JPS63113758A - Machine translation system - Google Patents

Abstract

PURPOSE:To simplify the editorial operation of an operator by presenting plural interpretations concerning only character string specified by the operator to the operator and translating a sentence with the interpretation selected by the operator. CONSTITUTION:In case of editing a translated sentence displayed on a display part 8, if an input part 1 indicates other interpretations concerning a certain character string of an original, an other interpretation control part receives the indication and retrieves interpretation information stored in a parsing syntactic analysis part so as to extract all the interpretation information concerning the specified character string and display it on the display part 8. And the input part 1 selects a piece of information and the other interpretation control part transmits the selected interpretation information to a construction conversion part. The construction conversion part examines cases according to the interpretation information and converts the case into the case of an object language. And a translated sentence generation part generates the translated sentence based on the case information of the objective language. Thus the operator can easily obtain the translated sentence with other interpretation.

Description

[Detailed description of the invention] [Purpose of the invention] (Industrial application field) The present invention relates to a machine translation system.

(Prior Art) In recent years, systems that automatically machine translate input original texts using computers have been attracting attention.

This machine translation system basically divides the human original text into predetermined processing units such as words (phrases) by morphologically and syntactically analyzing it, searches a translation dictionary for each processing unit, and searches the translation dictionary for the corresponding translation ( It is configured to find translated words (words and phrases) and combine them according to predetermined translation rules to obtain the translated text.

However, it is extremely difficult to reliably obtain appropriate translations, partly because the technology for interpreting the meaning of natural languages has not yet been fully established.

For this reason, systems have emerged that present multiple translation candidates to an operator when multiple interpretations of the original text are possible, and output the translation candidate that the operator selects and designates as the most appropriate.

For example, when the original text is in Japanese, there are expressions that can be interpreted in multiple ways, including the position of clause breaks, pronunciation, and part of speech.

The Japanese sentence ``I just met a Japanese person.'' can be divided into two ways: ``Today/the person'' or ``Now/Japanese.'' In conventional systems, for example, one of the two is selected and translated, and the other is translated based on the operator's instructions. Alternatively, the translation is performed in two ways from the beginning, the two translations are presented to the operator, and the operator is asked to select one of them.

(Problem to be solved by the invention) In the above machine translation system, the interpretation of the original text is 2.3.
In this case, the operator can obtain the desired interpretation in a short time. However, when several interpretations are possible, it sometimes takes a considerable amount of time to obtain the desired interpretation. For example, if there are two possible interpretations for each of the three words in the original text, the total number of possible interpretations is 2X2X2-8. It takes a long time for the system to translate all of this. Alternatively, even if the system sequentially translates with different interpretations based on the operator's instructions, the system does not necessarily present the translation expected by the operator immediately.

Therefore, in order to obtain an appropriate translation in a short period of time, the operator has had to either edit the original into an appropriate translation or rewrite the original into a sentence that can only be interpreted in a clear and unique manner. These operations are extremely troublesome and place a heavy burden on the operator.

The present invention has a simple configuration, presents multiple interpretations of only the character string specified by the operator to the operator, and translates the string using the interpretation selected by the operator.
The purpose of the present invention is to provide a machine translation system that simplifies editing operations for an operator and can easily obtain appropriate translation results.

[Structure of the Invention] (Means for Solving the Problems) In order to achieve the above object, the present invention includes an input unit and a translation dictionary that divides the original text inputted through the input unit into words. a morphological analysis unit that searches for the grammatical attributes of each word; a syntactic analysis unit that extracts and stores a syntactically correct interpretation from the information of each word examined by the morphological analysis unit; a structure conversion unit that examines case relationships from the extracted interpretation information and converts the case relationships into case relationships of the target language; a translation generation unit that generates a translated sentence based on the case relationship information of the target language; and the original and translated sentences. a display section that displays , and when editing the translated text displayed on the display section,
When a different interpretation of a certain character string in the original text is instructed via the human power section, the interpretation information stored in the syntax analysis section is searched and all interpretation information regarding the instructed string is extracted. , display this on the display section, and send the selected interpretation information to the structure conversion section via the input section;
Provided is a machine translation system characterized by comprising a separate interpretation control unit that prompts retranslation.

(Function) In a device configured as described above, when a different interpretation of a certain character string in the original text is instructed via the input section when editing the translated text displayed on the display section, the different interpretation control section Upon receiving the instruction, the interpretation information stored in the syntax analysis section is searched, all the interpretation information regarding the instructed character string is extracted, and this is displayed on the display section. When one of them is selected through the input section, the other interpretation control section sends the selected interpretation information to the structure conversion section. The structure conversion unit examines case relations according to this interpretation information and changes the case relations to those of the target language. The translation generation unit then generates a translation based on the case relationship information of the target language. In this way, the operator can easily obtain a translated text with a different interpretation.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings. 1st
The figure is an overall block diagram of a Japanese-English machine translation system that is an embodiment of the present invention. This system controls the entire system: an input section 1 consisting of a keyboard, an original text storage section 2 that stores human-keyed Japanese sentences as the original text, a translation storage section 3 that stores English sentences corresponding to the input Japanese sentences as translated texts. an editing control section 4 that executes the translation process, a translation section 5 that executes the translation process, a translation dictionary 6 that stores knowledge information used in the translation process, a display control section 7 that controls the display of the original text, the translated text, etc., a display section 8, and the original text and the translated text. It consists of a printing section for copying and outputting notebook copies.

FIG. 2 is a diagram showing the key arrangement of the input unit 1.

The input unit 1 is equipped with the following various keys in addition to character keys for Japanese literary skills.

Translation instruction key = 100 Edit key: 101~10G Function key: 110-117 Alternative interpretation key = 118 Cursor key: 130-133 Other keys FIG. 3 is a diagram showing the layout of the screen of the display section 8. As shown in FIG.

The input original text is displayed in the original text display area on the left side of the screen. The translated text obtained as a result of the translation process is displayed in the translated text display area on the right side of the screen at a position corresponding to the original text. Additionally, the editing area at the top of the screen can be used to display information necessary for various edits.

FIG. 4 is a flowchart showing the flow of dialog translation processing by the editing control section 4. In the interactive translation process, the operator can input the original text and edit and control the corresponding translated text as appropriate. In steps 81 to S5, the editing control unit 4 monitors whether there is any key input from the input unit 1 or whether a translation completion signal is received from the translation unit 5, and if there is any key input, Perform the corresponding processing.

If the operator operates any of the function keys, this is detected in step S5, and the process corresponding to that function key is performed in step 812.

When the operator operates the cursor key 130 or other control keys, processing for moving the cursor corresponding to the key operation and other processing are performed in step S13 via steps 81 to S5.

When the operator inputs the original text by operating the character keys, each character key is detected in step S3, and then the corresponding character code is sequentially set in the input buffer (not shown) in the editing control unit 4 (step S8). ). This input original text is displayed in the original text display area of the display unit 8 via the display control unit 7 (step S9). For example, when the operator finishes inputting a - sentence, the operator presses the translation instruction key 100. This key is detected in step S2, and the editing control unit 4 supplies the original text in the input buffer to the translation unit 5. and instructs the start of translation processing (step 510). still,
If input editing such as correction, insertion, deletion, etc. is required during the editing of the original text, move the cursor to the desired editing location using the cursor keys, and then use editing keys such as the insert key 101 and the delete key 102 to perform the input editing. It can be performed.

FIG. 5 is a flowchart showing the flow of translation processing in the translation section 5. The phrase analysis unit (S51) converts the input original text into phrase units using the words obtained from the dictionary search unit (852) and the adjunct connection/conjugation inflection dictionary 6a. The dictionary search unit (S52) searches the word dictionary 6b for words that can constitute the input original text.

Information such as its part of speech and translation is also requested.

The analysis control unit (S53) performs a process of sending translation candidates to the syntactic analysis unit. The parsing unit (S54) uses parsing grammar 6c.
is used to analyze the syntax of candidate translations and generate structures that still contain semantic ambiguity in interpretation.

If the analysis fails, the process returns to the analysis control section. Semantic analysis section (
S55'') uses the semantic analysis grammar 6d to generate a Japanese structure that eliminates ambiguity in interpretation.The structure conversion unit (95
B) converts Japanese structures into English structures using conversion grammar 6e. If rejected, the process returns to the parsing section.

The syntax generation unit (S57) determines the English word order from the English structure according to the generative grammar 6f and converts it into a word string. The morpheme generation unit (858) completes the translated sentence by changing the ending of the word using the morpheme generation grammar 6g.

Each of the above-mentioned processing units outputs a signal to the editing control unit 4 at the time of starting processing.

Referring again to FIG. 4, when the editing control unit 4 detects the translation completion signal in step S1, the process proceeds to step S7, where it sends the translation candidates and auxiliary information obtained from the translation unit 5 to the display control unit 7 and displays the translated text. (Step S7). The display control unit 7 displays the translation candidate in the translation display area at a position corresponding to the input original text (on the left side of the screen), and if there is auxiliary information, displays the translation with high brightness or in reverse video so that the operator can It is made so that it can be identified.

From this state, when the edit key is pressed, the edit control unit 4
detects this in step S4, and performs editing processing corresponding to each key on the word or phrase indicated by the cursor at this time (step Sll).

For example, by operating the insert key 101, a character is inserted before the cursor position. By operating the delete key 102, the character string in the range indicated by the cursor is deleted. By operating the movement key 103, the range indicated by the cursor is moved. By operating the cancel key 104, the key lo
t, nullify the effect of LQ2.103. Modifications are made by operating the key 105, using auxiliary information to display other modification candidates for the word or phrase indicated by the cursor.

Note that the following functions can be realized using the function keys.

By operating the translated word display key 110, the translated word can be displayed for each word in the translated sentence. By operating the dictionary display key 111, a dictionary can be displayed for the word in the current sentence. New words and phrases can be registered by operating the dictionary display key 112. By operating the dictionary deletion key 113, words and phrases registered through dictionary registration can be deleted. Partial translation key 1
By the operation 14, when the translation fails, the partial translation is displayed. By operating the different interpretation key 118, a character string for which the interpretation is to be changed is specified, all interpretations for that character string are displayed, and by selecting one of them, the translation can be performed again using the selected interpretation.

Further, as cursor keys, cursor movement keys 130 move the cursor in each direction, and the cursor moves. Unit switch °praise key 131 to switch 11th place
, an inter-area movement key 132 for moving the cursor between display areas, and an enlargement/reduction key for reducing the size of the cursor in units of characters or in units of words.

Therefore, the operator can interactively obtain a translated text while inputting the original text and editing the translated text as appropriate.

In the system configured as described above, another interpretation of editing processing will be explained in detail with reference to the flowchart shown in FIG. First, when the operator looks at the original text and translated text displayed on the display section 8 and finds that the original text has not been interpreted as the operator intended, the operator presses the different interpretation key 118 (801), and the editing control section 4 , waits for the operator to specify a character string to be interpreted differently via the input unit 1. Operator presses cursor key 130
, specify the character string that you want to interpret differently, and press the different interpretation key [18] again (802). The editing control unit 4 instructs the analysis control unit (S53) to perform another interpretation (803). The analysis control unit (353) searches the stored series development table (804).

Here, the series expansion table will be explained. The series development table is stored in the analysis control section (S53) and is configured as shown in FIG. This is an example of a translation of the original sentence, ``The Buddha bought Japan today.''

There are two syntactically valid clause break positions for "Kono Nihon." They are ``Now 7 Japan'' and ``Today 7 Books.'' The analysis control unit (353) stores these two interpretations for each clause in the series expansion table. “■I am,■now,■today,0
■I bought a ■book about Japan.■
~■ are clause numbers assigned sequentially from the beginning of the sentence, and these clause numbers indicate each clause. The connection destination number in the right column indicates the clause number of the connection destination. For example, “■Ima” has the destination number ■, so it shows that it connects to “■Japan” 2.The series number in the middle column is the maximum number of phrases in a certain series. The analysis control unit (S53) performs a search using this series number.

For example, as shown in Figure 8, the translation of the original sentence “I bought water today” is “I bought water today.”
Now,'' is assumed to have become ``now.''Here, the system interprets ``Today Wednesday'' as ``Now/Japan.'' The operator operates the cursor keys 130 and instructs r now'j3$'4J. Based on this instruction, the analysis control unit (853) searches the series development table shown in FIG. 7, and finds that the series numbers for "Now" and "Japan" are (2) and (3), respectively. All interpretations regarding these sequence numbers (2) and (3) are extracted. As a result, the lines ``■-■'' are extracted and displayed on the display section 8 as shown in FIG.

Returning to FIG. 6 again, the analysis control unit (953) searches the series expansion table (604) to find out which series number the specified character string corresponds to, and extracts all interpretations regarding that series number. Here, if there is no alternative interpretation for the specified string (605), <No alternative interpretation candidates.

> message is displayed (609), and the alternative interpretation process is ended. On the other hand, if there is another interpretation (805)
, the series is displayed on the display unit 8 (60B).

The operator looks at the displayed alternative interpretation candidates, selects one appropriate interpretation of the original text, and designates it via the input unit 1 (807). The analysis control unit (S53) re-translates with this specified interpretation (608).

If you select and specify "Today's 7 books" in Figure 9, it will be retranslated,
As shown in Figure 10, the translated sentence is “I bout ab
ook today.'' In this way, another interpretation process is performed.

In the above embodiment, the case where the interpretation differs depending on the break position of the phrase has been explained, but in addition to this, the interpretation may differ depending on the pronunciation, part of speech, etc.

For example, if there are two interpretations of the original sentence ``I went.'' depending on the different pronunciations of ``I went,'' the analysis control unit (S53) stores it in the series expansion table along with information on the same series reading. Just do it. Furthermore, even in cases where the interpretation of parts of speech is different, the analysis control unit (S53) may similarly store a plurality of clauses with the same sequence number in the sequence expansion table together with part of speech information.

As described above, the operator can instruct a different interpretation only for a certain character string, and can obtain an appropriate translation in a short time. In addition, editing operations have become easier, as the conventional work of rewriting and retranslating the original text is no longer necessary.

[Effects of the Invention] According to the present invention, the editing operation of the operator is simplified by presenting the operator with multiple interpretations regarding only the character string specified by the operator and translating the text using the interpretation selected by the operator. , you can easily obtain appropriate translation results.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing the key arrangement of the input section 1 shown in FIG. 1, and FIG. 3 is a screen of the display section 8 shown in FIG. 1. Diagram showing the layout of
4 is a flowchart showing the flow of processing by the editing control section 4 shown in FIG. 1, FIG. 5 is a flowchart showing the flow of processing by the translation section 5 shown in FIG. 1, and FIG. 6 is the flowchart shown in FIG. 7 is a flowchart showing the flow of the separate interpretation process by the editing control unit 4 shown in FIG.
53), and FIG. 8 to FIG. 1O are diagrams showing display screens in another interpretation process. 1: Input section 2: Original text storage section 3: Translation storage section 4-: Editing control section 5: Translation section
6: Translation dictionary 7: Display control section 8: Display section 9: Printing section

Claims (4)

[Claims] (1) an input section, a morphological analysis section which divides the original text inputted through the input section into word units, searches a translation dictionary to check the grammatical attributes of each word, and each word examined by the morphological analysis section; Extract syntactically valid interpretations from word information,
a structure converting unit that examines case relations from the interpretation information extracted by the syntactic analyzer and converts the case relations into case relations of the target language; a translation generation unit that generates a translation, a display unit that displays the original text and the translation, and when editing the translation displayed on the display unit, another interpretation of a certain character string in the original text is instructed via the input unit. and searches the interpretation information stored in the syntax analysis section, extracts all the interpretation information regarding the specified character string, displays this on the display section, and selects the string selected via the input section. A machine translation system comprising: a separate interpretation control unit that sends interpretation information to the structure conversion unit and prompts retranslation. (2) The syntactic analysis unit extracts and stores a syntactically valid connection relationship between each word from the information of each word examined by the morphological analysis unit. machine translation system. (3) The syntactic analysis unit extracts a syntactically valid part of speech of each word from the information of each word examined by the morphological analysis unit and stores it. Machine translation system. (4) The machine according to claim 1, wherein the syntactic analysis unit extracts a syntactically valid pronunciation of each word from the information of each word examined by the morphological analysis unit and stores it. translation system.