JPH11282845A - Machine translation apparatus and computer-readable recording medium recording machine translation processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a machine translation device applicable to a regular sentence even with no existence of a strict input sentence and obtaining a translated result sentence of high accuracy to reduce the load of a post-editing job by absorbing the slight variance of bath honorific and ordinary states of the regular sentence and the changes of auxiliary verbs etc. SOLUTION: It is decided whether a regular header sentence is included in an input sentence (S400) and also whether a regular sentence element exists at the head of the input sentence (S420). If the header sentence is included in the input sentence and the sentence element exists at the head of the input sentence, other sentence elements are detected (S430) and the extra adverbs and auxiliary verbs if included are given to an output sentence as the auxiliary information (S460). Thus, the translation processing is over. If no regular keyword is included, the ordinary translation processing is carried on.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a machine translation apparatus for translating an original sentence composed of a source language into a target language, and a computer-readable recording medium recording a machine translation processing program. The present invention relates to a technique for appropriately translating a fixed phrase as a fixed phrase.

[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. Specify the range of the vocabulary of the input sentence that the user wants to translate, and select and build an appropriate dictionary,
Provides more accurate translation results.

[0003] At the same time, registration of fixed phrases is also effective for providing high-level translation results. Fixed phrases are registered mainly for the following two purposes. First, regardless of the presence or absence of a user dictionary or specialized dictionary, and changes in translators,
This is a case where stable translation accuracy is obtained. Even in the manual translation process, it is often necessary to replace a specific phrase with a specific translation, instead of performing literal translation. A simple example is a greeting expression. This is very effective for those who want to always translate this sentence, from manual sentences to company documents.

[0004] The second is a case in which machine translation is impossible even though it is important and its appearance frequency is high.
If the fixed phrases are not registered, the fixed phrases, unique expressions, and idioms used by the user are literally translated, so that the translated sentences have no meaning or unnatural translations are generated. The greater these quantities, the more extensive the user has to do post-editing. User dictionaries and fixed phrases that users can freely edit are indispensable for obtaining natural translation results. In particular, fixed phrases are replaced with ideal translations, so it is difficult to find perfect translation accuracy. It is very effective in machine translation systems. Japanese Unexamined Patent Publication No. 7-152767 proposes the following technique as one technique of the registration method. That is, a fixed sentence is registered in a pattern file as a pair of an original sentence pattern expressed by a combination of a symbol representing an arbitrary phrase and a fixed phrase and a corresponding translated sentence pattern, and the pattern processing unit inputs the original sentence input from the input unit. Checks if matches the registered source pattern,
If there is a match, a word corresponding to the symbol of the original pattern is cut out. The translation control unit controls the translation unit to translate the extracted words one by one. The translated sentence synthesis unit receives a translated sentence pattern corresponding to the original sentence pattern from the pattern processing unit, and creates a final translated sentence by replacing the translation result of the cut-out phrase received via the translation control unit with the symbol of the translated sentence pattern. . The pattern file is such that the user can easily register and update.

[0005]

However, the fixed phrase may be replaced only when the whole sentence exactly matches in the input sentence, which is an advantage.
Conversely, it cannot respond to subtle fluctuations in expression. When there is no change in the structure of an input sentence such as a noun phrase, a wild card can be used with some restrictions. In addition, for the problem of swaying of notation due to differences in sending and character types, terms are being unified in manual sentences and the like by using various document revision support tools and the like. These actions are necessary for document processing, not just for machine translation. However, as a characteristic of Japanese, the ordinary body and the honorific body change every time a sentence is used. At the same time, when a sentence that is difficult to machine translation is registered as a fixed sentence, a pattern with an auxiliary verb such as a negative sentence must also be registered in some cases. There is a problem that registering a fixed phrase assuming all possible cases imposes a heavy load on a user. Also,
There is a problem that it is inefficient because the free space of the fixed form registration that is unnecessarily opened to the user is used.

The present invention has been made to solve the above-described problem. When a user registers a fixed phrase, the user performs a morphological analysis and registers the elements of the fixed phrase.
It is an object of the present invention to provide a machine translation device that can broaden a frame to which a fixed phrase can be applied, obtain a translation result of a target language with high accuracy, and reduce the load of post-editing by combining at the time of syntax analysis.

[0007]

In order to achieve this object, a machine translation apparatus according to a first aspect of the present invention comprises: an input means for inputting a source language sentence; and a source language input by the input means. Source language sentence storage means for storing sentences, headwords in the source language, part of speech of the headwords, word information including target language translations corresponding to the headwords, and connection information necessary for morphological analysis processing A translation dictionary storing grammatical information including: a fixed phrase sentence which is a sequence of arbitrary phrases in the source language; and a fixed translation sentence which is a sequence of translated words in the target language corresponding to the fixed phrase. A fixed sentence data storage unit storing a plurality of fixed sentence data consisting of pairs of the following, and using the translation dictionary and the fixed sentence data storage unit, the headword and the fixed headword for the source language sentence Morphological analysis that performs morphological analysis by searching A translation unit for translating the source language sentence into a target language sentence, comprising: a stage; and a syntax analysis unit that performs a syntax analysis based on a result of the morphological analysis of the source language sentence given by the morphological analysis unit. A display unit for displaying a target language sentence that is a final translation sentence, wherein the fixed sentence data storage unit adds and registers new fixed sentence data or is already registered It has a fixed phrase editing means for correcting and registering the fixed phrase data, wherein the fixed phrase editing means is a fixed phrase data comprising a pair of the newly added or corrected fixed entry and the fixed translation. When registering, the morphological analysis means performs a morphological analysis of the fixed headings, based on the results, the registered fixed headings to be registered as fixed headings and fixed sentence elements other than the fixed sentence elements Rule to divide into general elements A sentence dividing unit, when the fixed sentence dividing unit divides the fixed sentence into a fixed sentence element and a general element, a grouping unit that collects the fixed sentence element into a fixed sentence using only the fixed sentence element; And replacing means for replacing the standardized headline sentence compiled by the generalizing unit with the fixed-form translation corresponding to the standard headline sentence.

Therefore, according to this configuration, fixed phrases can be freely added and modified and registered in the fixed phrase data storage means, and the data stored in the fixed phrase data storage means can be used to machine translate the source language into the target language. Even if there is a difference in the honorific or ordinary body expressed in the tonality or the dana tongue, the difference in inflection, or the case where an auxiliary verb is added, It can be recognized and translated into a predetermined fixed translation, and a high quality translation can be obtained with the fixed translation as it is or with only a slight correction.

Further, the machine translation device according to claim 2 is
In addition to the configuration of the machine translation device according to claim 1, a translation generation auxiliary information storage unit that adds the general element as the translation generation auxiliary information to the fixed translation, and a translation associated with the fixed translation displayed on the display unit. And a translation generation auxiliary information display control means for displaying the translation generation auxiliary information on the display means.

Therefore, according to this configuration, when the auxiliary sentence, which is a general element, is added to the fixed phrase and the meaning is slightly different, the fixed verb is temporarily converted to a fixed translation, and the fixed translation is displayed by the translation generation auxiliary information displayed at the same time. The translation can be made a high quality translation with only minimal modification.

A computer-readable recording medium on which the machine translation processing program according to claim 3 is recorded stores a machine translation processing program for machine translation processing for translating a source language sentence into a target language sentence. Recording medium, a computer, an input procedure for inputting a source language sentence, a source language sentence storage procedure for storing the source language sentence input by the input procedure, a headword in the source language, A translation dictionary storing the part of speech of the headword, word information including a translation of the target language corresponding to the headword, and grammar information including connection information required for morphological analysis processing, A fixed phrase data recording unit that stores a plurality of fixed phrase data, each of which is a pair of a fixed phrase sentence that is a sequence of arbitrary phrases and a fixed phrase translation that is a sequence of translated phrases of the target language corresponding to the fixed phrase. Using A morphological analysis procedure for performing a morphological analysis by searching the headword and the fixed-form headword for the source language sentence, and a result of a morphological analysis of the source language sentence given by the morphological analysis procedure And performing a translation procedure for translating the source language sentence into a target language sentence and a display procedure for displaying the target language sentence that is the final translated sentence. A machine translation processing program, wherein the procedure for storing the fixed form data is a fixed form editing procedure for adding and registering new fixed form data, or correcting and registering already registered fixed form data. Has a fixed sentence editing procedure, when registering the fixed sentence data consisting of a pair of the newly added or modified fixed entry and fixed form translation, the fixed form data by the morphological analysis procedure Finding Morphological analysis, and based on the result, a fixed sentence element as a headword, and a fixed sentence dividing procedure for dividing the fixed fixed sentence into general elements other than the fixed sentence element; When the fixed entry is divided into the fixed sentence element and the general element according to the procedure of the above, the setting procedure of putting together the fixed entry by using only the fixed element and the fixed form put together by the setting procedure And replacing the entry sentence with the fixed translation corresponding to the fixed entry sentence.

[0012] Therefore, by using this recording medium, fixed phrases can be freely added and corrected in the computer and registered in the fixed phrase data storage means, and the data stored in the fixed phrase data storage means can be used in the source language. When translating into a target language, if there is a difference in the honorific or ordinary body that can be expressed in the tonality or the tonality, the difference in the use, or when an auxiliary verb is added Even if it is, it can be recognized as a fixed form sentence and can be translated into a predetermined fixed form translation, and a procedure for obtaining a high-quality translation can be executed with the fixed form translation as it is or with a slight modification.

Further, a computer readable recording medium on which the machine translation processing program according to claim 4 of the present invention is recorded is a computer readable recording medium on which the machine translation processing program according to claim 3 is recorded. In addition to the above, the translation element generating auxiliary information storage procedure of adding the general element to the fixed form translation as the translated sentence generation auxiliary information, and displaying the translated sentence generation auxiliary information in association with the fixed form translation displayed on the display means. A step of controlling the display of the translated sentence generation auxiliary information to be displayed by the means.

Therefore, when the recording medium is used, an auxiliary verb or the like, which is a general element, is added to the fixed sentence to the computer, and when the meaning is slightly different, the fixed sentence is temporarily converted into a fixed translated sentence, and the translated sentence generation assist displayed at the same time is displayed. With the information, it is possible to execute a procedure for converting a fixed-form translation into a high-quality translation with only minimal modification.

[0015]

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, a case where translation from Japanese (source language) to English (target language) will be described as an example.

First, FIG. 1 is a block diagram showing a schematic configuration of a machine translation apparatus according to the present embodiment. As shown in FIG. 1, a machine translation apparatus according to the present embodiment includes an input unit 2 that is an input unit including a keyboard or the like for inputting a Japanese original to be translated, various operation instructions, and the like. An external storage device 4 composed of a hard disk, an optical disk, or the like, and a C for displaying an input Japanese original and an English translation as a translation result.
A display unit 6 which is a display means composed of an RT or the like; a display control unit 7 which controls data to be displayed on the display unit 6; and a laser printer for printing and outputting English translations and the like as translation results. The translation unit performs translation processing based on data input from the input unit 2 or the external storage device 4, and displays the translation result on the display unit 6,
And a control unit 10 for performing control such as output from the controller.

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 fixed phrase editing key, an output sentence editing key, an end key, and a print execution key, cursor keys used for editing input and output sentences, and for operating the screen displayed on the display unit 6 , An insert key, a delete key, a decision key, a next candidate selection key, and the like.

Further, the external storage device 4 includes a translation dictionary 12 and a fixed phrase data storage unit 16 used for translation processing. This translation dictionary 12 constitutes the translation dictionary of the present invention. Also, the fixed phrase data storage unit 16 stores
This constitutes the standard sentence data storage means of the present invention.

The translation dictionary 1 used by the machine translation device
The contents of item 2 include an essential basic dictionary held in advance by the machine translation apparatus, and optionally, a specialized dictionary edited for each specialized field and a user dictionary independently constructed by the user are externally stored. By storing it in the device 4 and performing translation processing with reference to this, the translation accuracy can be improved.

FIG. 2 shows an example of the data structure of the translation dictionary 12 described above. Each dictionary registration information stores 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. It consists of fields that

FIG. 3 shows the fixed form data storage section 16 described above.
1 shows an example of the data configuration. Each fixed sentence data storage unit 16 stores a fixed entry sentence 40, a fixed sentence element 42, a translated word,
It is composed of fields for storing dictionary information 44 such as part of speech, fixed sentence element ID 46, and fixed translation 48.

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 pass line 26 connecting these components.

In the control unit 10, the ROM 2
Various kinds of processing are executed by the program stored in “0”. For example, a translation / 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 fixed form for constructing fixed sentence data used at the time of translation. A sentence editing process is performed.

The ROM 20 stores various kinds of information used in a translation process described later. More specifically, a control program 202 for controlling the entire machine translation program, a translation means for controlling the entire translation process A translation program 204, a morphological analysis program 206 as a morphological analysis means for performing morphological analysis, a syntactic analysis program 208 as a syntactic analysis means for performing syntactic analysis, and a conversion program 210 as a conversion means for converting Japanese into English. A generating program 212 which is a generating means for generating a translated sentence after conversion, a fixed sentence editing program 214 which is a fixed sentence editing means for editing a fixed sentence, and a fixed sentence dividing program 216 which is a fixed sentence dividing means for dividing the fixed sentence. , A fixed phrase data storage unit 16 and a fixed phrase data management unit Such as fixed text element recording program 218 is a unit are stored.

In the RAM 22, a work area 22 for performing work such as translation and temporarily retrieving and storing data.
2. A source language sentence storage area 224, which is a source language sentence storage means for recording a Japanese source sentence input from the input unit, and a translated sentence storage area 226 in which an English translation translated and edited by translation and editing processing is recorded. Is secured.

Next, the procedure of 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, in step 110 (hereinafter abbreviated as S110; other steps are the same), the user operates the character keys of the input unit 2 by the user. It is operated and waits for the input of the Japanese original 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 input sentence is displayed in a predetermined area of the display unit 6 (S120).

Next, in S130, it is determined whether or not the input sentence editing key provided in the input unit 2 has been operated. If not (S130: NO), the process proceeds to S150. If the input sentence edit key is operated (S13
0: YES), proceeds to S140, and executes 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.
In S150, 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 (S1).
50: NO), the process proceeds to S190, and if the translation execution key is operated (S150: YES), the process proceeds to S160 to convert the Japanese input sentence into an English translated sentence (hereinafter referred to as an output sentence). In step S170, the output sentence obtained by the translation process is stored in the RAM 22 as a pair with the input sentence. In step S180, the input sentence is stored in a predetermined area provided on the display unit 6. And the output statement.

The procedure of the translation process may be such that the entire input sentence is executed collectively, or a range is specified for each sentence, and only the sentences in the specified range may be executed.

In S190, it is determined whether or not the fixed phrase editing key provided on the input unit 2 has been operated, and if not (S190: N), the process proceeds to S210. If the fixed phrase editing key is operated (S190:
Y), and proceeds to S200 to execute a fixed phrase editing process.
When the fixed phrase editing process is executed, the user can correct or delete the content of the information registered in the fixed phrase data storage unit 16 or register a new required fixed phrase. Become. The fixed phrase editing program 214 for executing the procedure of S200 corresponds to a fixed phrase editing unit 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: NO), the process proceeds to S230. If the output sentence edit key is operated (S210: Y
ES), and proceeds to S220 to execute an output sentence editing process. When the output sentence editing process is executed, the user executes the translation process in S160 to display an output sentence (English translation) displayed in a predetermined area of the display unit 6.
Can be directly corrected using the cursor key, the delete key, the insert key, the character key, and the like provided in the input unit 2 as in the input sentence editing process in S140.

In S230, it is determined whether or not the end key provided on the input unit 2 has been operated. If it has not been operated (S230: NO), 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: YES), the process is terminated.

As described above, in the translation / editing process, after the Japanese original is input, the input sentence editing process, the translation process, the fixed phrase editing process, and the input process 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. After executing the input sentence editing process and re-editing the Japanese original sentence that is the input sentence, or operating the fixed phrase edit key to edit the fixed sentence data, operating the translation execution key again An output sentence editing process is executed by translating or operating an output sentence edit key, and then an English translated sentence, which is an output sentence, is directly edited and then 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 the translated text storage area 226 of the RAM 22 as text data. Or the external storage device 4, and then operate the end key to
The process ends.

Next, in FIG. 5, the process is executed in S200 in the flowchart shown in FIG. 4, and the user corrects or deletes the contents of the fixed phrase data storage unit 16, or registers a new required fixed phrase. The fixed phrase editing process will be described with reference to the flowchart.

First, in S510, it is determined whether or not the user has input a standard heading to be edited. If it has been input (S510: YES), the flow shifts to S520. S
At 520, it is determined whether or not the input standard sentence is already registered in the standard sentence data storage unit 16. If it has already been registered (S520: YES), the process proceeds to S540. In S540, the standard entry or the standard translation in the storage of the standard sentence data 16 is re-edited, and the process proceeds to S550. If not registered yet (S520: N
O), and proceed to S530. In S530, the routine edits the standard entry sentence and the standard translated sentence, and then proceeds to S550. S5
At 50, it is determined whether or not to replace the edited standard sentence with the standard translated sentence only when there is a perfect match. If the user wants replacement only when there is an exact match, select YES here. If only at the time of perfect match (S550: YE
S), and proceeds to S590. In S590, connection information that allows connection only when a punctuation mark, a colon, a semicolon, a line feed code, or the like, which is a sentence delimiter, comes to the fixed entry sentence, and the process proceeds to S610.

If a replacement other than an exact match is to be replaced,
(S550: NO), and proceeds to S560. In S560, it is determined whether the fixed entry is newly registered in S530 or re-edited in S540. If the fixed entry is not re-edited (S560: N
O), and proceed to S590. If the fixed headings are not rewritten, there is no need to perform morphological analysis again.
This is because it is sufficient to register as it is. If the standard entry is new or has been re-edited, the process proceeds to S570. S5
At 70, the morphological analysis processing S250 described in a flowchart described later is performed using the translation dictionary 12, and S580
Move to. In S580, a unique fixed sentence element ID is assigned to the fixed sentence element obtained from the result of the morphological analysis, together with dictionary information such as part of speech and translation, and the process proceeds to S600. The reason why the dictionary information such as the part of speech and the translated word is added is to continue the normal translation process when the sentence is not put together as a fixed phrase during the analysis process. Note that S560, S570, S58
The procedure 0 is executed by the fixed sentence dividing program 216 which is the fixed sentence dividing means of the present invention. In S600,
At the end of the fixed phrase element, add punctuation marks, colons, semicolons, line feed codes, etc., which are sentence breaks, and information that the polite auxiliary words "is" and "mas" can be connected.
The process moves to S610.

In S610, the fixed phrase sentence, the fixed phrase element, the dictionary information, the fixed phrase element ID, and the fixed translation obtained in each of the above steps are stored in each field 40 of the fixed phrase data storage unit 16,
42, 44, 46, and 48 are registered, and the process proceeds to S510. These pieces of connection information are for preventing a search in the middle of the input sentence during the execution of the morphological analysis.
The procedures of S590, S600, and S610 correspond to the fixed phrase element recording program 2 which is the fixed phrase element recording means of the present invention.
18.

In S510, when the headword and the part of speech have not been input (S510: NO), 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, using the predetermined judgment rule of the morphological analysis program 206, the connection information in the translation dictionary 12, and the fixed sentence data storage unit 16, the fixed sentence or the words constituting the input sentence are used. Translation dictionary 1
2 and the fixed phrase data storage unit 16 are searched. In the case where the fixed entry was found, the fixed translation was added to the morphological analysis result.
A morphological analysis process is performed for adding search information such as a translation word and a part of speech to the morphological analysis result. If a user dictionary and a specialized dictionary are used, priority is given to the user dictionary, the specialized dictionary, and the basic dictionary in this order. The procedure of S250 is executed by the morphological analysis program 206 as the morphological analysis means of the present invention.

In step S260, the result of the morphological analysis process is received. In step S260, the translation dictionary 1 in the external storage device 4 is transmitted by the syntax analysis program 208 as a syntax analysis unit.
Using a predetermined analysis grammar stored in the second grammar section, an analysis procedure for analyzing a Japanese structure is executed. When this is completed, the process moves to S400. In S400, it is detected whether or not the input sentence is a fixed entry sentence. If it is a fixed entry (S400: YES), S410
Move to. In S410, the input sentence is replaced with the fixed-form translation, and the translation process of the input sentence ends.

If it is not a fixed entry (S400: N
O), proceed to S420. In S420, it is detected whether or not the head entry of the morphological analysis result is the first fixed sentence element of the fixed head sentence. If it is a fixed phrase element (S
420: YES), and proceeds to S430.

At S430, it is detected whether the arrangement of the fixed phrase element IDs is appropriate. At this time, for example, an auxiliary verb other than the fixed phrase element ID may be included immediately after the declinable word. In addition, an auxiliary particle may appear immediately after the case particle or instead of the case particle. In this case, for example, even if an original case particle having an ID is replaced with a sub-particle, there is no significant change in meaning. Therefore, when a sub-particle without an ID appears, the case having the original ID is used. It is determined that the ID can be replaced with the particle, and it is considered that the ID is continuous for this part.
D is detected. However, if this part is a verb or a noun, it is expected that the meaning will fluctuate greatly by replacement, and the connection relationship will also be different. (S43
0: NO) The process proceeds to another analysis process (S265), which is a normal analysis process, without performing the process as a fixed phrase.
When the fixed phrase element IDs are continuous or when it is determined that they are continuous, the process proceeds to S440. In step S440, the process goes to step S450, excluding elements such as auxiliary verbs and sub-particles that are not fixed phrase elements. In S450, the original fixed heading sentence that has been put together is replaced with the corresponding fixed translation, and the process proceeds to S460. In step S460, auxiliary verbs, sub-particles, etc., which are not fixed phrase elements, are changed to information that can be easily understood by the user, added to the fixed translation as auxiliary information, and the input sentence translation process ends. The procedure of S460 is executed by the translated sentence generation auxiliary program 219 as the translated sentence generation auxiliary means of the present invention. The procedure of S440 is executed by a grouping program as a grouping unit.
The procedure of S450 is executed by a replacement program that is a replacement unit. These grouping program and replacement program are a part of the morphological analysis program 206.

When the input sentence is not a fixed entry sentence (S4
00: NO) or when the head entry of the morphological analysis result is not the first fixed sentence element of the fixed head sentence (S:
420: NO) Also, if the fixed phrase element IDs are not reliably consecutive (S430: NO), S is used to perform other translation processing as in the case of a normal input sentence not using fixed phrases.
The process proceeds to H.265, and normal analysis processing is continued, and S270 is performed.
Move to.

At S 270, the conversion program 210, which is the conversion means, uses the predetermined conversion grammar stored in the grammar information of the translation dictionary 12 of the external storage device 4 to convert the Japanese language analyzed by the analysis process. Perform the conversion procedure to convert the structure to English structure and replace individual words with English. In S280, using the predetermined generation grammar stored in the ROM 20 by the generation program 212 as the generation means, the English word order is determined from the English structure converted by the conversion processing,
Generates articles unique to English, supplements essential subjects in output sentences even if they do not exist in the input sentence, processes tense, aspect, person and numerical features, etc., and outputs an English word string A procedure of a generation process for generating a sentence is executed. Then, in S290, the generation program 212 of the ROM 20
Using a predetermined morpheme generation rule stored in the above, a plural form of a noun, a past verb, and an inflected form such as a past participle are generated for the English word string generated by the generation process of S280. Then, a morpheme generation process for obtaining an output sentence is executed, and the translation process ends.

Here, the processing performed in the above-described machine translation processing will be described with reference to example sentences.

For example, as shown in FIG. 3, as an example of always using the same expression without using the parts to be used, "Dear sir, I am grateful to your company for being more and more prosperous."
An example of registering is described.

If the Japanese sentence is not registered as a fixed head sentence, the translation is performed literally in accordance with the principle of machine translation, so that "It is glad that"
Dear, your company are mo
re well and prospering. "It's a very unnatural translation for the English sentence. This is in line with English customs," Dear Sir
Suppose that it is registered as fixed phrase data so that "s," is obtained.

In step S510 of the standard sentence editing process S200, "Dear sir, I am grateful to you for your continued pleasure" is input as a standard heading. This is S52
If it is not registered at 0 (S520: NO), the process proceeds to S530. Here, the headword translation “Dear Sirs,”
Is input, and the flow shifts to S550. In this case, there is no such a phrase as “to say”, so a fixed heading sentence that matches the whole can be used as it is (S550: YE
S), the process proceeds directly to S590 and S610. S61
0, the fixed sentence field 4 of the fixed sentence data storage unit 16
0 is registered in the fixed translation field 48, and the process returns to S510. If the next fixed heading is not input (S51)
0: NO), and ends the fixed phrase editing process.

In this state, a translation process S160 is executed by the user from the input unit 2 for the input sentence "Dear sir, I am grateful to you for your continued pleasure."

First, in S250, a morphological analysis process is performed to search for a typical headline sentence, and the morphological analysis result is obtained, "Dear Sirs, I am grateful that your company is becoming more and more prosperous <typical headline> /. <Phrase >>. , S260, and S400. In S400, since the fixed entry is searched (S400: YES), the process proceeds to S410. S410
Then, the input sentence is replaced with the fixed translation and the translation process for the input sentence ends.

Next, as another example sentence, as shown in FIG. 3, a case will be described in which "thread slack occurs at the time of square sewing" including a part to be utilized is registered. This "occurs" as a representation of the input sentence in the text you want to translate
Or, an expression such as "will occur" can be used. In addition, a negative sentence “does not occur” or a “slack of thread also occurs” accompanied by an auxiliary particle can be assumed as an input sentence. This sentence is frequently used in troubleshooting of manual sentences of industrial sewing machines. Even if the user registers “corner stitching” in the user dictionary, if this is translated by the machine translation system as it is. "During the corner
stitching, looseness of
the thread occurs. "Is the desired English sentence" Imper upper upper
d tension at the corner
s. Cannot be obtained. This is an expression that should be registered as a cliche.

In step S510 of the standard sentence editing process S200, "slack of thread occurs at the time of corner stitching" is input as a standard heading sentence. This is assumed to be unregistered, and S530
Move to. Here, the headword translation "Imperper u
paper thread tension at th
e corners. Is input and the process proceeds to S550. In this case, since there is an expression in which an auxiliary verb or an auxiliary particle appears, S550 is NO and the process proceeds to S560. In S560, since the current registration is new, S5
60: YES, proceeding to S570. In S570, morphological analysis is performed, and as a result, the fixed phrase elements “square stitch (noun)” “time (noun)” “ni (case particle)” “thread (noun)” “no (case particle)” “sag (noun) ), “Ga (case particle)” and “occurs (verb)”. The reason for dividing into elements in this way is that the auxiliary particle may appear immediately after each case particle or in place of the case particle, or that the sentence end is a verb and the auxiliary verb may appear later. . In the next S580, a fixed phrase element ID is assigned to each, and S590, S
At step 610, the standard sentence field 40, the standard sentence element field 42, the standard sentence element ID field 46, and the standard translation field 48 of the standard sentence data storage unit 16 are registered.
Return to 510. If the next fixed entry is not input (S510: NO), the fixed sentence editing process ends.

In this state, a translation process S160 is executed by the user from the input unit 2 for the input sentence "a slack of the thread also occurs at the time of square sewing".

First, in S250, the morphological analysis processing does not search for a fixed-form head sentence, but instead searches for a fixed-form sentence element. The result of the morphological analysis is "square stitch <noun / fixed-statement element> / time <noun / fixed-statement element> / To <case particle / formal sentence element> / thread <noun / formal sentence element> / of <case particle / formal sentence element> / sag
Noun / fixed sentence element> / also <adjunct particle> / occurs <verb /
Fixed phrase element> / Masu <Auxiliary verb> /. <Period> ”and S
260, proceed to S400. In S400, since the standard entry has not been searched (S400: NO), S42
Move to 0. In S420, "square stitching" is the first fixed sentence element of the fixed entry (S420: YES), and in S430, it is detected from the fixed sentence element ID that all the elements behind it are aligned. In this case, the standard sentence elements are “square stitching”, “time”, “ni”, “thread”, “no”, “sag”, and “occur”. , Words are attached after the final ID. In S430, "mo <sub-particle>" and "mas <auxiliary verb" which are elements other than the fixed phrase element are extracted, but "ga" having an ID can be replaced with "mo",
Since "masu" is semantically equivalent even if attached,
D is considered OK until the end of the sentence (S430: YES), and S
The original fixed entry is compiled again at 440, and is replaced with the fixed translation corresponding to the original fixed entry at S450. At this time, in S460, the auxiliary particle "mo"
Is added as supplementary information that exists in the input sentence,
The translation process ends for this input sentence. The user displays the output sentence “Imper upper th on the display unit 6.
read tension at the corner
s. (Also, impr)
upper upper thread tension
at the corners. At this time, the auxiliary verb "Masu" is unnecessary information for an English sentence, and is not intentionally presented to the user as auxiliary information.

If the input sentence is "... does not occur", the output sentence is provided with information that is a negative sentence such as (none) and is presented to the user. In some cases, the user effectively re-edits the fixed phrase while utilizing the information, and in some cases, changes the setting so that the fixed phrase can be used only in the case of overall matching in S550.
It is possible to improve the accuracy of the subsequent translation processing and increase the efficiency of post-editing.

Although the present invention has been described with reference to one embodiment, the present invention is not limited to the above-described embodiment, and various modifications can be made. For example,
In this case, the values of the auxiliary particles and auxiliary verbs that can be used as auxiliary information are determined by the internal processing in the procedure for summarizing S440 in the system and the procedure for adding auxiliary information in S460, but can be set in the menu of the user interface. You may do so. For example, if the headword is in the verb imperative form on the screen of the CRT, and "Tetsu" appears in the input sentence, it is possible to specify with the mouse operation whether or not to output as auxiliary information. Things. Boilerplate sentences are a rescue for machine translation systems where perfect translation is not possible, but at the same time it is inconvenient if there is too much. If you give users fine control over this,
The freedom of the user is expanded, and the translation accuracy can be improved and post-editing can be greatly reduced.

Further, the machine translation apparatus of the present embodiment
OM 20 has control program 202 and translation program 20
4. Morphological analysis program 206, syntax analysis program 208, conversion program 210, generation program 21
2, a fixed sentence editing program 214, a fixed sentence division program 216, a fixed sentence element recording program 218, and the like. In addition to these, the dictionary editing program, the part-of-speech determination program, the headword division program, the determination program, the before / after determination program, and the headword resurrection program are stored in the ROM in advance. It is not limited to this. For example, these programs are stored on a floppy disk or CD-
What is stored in a ROM or the like can be read by a reading device and installed directly on a hard disk or the like to operate. 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 above-mentioned floppy disk, CD-ROM, hard disk which is an external information processing device, and a memory storing the program constitute a recording medium of the present invention.

Further, even if the programs are not clearly distinguished and stored as in the present embodiment, the functions of the machine translation device and the machine translation processing program described in the claims may be achieved. .

In the present embodiment, the source language has been described using Japanese as an example, but the source language is not limited to Japanese. The type of language does not matter as long as the width of the fixed expression can be absorbed within the allowable range of the grammar of each language, processed as a fixed sentence, and displayed in a form that is easy for the user to understand.

The target language is not limited to English, and can be applied to various languages as long as the grammatical structure is not so different that application of the present invention is difficult.

The input means is exemplified by a keyboard.
It goes without saying that any type of recording medium, such as infrared rays, voice, and various recording media, whether wired or wireless, may be input to the computer.

Further, the display means is not limited to the CRT, but may be any display method such as sound, not to mention a liquid crystal display.

[0067]

As is apparent from the above description, according to the machine translation device of the first aspect, fixed form sentences can be freely added and corrected and registered in the fixed form data storage means. When the source language is machine-translated into the target language based on the data stored in the storage means, there are differences between the polite and ordinary forms expressed in the tonality, the tonality, etc., and the difference in utilization. Or even if an auxiliary verb is added, it can be recognized as a fixed phrase, and can be translated into a predetermined fixed translation. There is an effect that a translation can be obtained. Therefore, it is not recognized as a fixed sentence due to a slight difference, and the original machine translation is not performed, and the quality of the translation is not reduced. There is an effect that a translation result can be obtained.

Further, if it can be processed as a standard sentence, there is an effect that the speed of the translation process can be increased.

According to the machine translation device of the second aspect, an auxiliary verb or the like which is a general element is added to a fixed phrase,
Even if the meaning is slightly different, it is recognized as a fixed phrase, and once converted to a fixed phrase, and at the same time, the added translation information generation auxiliary information is displayed on the display device. A high-quality translation can be achieved by slightly modifying the fixed-form translation with reference to the translation generation assistance information without retranslating as in the case where the translation is performed. Therefore, there is an effect that the translation is not lost and a high-quality translated sentence can be efficiently obtained in a short time.

According to the computer-readable recording medium on which the machine translation processing program according to the third aspect is recorded, the procedure for freely adding and correcting fixed phrases in the computer and registering them in the fixed phrase data storage means is executed. When the source language is machine-translated into the target language by using the data stored in the fixed-form sentence data storage means, the polite / ordinary body is expressed in the form of a key or a key. Even if there is a difference in usage, a difference in usage, or an auxiliary verb is added, it can be recognized as a fixed phrase and a procedure for translating this into a predetermined fixed translation can be executed. There is an effect that a high-quality translation can be obtained by simply or slightly modifying. For this reason, it is not recognized as a fixed sentence due to a slight difference, and the original machine translation is not performed, so that the quality of the translation does not deteriorate and the translation is not redone. There is an effect that quality translation results can be obtained.

Further, if it can be processed as a standard sentence, there is an effect that the speed of the translation process can be increased.

According to the computer-readable recording medium on which the machine translation processing program according to claim 4 is recorded, an auxiliary verb or the like which is a general element is added to a fixed sentence, and the meaning is slightly different. Even if there is, a procedure for recognizing this as a fixed phrase, temporarily converting it to a fixed phrase, and displaying the translated sentence generation auxiliary information on the display device when the entire sentence is subjected to normal machine translation by the translated sentence generation auxiliary information displayed at the same time Can be executed, and there is an effect that a high quality translation can be obtained by only slightly correcting the standard translation with reference to this information without retranslation. Therefore, there is an effect that a high-quality translation can be efficiently obtained in a short time.

[Brief description of the drawings]

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

FIG. 2 is an explanatory diagram of a configuration of dictionary data according to the present embodiment.

FIG. 3 is an explanatory diagram of a data configuration of a fixed form data storage unit according to the present embodiment.

FIG. 4 is a flowchart illustrating a processing flow of the machine translation device according to the present embodiment.

FIG. 5 is a flowchart showing the flow of a fixed sentence editing process of the machine translation device of the present embodiment.

FIG. 6 is a flowchart illustrating a flow of a translation process of the machine translation device according to the present embodiment.

[Explanation of symbols]

 2 input unit 4 external storage device 6 display unit 7 display control unit 10 control unit 12 translation dictionary 16 fixed sentence data storage unit 18 CPU 20 ROM 22 RAM 204 translation program 206 morphological analysis program 208 syntax analysis program 210 conversion program 212 generation program 214 Fixed sentence editing program 216 Fixed sentence division program 218 Fixed sentence element recording program 219 Translation sentence generation auxiliary program 222 Work area 224 Source language sentence storage area 226 Translated sentence storage area

Claims (4)

[Claims] An input unit for inputting a source language sentence; a source language sentence storage unit for storing the source language sentence input by the input unit; a headword in the source language; a part of speech of the headword; A translation dictionary storing word information including a translation of the target language corresponding to the headword and grammatical information including connection information necessary for morphological analysis processing; and a fixed form which is an array of arbitrary phrases in the source language. A fixed sentence data storage unit storing a plurality of fixed sentence data each comprising a pair of an entry sentence and a fixed translation that is a sequence of translated phrases of the target language corresponding to the fixed entry, the translation dictionary and the fixed sentence A morphological analysis unit that performs a morphological analysis by searching the headword and the fixed headword for the source language sentence using a data storage unit;
A translation unit that performs a syntax analysis based on a result of the morphological analysis of the source language sentence given by the morphological analysis unit, and a translation unit that translates the source language sentence into a target language sentence; Display means for displaying the target language sentence, wherein the fixed form data storage means adds and registers new fixed form data, or the fixed form data already registered A fixed phrase editing means for correcting and registering the fixed phrase sentence data, wherein the fixed phrase editing means is for registering fixed phrase data comprising a pair of the newly added or corrected fixed phrase heading and the fixed phrase translation. Then, the morphological analysis unit performs a morphological analysis of the fixed-form head sentence, and divides the registered fixed-statement sentence into a fixed-form sentence element as a headword and a general element other than the fixed-form sentence element based on the result. Fixed phrase Means, when the fixed-form sentence is divided into a fixed-form sentence element and a general element by the fixed-form sentence dividing means, a consolidating means for consolidating the fixed-form sentence element into a fixed-form sentence only, and the consolidating means And a replacing means for replacing the standardized headline sentence compiled by the above with the standardized translation corresponding to the standard headline. 2. A translated sentence generation auxiliary information storage unit for adding the general element as the translated sentence generation auxiliary information to the fixed translation, and in association with the fixed translation displayed on the display unit,
2. The machine translation apparatus according to claim 1, further comprising a translation generation auxiliary information display control unit that displays the translation generation auxiliary information on the display unit. 3. A recording medium on which a machine translation processing program for machine translation processing for translating a source language sentence into a target language sentence is recorded. A source language sentence storage procedure for storing the source language sentence input by the input procedure; a headword in Japanese; a part of speech of the headword; and a word including a target language translation corresponding to the headword. A translation dictionary storing information and grammatical information including connection information necessary for morphological analysis processing, a fixed-form heading that is a sequence of arbitrary words and phrases in the source language, and corresponding to the fixed-form heading. Using the fixed phrase data recording means storing a plurality of fixed phrase data, which are pairs of fixed translations, which are a sequence of translated phrases of the target language, the headword and the fixed headword for the source language sentence Perform morphological analysis by searching A morpheme analysis procedure, and a syntax analysis procedure for performing a syntax analysis based on a result of the morphological analysis of the source language sentence given by the morphological analysis procedure, wherein the source language sentence is converted into a target language sentence. A machine translation processing program for executing a translation procedure for translating and a display procedure for displaying the target language sentence that is the final translation, wherein the fixed form data storage procedure includes adding new fixed form data. And has the procedure of editing the fixed form data to correct the already registered fixed form data and register the fixed form data, wherein the fixed form editing procedure is newly added or corrected. When registering the fixed phrase data composed of pairs of the sentence and the fixed translation, the morphological analysis of the fixed phrase is performed by the morphological analysis procedure, and the registered fixed phrase is registered based on the result. Fixed form as a word Elements, a procedure for dividing a fixed sentence into general elements other than fixed-form elements, and when the fixed entry is divided into fixed-form elements and general elements by the fixed-segment division procedure, only the fixed elements are included. A machine that executes a grouping procedure of summarizing into a fixed form heading using the above, and a replacement procedure of replacing the fixed form heading assembled by the grouping procedure with the fixed form translation corresponding to the fixed heading sentence. A computer-readable recording medium on which a translation processing program is recorded. 4. A procedure for storing the translated sentence generation auxiliary information for adding the general element as the translated sentence generation auxiliary information to the fixed translation, and in association with the fixed translation displayed on the display means,
The computer-readable recording medium according to claim 3, wherein the computer-readable recording medium executes a procedure for controlling display of the translated sentence generation auxiliary information in which the display unit displays the translated sentence generation auxiliary information.