JP2001060195A - Machine translation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To simplify operations, to improve hadleability and to provide the translated document of high quality by automatically setting a translation environment supply means for supplying a translation environment for translation corresponding to the correspondence relation of a translation origin document and a field instruction means or the like. SOLUTION: A communication control means 103 controls communications with an external network 108 and a perusal means 101, and a translation control means 104 is provided with a function for performing direct control to a machine translation means 105 such as which document is to be translated and which dictionary is to be used. That is, the setting of the translation environment to the machine translation means 105 is directly executed. The machine translation means 105 translates a supplied document to another language under the control of the translation control means 104. In this case, the field instruction means for indicating special fields to which the individual translation origin documents belong are correlated by a hierarchical structure corresponding to the relation of the respective special fields and managed, and the translation environment supply means for supplying the translation environment for the translation is automatically set in accordance with the correspondence relation of the translation document and the field instruction means and the relation among the field instruction means in the hierarchical structure.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a machine translation system, and can be applied to, for example, a case where a natural language is machine-translated using an appropriate technical term dictionary automatically selected.

[0002]

2. Description of the Related Art The following literature 1 describes a conventional machine translation method and a technical term dictionary selection method.

Reference 1: Japanese Patent Application Laid-Open No. H10-21222 The above reference 1 discloses document identification information (URL (Uniform Resource Locato) used for obtaining a document to be translated.
r)) is compared with the field determination data set in advance, and based on the number of characters matched in this comparison, etc.
Searching for the best matching specialty.

[0004] By automatically setting up a technical term dictionary corresponding to the most specialized technical field and performing machine translation, a high-quality translation result can be obtained without forcing the user to make complicated settings. it can.

[0005]

SUMMARY OF THE INVENTION However, URL
Is a notation method for uniquely representing a certain location of the document, and does not always use a character string representing the field to which the content of the document belongs in the URL. Therefore, it is often difficult to determine the field to which a document belongs simply by examining the URL.

That is, of the field determination data,
There is no guarantee that the field determination data that best matches a certain URL accurately corresponds to the field to which the translation target document belongs.

[0007] Further, from the viewpoint of the contents, an arbitrary document to be translated may belong to a plurality of fields, and the degree of specialization may include various contents. There is.

From the comparison between the URL and the data for field determination, it is considered that it is more difficult to appropriately address these problems than to accurately specify the field to which the document to be translated belongs.

[0009]

According to the present invention, in order to solve the above-mentioned problem, a received translation source document of a first language is converted into a translation destination document of a second language under a set translation environment. In a machine translation system for translation, (1) hierarchical field management means for managing field indicating means indicating a field of expertise to which each source document belongs in a hierarchical structure corresponding to the relationship of each field of specialization, and (2) Translation environment supply means for supplying a translation environment corresponding to each field instruction means in the hierarchical structure;
(3) At the time of the translation, one or more translation environments are provided for the translation in accordance with the correspondence between the source document and the field designating means and the relationship between the field designating means in the hierarchical structure. And a translation environment setting means for automatically setting the translation environment supply means.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS (A) Embodiment The machine translation communication system according to the first to fifth embodiments hierarchizes fields according to a predetermined industrial technology system, and sorts documents according to the hierarchized fields. In a directory service or the like, a (specialized) term dictionary used for translation is automatically and finely set so that high-quality translation results can be easily obtained.

That is, the first to fifth embodiments provide a machine translation communication system for a directory service capable of realizing high translation quality with a simple operation by cooperating with a directory service such as the Internet. .

Here, the field means any specialized field such as industrial technology such as "communication", "computer", and "Internet", or any technical field such as industrial technology or academic field. In a case where a general term dictionary is used because of lack of specialty, the “general” is one of the fields.

A directory server that determines a term dictionary and executes machine translation prepares a directory hierarchical structure (tree structure) that faithfully reflects the relationship between each category (field).

In this hierarchical structure, the categories and term dictionaries are arranged so that the specialty increases from the highest level to the lower level. Therefore, in order to access a target category, the higher the degree of specialty is, the lower the level is. It follows a hierarchical structure.

In an ordinary database, a hierarchical logical structure is suitable for tracing categories (data) from the top to the bottom. It is said that the procedure for finding the data is not suitable because it is complicated.

However, in the hierarchical structure of each embodiment of the present invention, the method of tracing from lower to higher is not to find target data but to construct a dictionary set used for one machine translation. The access from the lower level to the upper level is fixedly performed along each branch of the tree structure, and there is no danger of complicating the procedure. By creating a hierarchical structure corresponding to the position of each field in the industrial technology system in advance, a desired dictionary set can be constructed by such simple processing.

In the first to fifth embodiments, the URL
Is used to uniquely identify the location of the document on the network, as its original purpose.
L is also used to determine which category each document belongs to in the hierarchical structure.

(A-1) Configuration of First Embodiment FIG. 1 shows the configuration of a machine translation communication system 10 according to the present embodiment.

In FIG. 1, the machine translation communication system 10 includes a browsing unit (user terminal) 101, a directory server 102, and an external network 108.

The directory server 102 includes a communication control unit 103, a translation control unit 104, a machine translation unit 105, a hierarchical database 106, and a dictionary database 107. Dictionary database 10
7 is a general term dictionary D1 and an information dictionary D2 as an example.
, A computer dictionary D3, and an Internet dictionary D4
And a personal computer dictionary D5 and a personal computer communication dictionary D6.

The browsing means 101 is a user terminal equipped with a function (browsing software, for example, a Web browser) for displaying information of a designated address (URL) on a network on a browsing screen to inform the user. In the example of FIG. 1, the browsing unit 101 is directly connected to the directory server 102. However, the connection between the browsing unit 101 and the directory server 102 may be performed via the network 108.

The dictionary database 107 is a database for storing and managing translation dictionaries D1 to D6 classified by category. The external network 108 is a network such as the Internet. Are located in

The communication control means 103 is connected to the external network 1
08 and a function of controlling communication with the browsing unit 101.
The translation control unit 104 has a function of directly controlling the machine translation unit 105 such as which document is to be translated and which dictionary is to be used. That is, it is a part for directly setting the translation environment for the machine translation means 105.

The machine translation unit 105 translates the given document into another language under the control of the translation control unit 104. As the machine translation means 105, an existing machine translation system can be used. The document to be translated is stored in the directory server 10 based on the designation from the user terminal 101.
Reference numeral 2 denotes a document obtained from the network 108.

The hierarchical database 106 is functionally a central part of the directory server 102, and logically constitutes the hierarchical directory 11 as shown in FIG.
The attribute information (document address) of the document on the network 108 is centrally managed.

That is, the hierarchical database 106
The documents on the network 108 are classified into fine categories shown in FIG.

In FIG. 2, the ellipses correspond to one category, and each category has a unique address (U
RL). Each category has links to more detailed categories. Each category has its own terminology dictionary, but not all categories need to have a dictionary.

Each category has zero or more links URL1 to URL7 to the external network 108 (portions indicated by squares in the figure).

The hierarchical directory 11 shown in FIG.
Root CT1, which is a category, hobby CT2 belonging to the second highest hierarchy, computer CT3, information CT4
Internet CT5, computer magazine CT6, hardware (computer-related hardware) CT7, software CT8 belonging to the third category from the top, and Internet CT5 belonging to the fourth hierarchy from the top CT9, PC magazine C
It has three categories of T10 and Java magazine CT11.

In addition, each category CT3, CT5, C
At T7 and CT8, there are links URL1 to URL7 to the external network 108 described above.

The top category CT1 is connected to a general term dictionary D1 containing general terms and sentence expressions having no specialty, and the second highest computer category CT3 is a computer-related category. A computer dictionary D3 with low expertise is connected, and information category C
An information dictionary D3 related to information is connected to T4.

To the third highest Internet category CT3, there is connected an Internet dictionary D4 containing Internet-related terms and sentence expressions among computer-related items. Internet dictionary D4
Has a smaller area in the industrial technology system than the computer dictionary D3, and has a high degree of specialty.

FIG. 2 shows the dictionaries D5 and D6 of FIG.
Are not shown, but the dictionaries D5 and D6 are usually
Connected to some category not shown.

The directory server 102 having the databases 106 and 107 therein functions as a kind of general-purpose database relating to attribute information registered in the directory service as a whole. Here, as the attribute information, the addresses of the respective documents on the network 108 included in the respective categories are centrally managed.

Therefore, the directory server 102
With respect to the managed document addresses, the information (documents) on the network 108 are classified by category, and it is possible to assist the user in searching for desired information.

FIG. 4 shows a display example of a directory service page.

Referring to FIG. 4, a window 401 for displaying a browser includes a text field 402 for displaying an address and an information area 403 for displaying information on the address.

From the text field 402, the address of a document on the network 108 to be viewed or the address of a category in the hierarchical directory 11 can be directly input.

In the example of FIG. 4, the information displayed in the information area 403 is the category CT6 of the "computer magazine" in the directory service.

In the information area 403, under the notation M1 of "computer magazine", the notation M2 of "Internet magazine", the notation M3 of "PC magazine", and the notation M4 of "Java magazine" are respectively the current category. Category CT9 to CT1 that is one level lower than the computer magazine
1, the notation M2 to M4 can be moved to the category by, for example, clicking.

In the information area 403, the notation M5 of "XX magazine" shown below the center line L1,
The notation M6 of "PC @" indicates the title on the external network 108 classified into the current category CT6, and clicking this obtains information (contents of the document) on the external network 108. And can be displayed.

The "xx magazine" and "PC @
The symbol “Δ” can be displayed as notations M5 and M6 because the directory server 102 manages the document address and determines in advance which category of the hierarchical directory 11 the document content belongs to. Things.

Regarding general documents on the network 108 outside the management range of the directory server 102, it is not known whether or not they belong to each category, and cannot be displayed in this way.

Focusing on the function of the window 401 as described above, the procedure until the user obtains the display state of FIG. 4 by the browsing means 101 is as follows. First, the URL of the directory server 102 is specified and the information of the root category CT1 is obtained. Displayed in window 401, then click on the word "computer" to select computer category CT3,
Finally, the procedure is usually such that the notation "computer magazine" is clicked to select the computer magazine category CT6. In this case, the user follows the hierarchical directory 11 from the upper level to the lower level.

Of course, the computer magazine category C displayed in the text field 402 in the state of FIG.
If the user knows the URL of T6,
By inputting RL, the display state of FIG. 4 can be obtained from the beginning.

In the present embodiment, this function is implemented by CGI (Comm) so that the information of the currently displayed category (document) and the information of the document to be displayed (target document) can be understood.
mon Gateway (Mon. Gateway Interface).

Eventually, in the state of FIG. 4, the notations M5 and M6
In order to obtain the document address, the document address must be within the management range of the directory server 102 and must be a CGI.

When the directory server 102 receives a document to be translated from the network 108 and the document to be translated is a document within the management range of the directory server 102, the directory server 102 assigns a translation environment identifier to the document. Can be granted.

The translation environment identifier is stored in the directory server 1
02 for each document (to be translated)
The user terminal 1 together with the document in a form attached to the document
01 is sent.

If one document sent to the user terminal 101 belongs to, for example, a plurality of fields, only one of the plurality of fields that is most suitable for the contents of the document is displayed by the translation environment identifier.

The most suitable one field is set as one field corresponding to the most specialized part in the document, so that the plurality of fields exist in each branch of the tree structure (hierarchical structure). If there are multiple categories, they can be handled.

In this case, the dictionaries of each category existing in one branch can be integrated from the lower order to the upper order, and a dictionary set used for translating the one document can be constructed.

However, the present embodiment cannot cope with the case where the plurality of fields exist in different branches (branches other than the one-by-one branch).

Which field is most suitable for the document has already been determined at the stage of determining which document belongs to which category in the hierarchical directory 11.

For the translation of a part of a document having a different level of specialty, the dictionary of a plurality of categories existing in each branch in a tree structure (hierarchical structure) is translated into a lower order. This can be dealt with by constructing a dictionary set to be used for translating the one document by integrating from the first to the upper.

From this point, it is desirable to determine which category each document belongs to based on the highest specialty assumed for the document.

Hereinafter, the operation of the first embodiment having the above configuration will be described.

(A-2) Operation of the First Embodiment The operation procedure of the machine translation communication system 10 is shown in the flowchart of FIG.

In FIG. 3, the flowchart includes steps S301 to S312.

In the first step S301, the browsing means 1
In step 01, the user inputs the address of the document to be displayed and sends it to the directory server 102. This operation can also be realized by clicking a link.

Next, in the directory server 102, the communication control means 103 obtains the information of the transmitted address (S302), and checks whether the transmitted information is a translation request (S303).

In step S303, if the information is a translation request, the process proceeds to step S311; otherwise, the process proceeds to step S304.

In step S304, the communication control means 103 checks whether or not the received address is an address in the directory server 102 (an address within the management range of the directory server 102). Proceeds to S308, where the server 102
If the address is within the range, the process proceeds to step S305.

As an example of a method of determining whether or not the address is in the server, the address of the system is “http: //www.f
If it is oo.co.jp/search/ ”, a method of judging whether or not the address includes the address can be considered.

In step S305, the communication control means 10
It is checked whether 3 is an address of CGI. If not, the process proceeds to step S312.
Proceed to 306. Processing proceeds to S312 when a link to a lower layer than the current document is specified.

The determination as to whether or not the address is a CGI address can be made, for example, based on whether or not the address is “* .cgi”.

In step S306, the communication control means 10
3 acquires the address information of the current document (category (hypertext: linked document)) and the address of the target document (hypertext: linked document) from the CGI program.

Next, the translation control means 104 traces up the node of the current document address in the hierarchical database 106, and if there is a dictionary at that node, sets it in order as a dictionary to be used for translation (S307). ).

For example, in FIG. 2, when the document address designates the "Internet" category CT5, the dictionaries to be used are the Internet dictionary D4, the computer dictionary D3, and the general term dictionary D1, and are composed of D4, D3, and D1. Translation using a dictionary set.

In this case, at the time of translation, if the same word is registered in the Internet dictionary D4 and the computer dictionary D3, the translated word of the Internet dictionary D4 is used preferentially. That is, hierarchical directory 1
In 1, the closer to the category to which the document to be translated belongs, the lower the dictionary related to the category, the higher the priority.

In step S308, the dictionary to be used is the general term dictionary D1.

If it is not an address in the server 102, it is unknown which category in the hierarchical directory 11 the document belongs to, so the general term dictionary D1 is applied uniformly.

In step S 309, data at the designated address (ie, hypertext document data)
To the external network 1
08.

In step S310, the communication control means 103 adds the dictionary information to be used (translation environment identifier) and the translation button to the acquired information, and
Send to 1.

FIG. 5 shows an example of the transmitted screen.

In FIG. 5, this screen is displayed in window 5
01, a text field 502, and an information area 50
Contains three. FIG. 5 is substantially the same as FIG. 4, in which a window 501 corresponds to the window 401, a text field 502 corresponds to the text field 402, and an information area 503 corresponds to the information area 403.

In the information area 503, a notation M10 of “Welcome to the XX company homepage”, a notation M11 of “new arrival information”, a notation M12 of “company outline”, a notation M13 of “employment information”, There is a notation M14 of "related company". The notation M10 corresponds to the notation M1, and the notations M11 to M11.
14 corresponds to the above notations M2 to M4.

In addition, a translation button 504 is displayed in the information area 503. The user can access the browsing means 10
1 to browse this window 501, and
By clicking 04, a translation request can be transmitted to the directory server 102 and a translation result can be received.

As a method for obtaining a translation result by adding information for translation and transmitting the information, a method described in Japanese Patent Application Laid-Open No. 8-202934 (Reference 2) or the like can be used.

When the principle (pre-translation) of Document 2 is applied to the present embodiment, when a document address is input and the information of the address is obtained, machine translation of the document (within the management range) is automatically performed. If the translation is requested, the translation result is stored in the translation storage means (memory). If a translation request is issued, the stored translation result is immediately taken out. By supplying the translation to the user terminal 101, the apparent time required to obtain the translation result can be shortened, and the user's trouble can be reduced.

Since the translation result can be supplied to user terminals other than the user terminal 101, it is possible to reduce the amount of processing until the translation result is obtained in the entire system. If the translation is in progress at the time when the translation request is issued, the partially translated document may be supplied to the user terminal, or may be supplied to the user terminal after the translation is completed. . In any case, the time from when the translation request is transmitted to when the translated sentence is obtained becomes short for the user.

Returning to the flowchart of FIG. 3, in step S311, the machine translation means 105 performs machine translation using the specified dictionary set, and the communication control means 103
Transmits the translation result to the browsing unit 101.

In step S312, the CGI
Therefore, the information of the address of the current document (category (hypertext)) and the address of the target document (hypertext) cannot be obtained, and the data (document) of the specified address is simply stored in the user terminal 1.
Just send to 01.

However, the fact that the processing has proceeded to step S312 means that the branch of step S304 is on the Y side, and that the address of the document is within the management range of the directory server 102. Therefore, the directory server 102 knows which category the document belongs to, and when a user who browses the document on the user terminal 101 sends a translation request, the process of step S311 sets the dictionary set. And machine translation using the dictionary set can be performed at once.

Of the progress of the processing that can be performed in the flowchart of FIG. 3, the most preferable one for this embodiment is steps S301, S302, S303, and S30.
4, S305, S306, S307, S309, S31
In this case, the processing proceeds in the order of steps S301, S302, S303, and S311 in response to a translation request transmitted by a user who has viewed the screen shown in FIG.

The least desirable is step S304.
, And the machine translation is performed using only the general dictionary D1. In this case, the quality of the translation result is likely to be lower than in other cases.

On the other hand, in the above-described case where the branch of step S305 is made on the N side, there is a possibility that a problem may occur from the viewpoint of usability and operability. This is the same as the most preferable case.

As the management range of the directory server 102 in the network 108 increases, the possibility that the most unfavorable case occurs decreases, and the possibility that the user can obtain a high-quality translation result increases.

(A-3) Effects of the First Embodiment According to the present embodiment, one or a plurality of technical term dictionaries (dictionary sets) used for machine translation of one document are provided.
Since the setting can be made automatically and finely, it is possible to obtain a high-quality translation result without putting a burden on the user.

(B) Second Embodiment The present embodiment is characterized in that the terms and sentence expressions recorded in each dictionary of the first embodiment are automatically updated.

(B-1) Configuration and Operation of Second Embodiment In the configuration of the first embodiment and the configuration of the present embodiment, the main difference in operation is limited to the update of the contents of the dictionary.

Hereinafter, only the configuration and operation of the present embodiment that are different from those of the first embodiment will be described.

FIG. 11 shows the configuration of the machine translation communication system 20 of the present embodiment. The machine translation communication system 20 is the same as the machine translation communication system 10 of the first embodiment in that it has a browsing unit 101 and an external network 108, but the directory server 200 is different from the directory server 102. Having.

In FIG. 11, the browsing means 101, the external network 108, and the components 103, 104, 105, and 106 in the directory server 200 are the same as those of the first embodiment, including their functional aspects. Since they are the same, their detailed description is omitted.

The dictionary database in the directory server 200 has almost the same functions as the dictionary database 107, such as having the dictionaries D1 to D6, but receives the action from the dictionary input means 203 to update the dictionary contents. Since it is also provided with a function of performing, a reference numeral 204 is attached to distinguish it from the dictionary database 107.

In addition to the constituent elements 103 to 106 and 204,
Directory server 200 of machine translation communication system 20
Includes a bilingual URL registration unit 201, a bilingual knowledge acquisition unit 202, and a dictionary input unit 203.

The bilingual URL registration means 201 is a means for allowing a user to register his / her home page and the like.

The bilingual knowledge acquisition means 202 acquires knowledge from a bilingual document, and can automatically associate a sentence even if the bilingual document has no correspondence for each sentence.
Dictionary data can be created automatically. As the knowledge acquiring means 202, for example, Japanese Patent Laid-Open No. Hei 10-11445
Japanese Unexamined Patent Publication (Kokai) No. 3 (Document 3) can be used.

In Document 3, a translation pattern is created by performing statistical processing (translation word estimation processing) on a bilingual document. That is, in a bilingual document in which a source language document and a target language document, which is a translation of this document, are associated with each sentence unit, the number of occurrences of a specific word in the source language document and the target language Introducing the concept of a word correspondence that has a higher value as the number of occurrences of a specific word in a document and the number of occurrences of the specific word in the same translation are closer to each other. A word pair between words having the highest value is extracted as a bilingual word.

According to this, the linguistic information can be obtained from the entire document without depending on only the linguistic information in one bilingual sentence.
In practice, it is not necessary to perform a parsing process that is often difficult to execute, and it is possible to identify a bilingual relationship even in a document in which technical terms that are not registered in a general bilingual dictionary are frequently used.

The document 3 is based on the premise that a document and a document in the target language, which is a parallel translation, are respectively associated with each sentence, but sentences may be automatically associated with each other. It is possible.

For example, as a method of automatically associating sentences, there is the following method.

In some cases, one original sentence and three translated sentences or one original sentence and two translated sentences correspond to each other. In this method, first, all combinations (paths) of corresponding pairs of the original sentence and the translated sentence are obtained. And the number of sentences of the translated sentence and a combination having a high possibility of the optimal solution of the association are obtained.

Next, an evaluation value which becomes higher as the original sentence and the translated sentence correspond to each other is obtained, and a sum of the evaluation values is obtained for a combination which has a high possibility of the optimal solution of the correspondence between the original sentence and the translated sentence. Provisional solution.

After that, the sum of the evaluation values of the combination of the original sentence and the translated sentence is calculated sequentially from the first sentence to the last sentence of the combination, and even if the calculation is performed up to the last sentence, the sum exceeds the provisional solution. When it is determined that the combination is not satisfied, the calculation of the evaluation value of the combination is stopped, and the provisional solution is determined as the optimal solution and used as the solution for associating sentences.

In response to the supply of the output of the knowledge acquiring means 202 from the bilingual translation which can use such a method, the dictionary input means 203 of FIG. Register the data (translation).

FIG. 9 shows an example of a bilingual registration screen viewed by the user via the user terminal 101.

Referring to FIG. 9, a bilingual registration screen 900 includes
Form 90 describing English URL to be registered by user
1, a form 902 describing a Japanese URL to be registered by the user, a category list 903 for selecting a category to be registered by the user, and a button 904 for sending to the system when the description is completed is displayed. ing.

In this example, the bilingual registration screen 900 registers bilingual English and Japanese translations.

UR performed via bilingual registration screen 900
FIG. 6 shows a flowchart at the time of L registration. This flowchart includes steps S601 to S605.

In FIG. 6, first, the user inputs a bilingual URL.
By performing operations such as clicking on the registration link,
The bilingual URL registration means 201 displays a bilingual registration screen (S601). The bilingual registration screen is, for example, a screen such as the bilingual registration screen 900.

Next, the user inputs the forms 901 and 902
Translation documents to register (English and Japanese documents)
And the category to be registered are entered (S602), and a send button 904 is pressed.

The directory server 200, which has received the URL of the bilingual document and the information on the category to be registered,
The document of L is obtained from the external network 108 via the communication control means 103, and the document is sent to the knowledge acquisition means 202 (S603).

The knowledge acquiring means 202 analyzes the bilingual document and automatically creates dictionary data (S604).

The dictionary input means 203 registers the created dictionary data in the dictionary of the selected category (S605).

(B) Effects of the Second Embodiment According to the present embodiment, effects equivalent to those of the first embodiment can be obtained.

In addition, according to the present embodiment, the URL by the user, which exists in most directory services,
Since the dictionary data of each technical term dictionary is automatically updated using the registration, the ability of each technical term dictionary is improved, and a high-quality translation result can be obtained.

Therefore, a high-quality translation result can be obtained without preparing a technical term dictionary for each category.

In the present embodiment, a document (a bilingual document) written in a plurality of languages is required as a premise, but a home page describing the same contents in a plurality of languages is currently used in order to be seen by more people. It is increasing. The owner of such a home page will make a bilingual registration to make it known to many people.

(C) Third Embodiment The present embodiment is characterized in that a specialized dictionary can be automatically created even without a translation required in the second embodiment.

(C-1) Configuration and Operation of Third Embodiment In the configuration of the second embodiment and the configuration of the present embodiment, the main difference in operation is limited to the point related to dictionary data creation processing.

In the following, only the structure and operation of this embodiment different from those of the second embodiment will be described.

FIG. 7 shows the configuration of the machine translation communication system 30 of the present embodiment. The machine translation communication system 30 is the same as the machine translation communication system 10 of the second embodiment in that it has a browsing unit 101 and an external network 108, but the directory server 300 is different from the directory server 200. Having.

In FIG. 7, browsing means 101, external network 108, and components 103, 104, 105, 106, 203, and 20 in directory server 300 are provided.
4 is the same as that of the second embodiment, including its functional aspects, and will not be described in detail.

Components 103 to 106, 203 and 204
In addition, the directory server 300 of the machine translation communication system 30 includes a URL registration unit 301 and a knowledge acquisition unit 3
02.

The URL registration means 301 is a means for allowing a user to register his / her home page and the like.

The knowledge acquiring means 302 uses a predetermined method from one document in one language, while the knowledge acquiring means 202 of the second embodiment performs statistical processing (translation word estimation processing) on a bilingual document. To acquire dictionary knowledge and automatically create dictionary data.

As a predetermined method, for example, Japanese Patent Laid-Open No. 6-3
There is a method described in Japanese Patent No. 01717 (Reference 4).

Reference 4 introduces the concept of a long unit word composed of a character string of two or more morphemes sandwiched between functional words and punctuation marks, and converts a long unit word and long unit word bilingual information from an input sentence. And automatically extracts information such as the attributes of the words constituting the long unit words or the translations of the words constituting the long unit words. Therefore, when the information such as the extracted translated word is different from the dictionary data already stored in the dictionary, it is possible to update the contents of the dictionary by adding it as new dictionary data.

This method is different from the method of Reference 3 in that
Although the reliability of the extracted information may be low, it has the great advantage of not requiring a translation.

FIG. 10 shows an example of a URL registration screen.

Referring to FIG. 10, a URL registration screen 1000 is displayed.
Is a form 100 describing a URL to be registered by the user.
1 and button 1 for specifying the language to be registered by the user
002, a category list 1003 for the user to select a category to be registered, and a button 1004 for sending to the system when the description is completed.

FIG. 8 is a flowchart for registering a URL according to the present embodiment. This flowchart includes steps S801 to S805.

In FIG. 8, the user first clicks a URL registration link, and the URL registration means 301 causes the user terminal 101 to display a URL registration screen (URL).
An L input screen is displayed (S801).

The URL registration screen is, for example, like the URL registration screen 1000 described above.

Next, the user enters the URL of the document to be registered in the form 1001 and the category to be registered in the category list 1003 (S802), and presses the send button.

In response to this, the directory server 300 transmits the document of the URL and the document of another language (a language different from the registered URL) registered in the category from the external network 108 via the communication control means 103. Obtain and send the document to the knowledge acquisition means (knowledge acquisition system) 302 (S803).

Then, the knowledge acquiring unit 302 analyzes the document and automatically creates dictionary data (S804).

The created dictionary data is stored in the dictionary input means 203.
Is registered in the dictionary (any one of D1 to D6) of the selected category.

In the present embodiment, the translation is described in English and Japanese, but the translation can be similarly performed in other languages.

(C-2) Effects of the Third Embodiment According to the present embodiment, it is possible to obtain effects substantially equivalent to the effects of the second embodiment.

In this embodiment, however, the accuracy of the dictionary data to be updated may be lower than that in the second embodiment, but there is an advantage that a specialized dictionary can be automatically created without a translation. .

(D) Fourth Embodiment This embodiment is characterized in that a dictionary is registered in a dictionary database in a plurality of languages.

(D-1) Configuration and Operation of the Fourth Embodiment In the configuration of the second embodiment and the present embodiment, the main difference in operation is the presence or absence of the language determination means 401.

Hereinafter, only the configuration and operation of the present embodiment that are different from those of the second embodiment will be described.

FIG. 12 shows the configuration of the machine translation communication system 40 of the present embodiment.

In FIG. 12, constituent elements 101 and 10
3,104,105,106,108,201,20
The components 2 and 203 are the same as those of the second embodiment to which the same reference numerals are assigned, including the functional aspects, and therefore, detailed description thereof is omitted.

The dictionary database 402 is the same as the dictionary database 204 of the second embodiment in that it has a dictionary for translation classified for each category. However, the dictionary database 402 includes Japanese, English, Chinese, and French. It has dictionary data for each language.

That is, the general term dictionary P1 includes dictionary data that can be used for translation from English to Japanese, from Chinese to Japanese, from French to Japanese, and so on.

The same applies to the other dictionaries P2 to P6.

The language judging means 401 has a function of automatically judging in what language (Japanese, English, Chinese, etc.) the document is described by examining the language information and code system in the document. Equipped.

According to this determination, a part (dictionary data) used for the translation is determined within any of the dictionaries P1 to P6 corresponding to the dictionary set used for any one translation.

For example, when the document is described in English, it is determined that only the portion corresponding to the translation from English to Japanese in the dictionary of the dictionary set is used.

(D-2) Effects of the Fourth Embodiment As described above, according to the present embodiment, effects equivalent to the effects of the second embodiment can be obtained.

In addition, in this embodiment, the language of the document registered in the database can be a plurality of languages (can be mixed). This point is important when expanding the management scope of directory servers worldwide.
It is advantageous.

(E) Fifth Embodiment The relationship between the present embodiment and the third embodiment is the same as the relationship between the fourth embodiment and the second embodiment.

(E-1) Configuration and Operation of Fifth Embodiment In the configuration of the third embodiment and the present embodiment, the main difference in operation is the presence / absence of the language determination means 501.

In the following, only the configuration and operation of this embodiment different from those of the third embodiment will be described.

FIG. 13 shows the configuration of the machine translation communication system 50 of the present embodiment.

In FIG. 13, the components 101 and 10
3,104,105,106,108,301,30
2 and 203 are the same as those of the third embodiment, which are denoted by the same reference numerals, including the functional aspects, and therefore, detailed description thereof is omitted.

The dictionary database 40 of the present embodiment
The configuration and function of No. 2 are the same as those of the dictionary database of the fourth embodiment with the same reference numerals.

The function of the language determining means 501 is the same as that of the language determining means 401 of the fourth embodiment.

(E-2) Effects of the Fifth Embodiment According to the present embodiment, the same effects as those of the third embodiment can be obtained.

In addition, according to the present embodiment, the language of the document registered in the database can be a plurality of languages (can be mixed). This is also advantageous when expanding the management range of the directory server worldwide.

(F) Other Embodiments In the first to fifth embodiments, the browsing means, the directory server, and the external network are connected to each other by a network. I do not care.

In the first to fifth embodiments, the CGI
However, the same function can be realized by using Java or the like. If the address information can be transmitted only by the link on the screen of the browsing means, the node of the hierarchical database can be realized on the server. Can be realized by a normal hyperlink mechanism.

Further, in the first to fifth embodiments, the translation environment identifier is assigned to the document to be translated by the directory server 102 and is sent to the user terminal 101. This is not always necessary, as the dictionary used for translation is automatically determined.

In the first to fifth embodiments, the dictionary D
Although 1 to D6 are provided in the directory server 102, dictionaries existing on the network 108 may be used for machine translation of the directory server 102 as necessary.

Although the number of dictionaries in the first to fifth embodiments is six, the number of dictionaries may be more or less. The hierarchical directory handled is shown in FIG.
May have more categories and levels than
It may be less.

When the number of categories and the number of hierarchies increases and the hierarchical directory becomes large, it is not always necessary to construct a dictionary set by going back to the highest root (generic term dictionary). The dictionary set may be constructed halfway back, or the dictionary set may be constructed to include a category lower than the corresponding category by, for example, one hierarchy level.

Further, a dictionary included in the dictionary set may be selected by tracing another branch, for example, one level lower, from the category in the middle of tracing from lower to higher. This makes it possible to cope to some extent with the above-described case in which the document to be translated belongs to a plurality of fields and the plurality of fields exist in different branches.

Further, in the fourth and fifth embodiments, Japanese, English, Chinese and the like are exemplified, but the type of language is not limited to this.

Further, in the fourth and fifth embodiments,
Although the language of the translation source document is plural and the language of the translation destination document is one type, the language of the translation destination document may be plural.

In the first embodiment and the like, pre-translation is performed, but this is not always necessary.

That is, the present invention can be widely applied to a machine translation system that translates a received source document in a first language into a destination document in a second language under a set translation environment. it can.

[0176]

As described above, according to the present invention, one or a plurality of translation environment supply means for supplying a translation environment can be automatically set according to the specialty to which the source document belongs. It is possible to obtain a high-quality translated document that is easy to operate, easy to use, and high in quality.

[Brief description of the drawings]

FIG. 1 is a schematic diagram illustrating a configuration of a machine translation communication system according to a first embodiment.

FIG. 2 is a schematic diagram showing a configuration of a hierarchical directory according to the first to fifth embodiments.

FIG. 3 is a flowchart showing an operation of the first to fifth embodiments.

FIG. 4 is an example of a screen display showing an operation of the first to fifth embodiments.

FIG. 5 is an example of a screen display showing the operation of the first to fifth embodiments.

FIG. 6 is a flowchart showing the operation of the second and fourth embodiments.

FIG. 7 is a schematic diagram illustrating a configuration of a machine translation communication system according to a third embodiment.

FIG. 8 is a flowchart showing the operation of the third and fifth embodiments.

FIG. 9 is an example of a screen display showing the operation of the second and fourth embodiments.

FIG. 10 is an example of a screen display showing operations of the third and fifth embodiments.

FIG. 11 is a schematic diagram illustrating a configuration of a machine translation communication system according to a second embodiment.

FIG. 12 is a schematic diagram illustrating a configuration of a machine translation communication system according to a fourth embodiment.

FIG. 13 is a schematic diagram illustrating a configuration of a machine translation communication system according to a fifth embodiment.

[Description of References] 10, 20, 30, 40, 50: Machine translation communication system, 11: Hierarchical directory, 101: User terminal (browsing means), 102, 200, 300, 400, 500 ...
Directory server, 107, 204, 402 dictionary database, 103 communication control means, 106 hierarchical database, 108 (external) network, D1-D
6: Technical term dictionary, P1 to P6: Technical term dictionary, CT1
~ CT11 ... category.

Claims (6)

[Claims] In a machine translation system for translating a received source document in a first language into a target document in a second language under a set translation environment, a field of expertise to which each source document belongs is specified. The field indicating means shown
A hierarchical field management unit that manages the fields in association with each other in a hierarchical structure according to the relationship between the specialized fields; a translation environment supply unit that supplies a translation environment corresponding to each field designating unit in the hierarchical structure; Automatically activates one or more translation environment supply means for supplying a translation environment for the translation in accordance with the correspondence between the source document and the field designation means and the relationship between the field designation means in the hierarchical structure. And a translation environment setting means for setting the translation environment. 2. The machine translation system according to claim 1, wherein the translation environment supply unit is a technical term dictionary. 3. A machine translation system according to claim 2, wherein a document in the source language and a document in the target language, which is a translation of the document, are registered in a bilingual document corresponding to each sentence unit. Means, a bilingual dictionary data obtaining means for automatically obtaining dictionary data by comparing the bilingual document, and a bilingual translation for reflecting the obtained dictionary data to the contents of a technical term dictionary in the same technical field as the bilingual document. A machine translation system comprising: a dictionary content changing unit. 4. A machine translation system according to claim 2, wherein a single document registration means for registering a single document, and a single dictionary data acquisition for automatically obtaining dictionary data based on a predetermined process for said document A machine translation system comprising: a unit; and a single dictionary content changing unit that reflects the dictionary data obtained here to a technical term dictionary in the same technical field as the document. 5. The machine translation system according to claim 2, wherein a document in the source language and a document in the target language, which is a translation of the document, are registered in a bilingual document corresponding to each sentence. Means, a bilingual dictionary data acquiring means for automatically obtaining dictionary data by comparing the bilingual documents, and a bilingual translation for reflecting the obtained dictionary data to the contents of a technical term dictionary in the same technical field as the bilingual document. Single-document registration means for providing a bilingual translation processing means having dictionary content changing means and receiving registration of a single document; and single-dictionary data for automatically obtaining dictionary data based on predetermined processing for the document. A single document processing means comprising: an acquisition means; and a single dictionary content changing means for reflecting the dictionary data obtained here in a technical term dictionary in the same technical field as the document. Machine translation system characterized in that it comprises selecting means for operating to select one of translation processing means or a single document processor. 6. The machine translation system according to claim 2, wherein the technical term dictionary is a dictionary for a plurality of languages corresponding to a plurality of types of languages, at least for a first language. A machine translation system that includes a translation source language determining unit that determines a type of a first language of the translation source document, and that performs translation from a plurality of first languages to a second language.