JPH0582626B2 - - Google Patents

Description

TECHNICAL FIELD The present invention relates to a machine translation system, and particularly to a machine translation system suitable for clarifying the usage of adjuncts in Japanese sentences.

BACKGROUND OF THE INVENTION Research and development is actively underway on machine translation systems that use computers to translate sentences from Japanese to a foreign language (especially English) or from a foreign language to Japanese. If the translation is limited to texts in a specific field such as technical documents and manuals, and it is assumed that the user will intervene during the translation or check the output text, it will be possible to put it into practical use. coming.

That is, when a Japanese sentence is translated into a foreign language by a machine translation system, the Japanese language is analyzed (morphological analysis and syntactic analysis) using a word dictionary and a grammar dictionary. However, since this analysis uses only part-of-speech information of words, there is ambiguity in the usage of adjuncts. Therefore, in conventional machine translation systems, users are allowed to decide how to use attached words.

Conventionally, the following two methods have been used to allow the user to determine the usage of adjuncts.

Replace ancillary words with different expressions and show them to the user so that the usage is clear.

Show users multiple uses for each adjunct.

However, with this method, it is difficult to replace all the usages of adjunct words with different expressions, and the method also makes it difficult for non-expert users to understand the subtle differences between multiple usages of the same adjunct. The problem is that it is difficult to judge.

Purpose It is an object of the present invention to provide a machine translation system that solves the drawbacks of the prior art as described above and allows even non-specialist users to clarify the usage of adjuncts in Japanese.

Configuration The above-mentioned object of the present invention is to provide a machine translation system having dictionary documents such as a word dictionary, an adjunct dictionary, a grammar dictionary, etc., a morphological analysis device, a syntactic analysis device, an input device, a display device, etc. An adjunct word extraction unit that extracts adjunct words from morphologically analyzed Japanese sentences, an adjunct usage correspondence table that shows the presence or absence of adjunct usage types for each adjunct word, and an adjunct usage correspondence table for each adjunct usage type. has an attached word usage dictionary having explanations of its usage and example sentences including attached words of its usage, and searches the attached word usage correspondence table for the attached words extracted by the attached word extracting unit and extracts the attached words. detecting the type of usage that the attached word has, and according to the detected type of usage, reading an explanation of the usage and an example sentence including the attached word of the usage from the dictionary of usage, and displaying it on the display device. , an adjunct word usage selection device configured to select the usage of the adjunct word based on the input device when the user indicates whether or not the displayed usage is acceptable; the usage of the adjunct word selected by the adjunct word usage selection device, and the morphological analysis result in which the usage of the adjunct word is set is input to the syntactic analysis device. achieved by the system.

Embodiments Hereinafter, embodiments of the present invention will be described in detail based on the drawings.

FIG. 2 is a block diagram showing a schematic configuration of a machine translation system according to an embodiment of the present invention.

In Figure 2, 1 is an input device, 2 is a morphological analyzer, 3 is a dictionary for analyzing Japanese, and 4 is a morphological analyzer.
Reference numeral denotes an adjunct word usage selection device, 5 a syntax analysis device, 6 a conversion device, 7 a syntax generation device, 8 a dictionary for parsing a foreign language, 9 a morpheme generation device, and 10 a display device. Note that dictionary document 3 is a word dictionary,
It has a grammar dictionary, ancillary word dictionary, etc.

The Japanese language to be translated is input from the input device 1. The input Japanese is divided into words by the morphological analyzer 2 using the dictionary 3.
Morphological analysis is performed by determining each part of speech. Of the Japanese words that were morphologically analyzed,
Regarding subordinate words, the subordinate word usage selection device 4, which is a feature of the present invention, sequentially displays the usages of the subordinate words on the display device 10.

Note that the usage of adjuncts displayed is not that multiple usages of each adjunct are displayed all at once, but for each type of usage, different adjuncts with that usage type are displayed along with their example sentences. . for example,
The case particle ``no'' has usages such as ``subject'', ``restriction'', and ``place'', but these usages are not displayed all together under the case particle ``no'', but rather as ``subject''.
, ``Limited'', and ``Location'', and are displayed with other adjuncts and the same usage of other words. That is, the display is performed not by the type of attached word but by the type of usage.

When the user selects an appropriate one from among the usages of the accessory word displayed on the display device 10, the accessory word usage selection device 4 learns the corresponding translation, if any. The auxiliary word usage selection device 4 will be explained in detail later.

Next, the syntactic analysis device 5 uses the dictionary document 3 to analyze the structure of the sentence, such as how words are connected. However, since the adjunct word has already been analyzed by the adjunct word usage selection device 4, it is not analyzed by the syntactic analysis device 5.

The analysis result by the syntax analysis device 5 is a Japanese intermediate representation, and this Japanese intermediate representation is output to the conversion device 6. The format of the Japanese intermediate expression may be either a parse tree structure or a case structure (focusing on a verb and indicating the case of each word related to it).

The conversion device 6 converts the Japanese intermediate representation into production rules (IF-THEN format rules).
Once the pattern matching between the input data and the IF part is achieved, it is converted into an intermediate representation of the target language (foreign language) using the rules that execute the actions in the THEN part.

The syntax generation device 7 and the morpheme generation device 9 perform respective processing on the intermediate representation of the target language using the dictionary document 8, and the translated sentence in the target language is displayed on the display device 10. Note that each generation process is a process opposite to the analysis process.

FIG. 1 is a block diagram showing in detail the structure of the adjunct usage selection device 4 in FIG. 2. As shown in FIG.

In FIG. 1, 4a is an adjunct word extraction unit, 4b is a control unit, 4e is an adjunct word usage correspondence table, 4f is an adjunct word usage dictionary, and 4c is a line pointer storage unit that stores the line number of the adjunct word usage correspondence table 4e. , 4d is a column pointer storage unit that stores column numbers of the adjunct usage correspondence table 4e.

FIG. 3 is a diagram showing an example of the contents of the adjunct word dictionary of the dictionary document 3 in FIG. 2.

As shown in FIG. 3, the adjunct word dictionary has an "adjunct word notation" column, a "morpheme, parsing information" column, and a "line pointer" column. The "Row Pointer" column is a column that indicates which line number of the attached word usage correspondence table 4e the attached word written in the "Adjunct Word Description" column corresponds to.

Figure 4 is ancillary word usage correspondence table 4 in Figure 1.
It is a figure which shows an example of the content of e.

As shown in Figure 4, attached word usage correspondence table 4e
, the adjunct words are arranged in correspondence with the row numbers, and the types of usage of the adjunct words are arranged in correspondence with the column numbers. The usage of adjuncts is divided into detailed categories, such as whether the adjunct represents a subject, an object, a cause, etc. '1' at the intersection of the row number and column number means that the usage shown in the column exists for the adjunct shown in that row; '0' means the usage shown in the column exists for the adjunct shown in that row. about,
This means that the usage indicated in the column does not exist.

Figure 5 shows the attached word usage dictionary 4f in Figure 1.
It is a figure which shows an example of the content.

As shown in FIG. 5, the attached word usage dictionary 4f is
"Column pointer" field, "Usage explanation" field, "Foreign language notation"
column, and “Foreign Language Information” column. In addition, in the figure,
Words marked with are used in the same way as the adjunct in question (adjuncts and other words);
The × mark means that there is no parallel translation.

In the "Column pointer" column, the attached word usage correspondence table 4e
Indicates the number corresponding to the column number. The "Usage Explanation" column specifically describes the usage of adjunct words (for example, representing a subject, representing a place, etc.) along with example sentences for each usage. In the "Foreign language notation" column, the foreign language translation corresponding to the usage is listed. The "Foreign language information" column contains information regarding the translation listed in the "Foreign language notation" column.

Hereinafter, the operation of the adjunct usage selection device 4 will be explained using FIGS. 1, 3, 4, and 5, taking the case where the input sentence is "in Tokyo..." as an example.

The following sentence, which has been morphologically analyzed by the morphological analysis device 2, is input to the adjunct word extraction unit 4a.

((Tokyo ((noun),…,)(de(particle),…,
((14)))) Information (14) in the above morpheme (de) corresponds to the contents of the "line pointer" column listed in the adjunct word dictionary shown in Figure 3, that is, the corresponding adjunct usage. This is the row number of Table 4e.

The adjunct word extraction unit 4a examines the part of speech of each morpheme in the input sentence subjected to morphological analysis as described above, and extracts an adjunct word (a word corresponding to a particle or an auxiliary verb).
In the example sentence above, "Tokyo" is not extracted because its part of speech is a "noun." Next, when the part of speech of "de" is examined, it is an adjunct "particle," so "de" is extracted, and the control unit Sent to 4b.

The control unit 4b takes out the pointer to the adjunct usage correspondence table 4e, that is, the line number (14), from the information on the adjunct word "de" and sets it in the line pointer storage part 4c. Also, the counter of the column pointer storage section 4d is set to '1'.

Next, the control section 4b searches the adjunct term usage correspondence table 4e while incrementing the counter in the column pointer storage section 4d by '1' based on the row number set in the row pointer storage section 4c. At this time, if the value at the intersection of the row and column in the attached word usage correspondence table 4e is '0', ignore it, and continue until the value at the intersection of the row and column becomes '1'.
The counter in the column pointer storage section 4d continues to be incremented by '1'.

If the value '1' at the intersection of the row and column is detected, the attached word usage dictionary 4f is searched using the counter number in the column pointer storage unit 4d at that time as a pointer. In the case of the particle "de" in line number 14, first, column number 5 (subject) is '1', and the adjunct usage dictionary 4f is searched using column number 5 as a pointer.

The control unit 4b reads out the item (subject) indicated by column number 5 from the attached word usage dictionary 4f, and outputs it to the output device 10.

The user inputs information that means "YES" from the input device 1 when the usage of the attached word shown on the output device 10 is the intended one, and when it is not the intended usage, the user inputs information that means "NO". Enter the information you want.
In the case of the example sentence above, it is indicated by column number 5 (subject)
Since the intended usage (which is currently under investigation at Omori Police Station) is indicated in the field, the user enters information that means "YES."

When the user inputs information meaning "YES", the control unit 4b adds the name of the "subject" to the information of "de" and the bilingual translation in the "foreign language notation" column obtained from the attached word usage dictionary 4f. and the information in the "Foreign language information" column is added.

The result is as follows.

(((particle)(subject)(...)(14))) If the input from the input device 1 is information that means "NO", the control unit 4b further sets the counter of the column pointer storage unit 4d to ' Increment by 1' and continue searching the attached word usage dictionary 4f. Thereafter, the same operations as above are performed.

In this way, by displaying the usage of adjuncts to the user by usage rather than for each adjunct, it is possible to display the usage of adjuncts other than the adjunct of interest for a certain type of usage, as well as other adjuncts. Since the same usage of words can be viewed at the same time, even non-expert users can easily judge subtle differences in the usage of the same word between adjuncts, eliminating ambiguity in the usage of adjuncts. , can be clarified.

Note that the morphological analysis result in which the selected usage etc. are set is output to the syntactic analysis device 5 and becomes an input to the syntactic analysis device 5. Since the syntactic analysis device 5 indicates the usage of the adjunct word in the morphological analysis result of the input, the burden of analyzing the usage of the adjunct word, which was conventionally performed by the syntactic analysis device, is eliminated.

FIG. 6 is a block diagram showing an adjunct usage selection device 4 according to another embodiment of the present invention.

This embodiment is almost the same as the embodiment shown in FIG. 1, but the difference is that a temporary storage section 4g is added.

That is, in the embodiment shown in FIG.
When '1' is detected in the search of the accessory word usage correspondence table 4e, the corresponding "usage explanation" column in the accessory word usage dictionary 4f is displayed each time.
On the other hand, in the embodiment shown in FIG. 6, the rows corresponding to the adjunct word of interest are searched at once, and the column numbers of '1' detected in the process are sequentially searched.
The rows are stored in the temporary storage section 4g, and when the search for that row is completed, the "usage explanation" column in the attached vocabulary usage dictionary 4f corresponding to the column number stored in the temporary storage section 4g is displayed in descending order of column number. go.

In this way, in this embodiment, after searching all the accessory word usage dictionaries 4f, the usage of the selected accessory words is sequentially displayed to the user. can be speeded up.

Effects As explained above, according to the machine translation system of the present invention, even users other than experts can clarify the usage of subordinate words in Japanese.

[Brief explanation of the drawing]

1 and 6 are block diagrams of adjunct usage selection devices according to different embodiments of the present invention, and FIG.
The figure is a block diagram of a machine translation system to which Figure 1 or Figure 6 is applied, Figure 3 is a diagram showing the contents of the ancillary word dictionary in the dictionary document 3 in Figure 2, and Figure 4 is the same as Figure 1 or Figure 6. FIG. 5 is a diagram showing the contents of the accessory word usage correspondence table in FIG. 6, and FIG. 5 is a diagram showing the contents of the accessory word usage dictionary in FIG. 1 or FIG. 6. 4: Adjunct word usage selection device, 4a: Adjunct word extraction unit, 4b: Control unit, 4c: Line pointer storage unit, 4
d: Column pointer storage section, 4e: Attached word usage correspondence table, 4f: Attached word usage dictionary, 4g: Temporary storage section.

Claims (1)

[Claims] 1 Dictionaries such as word dictionaries, subsidiary word dictionaries, grammar dictionaries, etc., morphological analysis devices, syntactic analysis devices, input devices,
In a machine translation system that has a display device, etc.
an adjunct word extraction unit that extracts an adjunct word from the Japanese sentence morphologically analyzed by the morphological analysis device;
An adjunct word usage correspondence table that shows whether there is a type of adjunct usage for each adjunct word, an explanation of the usage for each type of adjunct usage, and a dictionary of adjunct words that includes example sentences that include the adjunct word. and detecting the type of usage of the extracted accessory word by searching the accessory word usage correspondence table for the accessory word extracted by the accessory word extraction unit, and detecting the type of usage of the extracted accessory word, and detecting the type of usage of the extracted accessory word. read out an explanation of the usage and an example sentence including the accessory word of the usage from the accessory word usage dictionary and display it on the display device, and when the user instructs whether or not the displayed usage is appropriate using the input device; , an accessory word usage selection device configured to select the usage of the accessory word based on this is provided, and the usage of the accessory word selected by the accessory word usage selection device is set in the result of the morphological analysis. , a machine translation system characterized in that the morphological analysis result in which the usage of the attached word is set is input to the syntactic analysis device.