KR100248385B1 - Automatic translation system for formatted documents - Google Patents

Abstract

The present invention provides information necessary for the items of the first language form so that the first language user can conveniently form a form document of a formal form such as a trade transaction or a general form form document such as an invitation or a hotel reservation as a second language form. It is an invention for a form document automatic translation system in which form documents of a formal language and a general form of a target second language are automatically created simply by inputting. The present invention has the advantage of supporting not only the text of the second language required by the user but also the form of the document language of the second language.

Description

Form document automatic translation system

The present invention relates to a form document automatic translation system, and more specifically, a formal form document such as a trade transaction between countries or a general form form document such as an invitation or reservation made between a company or an individual, as well as text form. By automatically translating from the first language to the second language, it is possible to provide the convenience of eliminating language barriers between countries and writing foreign documents. In this regard, the present invention relates to a format document automatic translation system that can provide economic convenience of trade and socially contribute to the understanding of document style culture.

The reason why the automatic translation system still does not sufficiently support the user's demand for the high quality translation is because of the following important problems related to the automatic translation technology development.

(1) Pursuit of a fully automatic translation system without human intervention

(2) Pursuing an open translation area where the translation area is not limited

(3) incomplete translation of culture-dependent foreign language formatting documents;

The first problem, the pursuit of a fully automatic translation system in which human beings are not involved, refers to the ideal form that the automatic translation system should continue to pursue. However, high quality translations require continuous research for a long time in practical terms. Therefore, appropriate technical measures must be taken for commercialization.

The second problem, the pursuit of an open translation area, in which the translation area is not limited, is the factor that lowers the quality of translation because it causes the multiplicity problem that occurs between languages. Although the method of automatic translation is being considered, the problem is that a limited area automatic discrimination method is still being studied and that it depends on the terminology dictionary. There is a problem with knocking down.

The third problem, 'incomplete translation of culture-dependent foreign language format document' refers to an automatic translation methodology that corresponds to a specific document format that can appear in each language, and the existing automatic translation system does not consider the document format. It could not be translated correctly and caused inconvenience to the user to understand the format of the other country. In addition, although there is a format document formatter that arranges sentences by selecting English from the database for users with basic knowledge of English, it is very monotonous and cannot be used by non-English speakers. It can not be called.

In order to improve the problems of the existing automatic translation system, the present invention allows the first language user to automatically generate a second language format document of a desired form simply by inputting a small amount of information according to the form provided. Its purpose is to provide an automatic translation system for form documents.

Such automatic translation of the format document can enhance the contact effect of translation and culture by presenting a form of a document that can exist only in an English-speaking culture, not just a text translation.

1 is a block diagram showing a format document automatic translation system according to the present invention.

Figure 2 is an exemplary view showing a screen layout of the workbench used in the format document automatic translation system of the present invention.

Figure 3 is a flow chart showing an example of the process of the format finder used in the automatic form document translation system of the present invention.

Figure 4 is a flow chart showing an example of the process of the formatter used in the automatic form document translation system of the present invention.

5 is a flowchart illustrating an example of a process of a form controller used in the form document automatic translation system according to the present invention.

6 is a flow chart showing an example of the process of the format generator used in the format document automatic translation system of the present invention.

7 is an input / output state diagram of the form document automatic translation system in the workbench shown in FIG. 2.

* Explanation of symbols for main parts of the drawings

1: workbench

2: Form Finder

3: Formatter

4: Form controller

5: formatter

6: format saver

7: First Language DB

8: Second Language DB

9: proper noun band dictionary

10: Terminology Band Dictionary

Form document automatic translation system according to the present invention for achieving the above object is a workbench for an interface with a first language user; A format searcher for searching for one form type selected by the format menu in the first language form DB and providing a form requested to the first language user; A format generator for supporting a user's form creation so that a first language user can fill out a form in a first language; A format controller for presenting stored form translations to the user when the user wants to repeat the form translation, eliminating repetitive work, and inducing a form translation when the user creates a new form; A format generator for preparing a second language form document in accordance with contents given in the first language form; A format saver for storing translated form documents; A first language form DB in which the first language form is stored; A second language form DB in which a second language form is stored; Proper noun band dictionaries containing band words of proper nouns, such as names and names; Consists of a terminology band dictionary that stores the terminology of the second language for the terminology of the first language.

According to the present invention having the above-described configuration, a first language user first selects a format type of a first language to be translated in a format searcher and then inputs contents into form items of the selected format type. The contents of the filled out form items are translated into the form contents of the second language in the form generator, and are then written in the second language form corresponding to the first language form. The form thus created is stored in the form store, and the first language user gets the desired second language form document. If the first language user selects the same type of content, the form controller presents the existing form document stored in the form store to the user.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

1 is a block diagram showing an automatic form document translation system according to the present invention. The form document automatic translation system shown in FIG. 1 includes a workbench 1 for interfacing with a first language user and a first language format menu. A format finder (2) for searching the format type selected in the second language format and a format finder (3) for supporting a user's form creation so that the first language user can create a form in the first language; A format controller (4) for presenting the stored form translations to the user to eliminate the repetitive work when the user wants to repeat the form translation and inducing the form translation when the user creates a new form; A format generator 5 for preparing a second language format document in accordance with the contents given in the first language format; A format saver 6 for storing translated form documents; A first language format DB 7 in which the first language format is stored; A second language format DB 8 in which a second language format is stored; A proper noun band dictionary 9, which stores a band word of a proper noun such as a name or a place name; It consists of a terminology band dictionary 10 that stores the terminology of the second language for the terminology of the first language.

2 shows an example of screen arrangement of the workbench 1, wherein the format type of the first language is displayed in the form menu area at the top left of the screen of the workbench 1, and the format of the first language specified in the form menu area is shown. The content of the form type is displayed in the first language format input area in the middle of the screen of the workbench 1, and the second language format corresponding to the selected first language format is the second language format on the right side of the workbench 1 screen. It appears in the output area, and the type of processed second language form appears in the storage form area at the bottom left of the screen of the workbench 1.

3 is a flow chart showing an example of a form search in the form finder 2, which is a normal form creation method operating when there is no existing form. The form type desired by the first language user in the workbench 1 is shown in FIG. When the is selected, the selected first language format type is stored as the first language format type display (S21), and the first language format type display (S21) formats the content in the first language format DB (7). The first language form type mark is searched for (S22), and the form type label transfer (S23) is performed for the second language form search (S22). The first language type mark is delivered from the second language DB (8), and the search for the second language format (S24) is performed, and the content of the second language is retrieved. The retrieved first language form type indication, the first language form type content, the second language form type indication, and the second language form type content are transferred to the form generator 3, which is the next process.

FIG. 4 is a flow chart showing an example of form creation in the formatter 3, wherein the corresponding information of the first form item of the form item constituted by the first language form type contents received from the form searcher 2 is shown. The language form item is entered in the content input process (S32) and the form item and the content of the form item are stored in the item and the content storage (S33). Until the form item constituting the first language form type content is terminated (S31), the above method continues to accumulate the form item and the content of the form item in the item and content storage (S33). When the form item is completed, the first language form type display, the first language form type items and contents, the second language form type display, and the second language form type content accumulated in the above items and contents storage (S33) are moved to the next step. Delivered.

5 is a flowchart showing an example of form control in the form controller 4, and performs three types of control. Firstly, when the first language user selects a form that has been previously written by the first language user directly from the workbench (2), and secondly, when the first language user wants to perform normal format translation, the same content as the previous form translation result is created. If the user is inputting the word, the user can control it and present it to the first language user after modifying a few items, and the third is normal format translation if there is no previous format translation. It is a way to induce. Each of these is described next.

The first method is that when the first language user selects a form previously created by the first language user directly from the storage form area of the workbench (1) menu, the form previously created by the question of the storage form menu selection (S41). The translation is extracted from the format saveer 6, and the previously written date is changed to the current date (S44) to provide the second language format translation to the first language user.

The second method is to reduce the burden on the user by controlling when the first language user wants to perform the format translation of the same content without being aware of the existing format translation. In 1), when the user selects the format type to be translated, the user searches for whether the format type is the same (S42) from the format saver 6, and if the content type is identical to the content of the format item 1 of the same format type (S42), S43) is compared with the format saver 6 again, and if the content of form 1 is the same, the second language format translation stored in the storage formatter 6 is extracted from the format saver 6, and the date previously created. Change to the current date (S44) to provide a second language format translation to the first language user.

The third method is a normal form translation method, which is implemented in two ways. One is that when the first language user does not have a form type to be translated in the workbench 1 in the form storage unit 6, the first language type display form, 1 The language format type items and contents, the display of the second language form type, and the content of the second language form type are transmitted to the format generator (5), and the format translation is performed. The other is the second method of form control. When a language user selects a form type to be translated in the workbench (1), the form type is the same (S42), but the same content (S43) of the form item 1 is broken, so that the formatter can be processed normally. The first language typetype indication, the first language typetype items and contents, the second language typetype indication, and the second language typetype content transmitted from (3) are transmitted to the format generator 5.

6 is a flowchart showing an example of format translation and format generation in the format generator 5, wherein the first language type indication, the first language type information items and contents, and the second language transmitted from the format controller 4 are shown. Format type display, the second language will be covered. In the first language type items and contents, it is checked whether the contents of the first language item in the terminology dictionary are already stored in the terminology dictionary 9 (S52). If the form item 1 content is in the terminology band dictionary (9), the first language terminology is replaced with the second language terminology terminology by lexical conversion (S54), and the content of the first language item is replaced by the corresponding language of the second language expression. If the first language entry content is not in the terminology band dictionary (9), the first foreign language entry content is in the proper noun dictionary (S53). 10). If the form item 1 content is in the proper noun band dictionary 10, the first language-specific noun is replaced with the second language proper noun by the vocabulary conversion (S54) to replace the content of the first language expression item with the second language expression. Is written to (S57). If the content of the first language expression item is not in the terminology band dictionary (9) or in the proper noun band dictionary (10), a character conversion (S55) is performed on the content of the first language expression item (S55). The result is again improved by the alphabet correction using the heuristic (S56) is to write the contents of the first language expression items in the corresponding column of the second language expression (S57). The overall flow of the contents of the format item 1 above is repeated until all of the format items of the selected first language type are terminated (S51). The first language format type display, the first language format type items and contents, the second language format type display, and the second language format type content completion result output from the format generator 5 are stored in the format saver 6. Will be saved.

FIG. 7 is a screen showing an example of an input and an output of the workbench 1. When a birth certificate which is an example of a form type is selected from the form menu area located in the upper left of the workbench 1, the workbench 1 ) The contents and items of the birth certificate appear in the first language area located in the middle of the screen. Here, the first language user inputs an appropriate content for the item using the first language. The contents of the birth certificate and the output of the items will be displayed as the second language birth certificate on the right side of the workbench screen. In the example of the birth certificate in FIG. 7, the first language is Korean and the second language is English. The birth certificate of English, the second language form, appears as a form type indication in the storage form area at the bottom left of the workbench screen.

As described above, according to the present invention, even if the first language user who does not know or are not familiar with the form and contents of the second language document document and wishes to send the form document in the second language, the content item in the form item is only used using the first language. By inputting the form and contents of the document of the second language, a convenient and complete second form document can be obtained.

Claims (8)

A workbench (1) for interfacing with a first language user and a format finder (2) for searching a form type selected by the form menu in the first language form DB and providing a form required for the first language user. A formatter that supports the user's form creation so that the first language user can fill out the form in the first language, and repeats the work by presenting the format translations saved to the user when the user wishes to repeat the format translation. When the user creates a new form, the form controller (4) for inducing the form translation, the form generator (5) for preparing the second language form document according to the contents given in the first language form, and the translated form document And a format storage unit (6) for storing, a first language form DB (7) in which the first language form is stored, and a second language form DB (8) in which the second language form is stored. 1 language A form document automatic translation system, comprising: translating a form document into a form document of a second language. The system of claim 1, wherein the first language is Korean and the second language is English. 2. The first language form according to claim 1, wherein a form menu screen area indicating the type of the form, a storage form screen area indicating the type of the stored form, and a first language form indicating the item of the first language form are used as screen areas constituting the workbench 1; A form document automatic translation system, characterized by a screen structure consisting of a second language output area representing an input area and a content of a second language expression. 2. The form searcher according to claim 1, wherein the form searcher (2) searches the first language form DB (7) to link the searched first language form type with the second language form type of the second language form DB (8). A form document automatic translation system characterized by a means for delivering a type mark (S23). 2. The format document automatic translation system according to claim 1, comprising means (S32) for inputting first language content into each item of the searched first language form type and means (S33) for storing each item and content. The method according to claim 1, further comprising: outputting the stored second language form translation if the form is already stored through the form storage unit (6), and determining whether to continue the form translation if not stored (S41, S42, S43). S44) format document automatic translation system characterized in that. 7. The system of claim 6, wherein the first language is Korean and the second language is English. The method according to claim 1, wherein the terminology band dictionary (9) and the proper noun band dictionary (10) are used to generate a form document of a second language according to the contents of the form document item of the first language given (S54). Character format conversion (S55), the alphabet correction (S56) and the format document automatic translation system, characterized in that the method for writing the content information of the translated first language in the corresponding form of the second language (S57).

Images