist, an output sentence is supplemented with an indispensable subject, and a process of generating an output sentence composed of an English word string is performed by processing a tense, an aspect, a person, a numerical feature, and the like. Then, in S290, using a predetermined morpheme generation rule stored in the ROM 20, the plural forms of the noun, the past of the verb,
A morpheme generation process for generating an inflected form such as a past participle and obtaining an output sentence is executed, and the translation process ends.

Next, details of the processing executed in S400 in FIG. 6 will be described with reference to the flowchart shown in FIG.

First, in S410, it is detected whether or not there is a headword in which a division flag is set in the input sentence, that is, whether an element headword exists. If it exists (S410:
Y), and proceed to S420. In S420, it is detected whether or not the element headword is a particle of a continuous modification. Information on whether or not the particle is a noun modification is stored in the dictionary information 46. If the particle is a noun modification particle (S420: Y), S43
Move to 0. In S430, it is determined whether or not the detected element headword includes a verb, and whether the right side of the element headword appearing at the rightmost is the end of a sentence, a period, or an auxiliary verb. To detect. That is, in these cases, there is a high probability that it is more appropriate to use the element headwords as verbs than to combine them as particles. The processing of S420 and S430 functions as the part of speech judgment means or the part of speech judgment program of the present invention. If it is not a sentence end, a period, or an auxiliary verb (S430: N), the process proceeds to S440.

If it is determined in S420 that the element headword is not a particle that has a continuous modification, the process proceeds to before-and-after judgment processing S450. In S450, it is detected whether or not the detected element headword is a particle for continuous modification. Information as to whether or not the particle is a continuous modification particle is stored in the dictionary information 46. If it is a consecutively modified particle (S450: Y), the process proceeds to S460. In S460, it is determined whether or not the detected element headword includes a verb, and whether the right side of the element headword appearing at the rightmost is an auxiliary verb or a verb "Sarusaru" is applicable. To detect. That is, in these cases, there is a high probability that it is more appropriate to use the element headwords as verbs than to combine them as particles. If the right side does not correspond to either an auxiliary verb or “subject” (S460: N), the headword resurrection processing S
Move to 440.

In headword resurrection processing S440, the plurality of element headwords are deleted leaving only the leftmost element, and the remaining headwords are replaced with original headwords stored in the dictionary information, and the process proceeds to S480. I do. That is, it is possible to perform the same processing as if the particle was closed in the morphological analysis processing S250 thereafter. Note that the process of S440 functions as a replacement unit or a replacement program of the present invention.

If the sentence is a sentence end, a punctuation mark, or an auxiliary verb in S430 (S430: Y), and if it is an auxiliary verb or "subscribe" in S450 (S460: Y), the flow shifts to S470. In S470, all the information on the element headwords is discarded, and the state is returned to a state in which there is no information on the particle registered by the user, and the process proceeds to S480.

If it is not a combined particle in S450 (S450: N), the flow shifts to S480.

In S480, the division flag is set to NIL, and the flow returns to S410.

In S410, if no part where the division flag is set is detected (S410: N), the processing of S400 is ended and the translation processing S260 is continued.

Here, the processing performed in S400 will be described with reference to example sentences.

For example, if the input sentence is "a problem relating to her", the following processing is performed.

If nothing is registered in the user dictionary 14, first, the translation processing S160 is performed by the user through the input unit 2.
Is executed. Morphological analysis processing S in translation processing S160
The morphological analysis result obtained in 250 is “She (noun) / ni (particle) / related (verb) / problem (noun)”. For this, analysis processing S260 and subsequent steps are executed. The text output by the translation process S160 at this time is “T
he problemwitch involves
her ". This is a normal embedded sentence (a sentence in which a verb modifies a noun). An attempt is made to register this by using the particle" = on ".

In S510 of the dictionary editing process S200, a headword "related to" and a part of speech "particle" are input. Since this has not been registered, the process moves to new creation in S530,
Other dictionary information such as “adjunct modification” is added. Since the part-of-speech is a particle (S560: Y), a tentative noun is added to the front and a morphological analysis is performed, and "ni (particle) / related (verb)" is performed.
Get. Then, the divided elements “ni” and “related” are registered in the user dictionary 14 as new headwords,
The information of the original headword "related to" and the division flag are added as dictionary information. FIG. 3 shows a state where “ni” and “related” are registered in this way.

The translation process S160 is executed again from the input unit 2 by the user.

At this time, the result of the morphological analysis obtained in the morphological analysis processing S250 is “She (noun) / (native particle) / related (verb) / problem (noun)”, which is the same as the previous time. However, the information on “ni” and “related” is given information that collectively becomes a particle by “related”. In processing S410 of analysis processing S260,
It is determined that the division flag is set to “ni” (S4).
10: Y) Then, it is determined that the particle is a particle which is a continuous modification (S420: Y). In the next processing of S430, since the right side of the rightmost element headword "relevant" is not the end of a sentence, a punctuation mark, or an auxiliary verb (S430: N), it is replaced with "relevant (particle)" in the headword resurrection processing S440. And
"Ni" and "Related" are discarded. After this, the analysis processing S
260 is continued and finally the output sentence "The probe
"lem on her" is generated. In this way, it is possible to obtain a much clearer English-like sentence than before registering the user dictionary.

Next, the processing when the input sentence is "She will be involved in this problem."

At this time, the result of the morphological analysis obtained in the morphological analysis processing S250 is “She (noun) / wa (particle) / this (adnominal) / problem (noun) / (particle) / related (verb) / (Auxiliary verb) ". The processing from S410 to S430 in S400 is the same as in the above example. In S430, since the right side of the rightmost element headword "relevant" is the auxiliary verb "da" (S430: Y), the processing shifts to S470 instead of the headword resurrection processing S440, and the information of the particle of "relevant" is discarded. Is done. Thereafter, analysis processing S26
0 is continued and finally the correct output statement "She wil
l inlove the problem. Is generated.

The present invention is not limited to the embodiment described in detail above, and various changes can be made without departing from the gist of the present invention.

For example, in the above-described embodiment, the embodiment in which particles are registered has been described. However, the present invention can be extended so that other functional words such as auxiliary verbs can be registered.

Also, the situation before and after the particle is strict,
By making it looser, it is possible to respond more flexibly. Further, by releasing the setting to the user, it is possible to provide a machine translation device having a user dictionary environment that is easier to use, and a translation result with higher accuracy can be obtained.

Further, in the above-described embodiment, the result of morphological analysis of the input original text is used as it is to determine whether or not an element headword exists. The determination may be made using a parse tree obtained by performing the parse.

In the above embodiment, the case where Japanese is used as the first language and English is used as the second language has been described, but it goes without saying that the present invention can be applied to other languages.

Further, in the above embodiment, a dictionary editing program, a morphological analysis program, a syntax analysis program, a specific part of speech determination program, a headword division program, a registration program, an element headword determination program, a part of speech The judgment program and the replacement program are RO
Although stored in advance in M, the present invention is not necessarily limited to this. For example, a program in which these programs are stored in a floppy disk, CD-ROM, or the like can be read and operated by a reading device. In addition, a program can be read from an external information processing device using a wired or wireless line and operated. In this case, the floppy disk or CD
-The ROM or the memory of the external information processing device storing the program constitutes the recording medium of the present invention.

[0079]

As is apparent from the above description,
The machine translation device according to claim 1 of the present invention, wherein the dictionary editing means determines whether a headword to be registered in the dictionary is a specific part of speech, and a specific part of speech part. When the determination unit determines that the word is the specific part of speech, the morphological analysis unit performs a morphological analysis of the headword and divides the headword into a plurality of elements. While registering each of the plurality of elements as a headword in the dictionary,
A registration unit that adds element headword information indicating that it is an element of the original headword before being divided by the headword division unit as dictionary registration information of each element. The user can also register a specific part of speech, such as a functional word, for which registration of the dictionary was prohibited by the user, and as a result, a natural translation result with high accuracy can be finally obtained. It is possible to reduce the work of later post-editing.

Further, the machine translation device according to claim 2 is
The specific part-of-speech determining means is configured to determine whether the headword to be registered is a particle, and the headword dividing means converts a temporary noun immediately before the headword to be registered. Since the morphological analysis is performed by connecting the particles, the particles can be accurately morphologically analyzed, and each element can be registered as a headword, so that there is no side effect in the translation process.

The machine translation device according to claim 3 is
The syntax analysis means includes: an element headword determination means for determining whether a headword having the element headword information is present in a morphological analysis result of the input original text by the morphological analysis means; A part-of-speech determining means for determining whether a plurality of headwords having the element headword information are used as the specific part-of-speech in the original text when the headword is determined to be present by the word determining means; When the part-of-speech determining unit determines that the part-of-speech is used as the specific part-of-speech, the replacing unit replaces a plurality of headwords having the element headword information with the original headword. It is possible to accurately determine the situation in which the function word is used, and replace the function word with the function word registered in the dictionary only when necessary. High accuracy it is possible to obtain a natural translation results, it is possible to reduce the time and effort of editing after the post-translational.

According to a computer-readable recording medium on which a machine translation processing program according to claim 4 is recorded, a specific word for determining whether or not a headword to be registered in the dictionary is a specific part of speech is determined. A part-of-speech determination program, and a headword division program that, when the specific part-of-speech determination program determines that the specific part of speech is determined, performs a morphological analysis of the headword by the morphological analysis program to divide the headword into a plurality of elements; Each of the plurality of elements divided by the headword division program is registered as the headword in the dictionary, and the element headword information indicating that the element is the element of the original headword before being divided by the headword division program. A dictionary editing program with a registration program to add as dictionary registration information for each element It can be stored from various media such as disk or CD-ROM to a recording medium suitable for machine translation apparatus. Then, by executing the program using this recording medium, the user can register a dictionary even for a specific part of speech such as a functional word, which is prohibited from being registered by the user in the conventional machine translator, As a result, a natural translation result with high accuracy can be finally obtained, and the post-translation post-editing effort can be reduced.

According to a computer-readable recording medium on which the machine translation processing program according to claim 5 is recorded, it is possible to specify whether or not the headword to be registered is a particle. A part-of-speech determination program and a headword division program adapted to perform a morphological analysis by connecting a temporary noun immediately before the headword to be registered in various media such as a floppy disk and a CD-ROM. Thus, it can be stored in a recording medium suitable for a machine translation device.

By executing a program using this recording medium, the particles can be accurately morphologically analyzed, and each element can be registered as a headword, so that there is no side effect in the translation process.

According to a computer-readable recording medium on which the machine translation processing program according to claim 6 is recorded, the element headword information is stored in the morphological analysis result of the input original sentence by the morphological analysis program. An element headword determination program for determining whether or not a headword having the same is present, and a plurality of headwords having the element headword information when the headword is determined to be present by the element headword determination program. A part-of-speech determination program that determines whether or not the part-of-speech is used as the specific part-of-speech in the original text, and when it is determined that the part-of-speech is used as the specific part of speech, the element headword information is And a replacement program for replacing a plurality of headwords having the same with the original headword. A floppy disk or CD-RO
It can be stored in a recording medium suitable for a machine translation device from various media such as M. Then, by executing the program using this recording medium, it is possible to accurately determine the situation where the function word is used, and to replace the function word with the function word registered in the dictionary only when necessary, As a result, a natural translation result with high accuracy can be finally obtained, and the post-translation post-editing effort can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an overall configuration of a machine translation device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram illustrating a data configuration of a basic dictionary.

FIG. 3 is an explanatory diagram illustrating a data configuration of a user dictionary.

FIG. 4 is a flowchart illustrating an entire operation of the machine translation apparatus.

FIG. 5 is a flowchart illustrating a flow of a dictionary editing process.

FIG. 6 is a flowchart illustrating a flow of a translation process.

FIG. 7 is a flowchart illustrating a flow of an element headword determination process.

[Explanation of symbols]

 4 External storage device 10 Control unit 18 CPU 20 ROM 22 RAM 40 Headword 42 Part of speech 44 Translated word 46 Dictionary information

Claims (6)

[Claims] 1. An entry word in a first language, a part of speech, a translation word in a second language corresponding to each entry word, and various dictionary registration information used for translation processing such as connection information necessary for morphological analysis processing. And a dictionary editing means for adding, correcting, deleting, etc., information such as the headwords of the first language to the dictionary, and inputting the original text of the first language registered in the dictionary. Morphological analysis means for performing a morphological analysis by searching for a headword, and syntactic analysis means for performing a syntactic analysis on a morphological analysis result by the morphological analysis means, and converts the original text of the first language into the second language In the machine translation apparatus, the dictionary editing means includes: a specific part-of-speech determining means for determining whether a headword to be registered in the dictionary is a specific part-of-speech; Is determined to be the specific part of speech, when the morphological analysis unit performs morphological analysis of the headword and divides the headword into a plurality of elements, a plurality of headwords divided by the headword division unit Registering each of the elements as headwords in the dictionary and adding element headword information indicating that the element is an element of the original headword before being divided by the headword division means as dictionary registration information of each element Machine translation device comprising: 2. The specific part-of-speech determining means is configured to determine whether or not the headword to be registered is a particle, and the headword dividing means is provided immediately before the headword to be registered. 2. The machine translation apparatus according to claim 1, wherein a morphological analysis is performed by connecting a tentative noun to the first term. 3. The element headword determining means for determining whether or not a headword having the element headword information exists in a morphological analysis result of the input original text by the morphological analysis means. When the element headword determination means determines that the headword is present, it is determined whether or not a plurality of headwords having the element headword information are used as the specific part of speech in the original text. A part-of-speech determining means for performing, and a replacing means for replacing a plurality of headwords having the element headword information with the original headword when the part-of-speech determining means determines that the specific headword is used as the specific part of speech. The machine translation device according to claim 1, further comprising: 4. Various dictionary registration information used for translation processing such as a headword of the first language, a part of speech, a translated word of the second language corresponding to each headword, and connection information necessary for morphological analysis processing. A dictionary editing program for adding, modifying, deleting, etc., information such as the headwords of the first language to the dictionary storing, and a headword registered in the dictionary for the input original text of the first language. A morphological analysis program that performs a morphological analysis by searching, and a syntactic analysis program that performs a syntactic analysis on a morphological analysis result by the morphological analysis program, wherein the original text of the first language is translated into the second language. In a computer-readable recording medium recording a machine translation processing program according to the present invention, the dictionary editing program may be configured such that the headword to be registered in the dictionary is A specific part-of-speech determination program for determining whether or not the part-of-speech is a specific part-of-speech; and when the specific part-of-speech determination program determines that the part of speech is the specific part of speech, a plurality of morphological analyzes of the headword are performed by the morphological analysis program. A headword division program that divides each of the plurality of elements divided by the headword division program into the dictionary as a headword, and an original headword before being divided by the headword division program. A computer-readable recording medium storing a machine translation processing program, comprising: a registration program for adding element headword information indicating that the element is a dictionary registration information of each element. 5. The specific part-of-speech determination program is configured to determine whether the headword to be registered is a particle, and the headword division program immediately before the headword to be registered. 5. A computer-readable recording medium recording a machine translation processing program according to claim 4, wherein morphological analysis is performed by connecting temporary nouns. 6. An element headword determination program for determining whether or not a headword having the element headword information exists in a morphological analysis result of the input original sentence by the morphological analysis program. When the element headword determination program determines that the headword exists, it is determined whether or not a plurality of headwords having the element headword information are used as the specific part of speech in the original text. A part-of-speech determination program to perform, and a replacement program that replaces a plurality of headwords having the element headword information with the original headword when the part-of-speech determination program determines that the headword is used as the specific part of speech. 6. A computer recording the machine translation processing program according to claim 4 or claim 5. A readable recording medium.