JPH02281374A - Hyphenated word processing method in machine translation equipment - Google Patents

Abstract

PURPOSE:To obtain a correct translation result of a hyphonated word by translating only the words before and after a hyphen and synthesizing them, in the case of a word before the hyphen is a noun, and a word after the hyphen is a present participle or a past participle. CONSTITUTION:As a result of consulting a dictionary, in the case of a word before a hyphen of a divided composite word is a noun, and also, a word after the hyphen is a present participle or a past participle, a dictionary consulting morpheme analyzing part 11 executes a translation by only the words before and after the hyphen, based on dictionary consulting information which is held in accordance with the composite word. Subsequently, the result of this translation is allowed to correspond to the composite word and held in a memory. Next, a translation generating part 14 outputs a translated word, based on translation result information held in the memory in accordance with this composite word, and generates a translation. In such a way, even if the composite word is not registered in a user dictionary, a correct translation result can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to a method for processing hyphen compound words in a machine translation device.

<Prior art> In conventional machine translation devices, when an input sentence contains a compound word composed with a hyphen, the compound word with a hyphen is not included in the dictionary, so it is processed as a noun, and the compound word is processed as a noun. Divide the word before and after the hyphen, look it up in a dictionary for each word before and after the hyphen, and generate translations for each word.
The translated words were simply lined up and output.

Another method is to have the user register a hyphen compound word in a user dictionary in advance, and refer to the user dictionary when translating the compound word.

<Problems to be Solved by the Invention> As described above, the machine translation device processes the input hyphen compound word as a noun if it is not included in the dictionary. Therefore, for example, if a hyphen is used to combine a noun and a present participle or a noun and a past participle to form an adjective phrase, the translation will either fail and produce an incorrect result, or the original word will not be output. The correct translation result could not be obtained.

On the other hand, if hyphen compound words are registered in the user dictionary in advance, correct translation results can be obtained, but in this case, the user is burdened with the registration work, and also the user has to guess all the compound words that appear and do not forget them. It was very difficult to register.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a hyphen compound word processing method that can correctly translate a hyphen compound word in which a noun and a present participle or a noun and a past participle are compounded by a hyphen without registering the word in a dictionary in advance in a machine translation device. be.

Means for Solving the Problems> In order to achieve the above object, the present invention provides a dictionary lookup morpheme system that looks up each word in an input sentence in a dictionary and obtains a morpheme sequence of the input sentence from the dictionary lookup information. an analysis section; a syntactic analysis section that analyzes the syntax of the morpheme sequence obtained by the morphological analysis section using a dictionary and grammar rules; and a syntactic analysis section that converts the syntax obtained by the syntactic analysis section into a syntactic structure in the target language. In a machine translation device comprising a conversion unit and a translated sentence generation unit that generates a translated sentence from the dictionary lookup information according to the syntactic structure of the target language obtained by the conversion unit, the dictionary lookup morphological analysis unit If there is a hyphenated compound word synthesized by a hyphen in the sentence, the compound word is divided into words before and after the hyphen and kept in Kishibanana medium format, and furthermore, the word before the hyphen is a noun, In addition, if the word after the hyphen is a present participle or past participle, translate only the words before and after the hyphen according to the information obtained from the dictionary, match the translation result to the compound word, and write Jusushichi≠= ＃It is something to keep well.

<Operation> According to the present invention, if there is a compound word compounded with a hyphen in any input sentence, the dictionary lookup morphological analysis section divides the compound word into words before and after the hyphen, and divides the compound word into words before and after the hyphen. A dictionary search is performed for the compound word, and the obtained information is stored in correspondence with the compound word.

Then, as a result of the above dictionary lookup, if the word before the hyphen in the compound word is a noun and the word after the hyphen is a present participle or a past participle, translation is performed based on the information held only for the words before and after the hyphen. , the translation result is held in correspondence with the compound word.

When generating a translation of the compound word, the translated sentence generation section reads out the translation result held by the dictionary lookup section corresponding to the compound word and generates the translation.

<Example> Hereinafter, the present invention will be explained in detail with reference to illustrated embodiments.

FIG. 1 is a block diagram of an embodiment of a machine translation device to which the present invention is applied. 1 is the main CPU (central processing unit), 2 is the main memory 3 is the CRT (cathode ray tube), 4 is the keyboard, 5 is the translation module, 6 is the dictionary for translation, grammar rules, tree structure conversion rules, etc. This is the memory that stores it.

When the source language is input, the translation module 5
It translates it and outputs the target language.

That is, the source language input from the keyboard 4 is sent to the translation module 5 under the control of the main CPUI. The translation module 5 uses the dictionary, grammar rules, tree structure conversion rules, etc. stored in the memory 6 to translate the input source language into the target language as will be detailed later. The results are temporarily stored in the main memory 2 and displayed on the CRT 3.

The machine translation performed by the translation module 5 has analysis levels as shown in FIG. When the source language is input in the upper left of the figure, dictionary lookup at level L1, morphological analysis at level L2, syntactic analysis at level L3, etc.
The analysis continues.

Machine translation can be broadly divided into the following two types depending on the level of analysis. One is to analyze concepts (called intermediate language) that do not depend on either the source language or the target language at level L6, and from there, generate context at level L7, generate meaning at level L8, generate syntax at level L9, Level L
It is a pivot method that generates the target language by proceeding with morpheme generation and generation of IO.

The other is the above-mentioned level L2 morphological analysis, level L
The internal structure of the source language is obtained by performing any of the syntactic analysis at level 3, the semantic analysis at level 4, and the context analysis at level 5, and then the internal structure of the target language at the same level as the internal structure of the source language obtained is analyzed. It is a transfer method that generates the target language after converting it to the internal structure.

Here, each of the above analyzes will be explained.

・Dictionary lookup, morphological analysis - Look up the dictionary in the memory 6 shown in Figure 1, divide the input sentence into each morpheme string (word string), obtain grammatical information such as part of speech and translation for each word, and further , analyze tense, person, number, etc.

- Syntactic analysis - Determines the structure of the sentence (structural analysis tree), such as dependencies between words, as will be described later.

・Semantic analysis - Distinguish between semantically correct and incorrect results from multiple syntactic analysis results.

・Context analysis – understand the topic and eliminate omissions and ambiguity.

It is assumed that the translation module 5 used in this invention performs analysis up to at least level L3 parsing. That is, the translation module 5 of this embodiment is considered to be a translation module consisting of parts as shown in FIG.

FIG. 4 is a block diagram of the translation module 5. Also, Figures 5 to 9 are ['rhis is a pen
, J are to be translated into Japanese, the contents of each buffer A to E shown in FIG. 4 will be exemplified.

The operation of English-Japanese translation will be explained below using FIGS. 3 to 9.

First, the read original text is stored in the buffer AK of FIG. 4, as shown in FIG. Under the control of the translation CPU 15 based on the translation program 16, the dictionary lookup morphological analysis unit 11 shown in FIG. 3 obtains necessary information for each word using the dictionary in the memory 6 according to the original text stored in the buffer 7A, The data is stored in buffer B in FIG. Here, Th1
Although s is a multipart speech word, it is uniquely determined by the syntactic analysis unit 12 shown in FIG. 3 below. In the syntax analysis unit 12, memo I)
In accordance with the dictionary and grammar rules of No.-6, a structural analysis tree showing the relationships between words is determined as shown in FIG. 7, and stored in buffer C in FIG.

This structural analysis tree is determined as follows.

That is, from the grammar rules in the memory 6, the following rules are obtained: sentence → subject, predicate subject → noun phrase predicate → verb, noun phrase noun phrase → pronoun noun phrase → article, and noun. For example, the first rule indicates that "a sentence is made up of a subject and a predicate." Thereafter, a structural analysis tree is determined according to this rule.

In the conversion unit 13 of FIG. 3, the structure of the parse tree (see FIG. 7) for the input English sentence is created as shown in FIG. It is converted into the structure for Japanese sentences like this. Then, the obtained result is stored in the buffer ID shown in FIG. The translated sentence generation unit 14 in FIG.
Store it in the buffer IE shown in Figure 4 in the Japanese format as shown in the figure. This Japanese sentence "This is a pen." is output from the translation module 5, stored in the main memory 2, and displayed on the CRT 8.

Next, the processing operation of hyphen compound words by the dictionary lookup morphological analysis unit 11 according to the present invention is shown in the flowchart of FIG.

The processing operation of the dictionary lookup morphological analysis unit 11 will be explained below according to this flowchart.

In step S1, it is determined whether the input word includes a hyphen. If the compound word contains a hyphen, the process proceeds to step S3, and if not, the process proceeds to step S2. In steps 53 to S7, the hyphenated compound word is divided into words before and after the hyphen, a dictionary lookup is performed for each word, and the obtained results are stored in buffer B in correspondence with the compound word.

Then, as a result of the dictionary lookup in step S8, if the word before the hyphen is a noun and the word after the hyphen is a present participle or a past participle, the process advances to step S9; otherwise, the process advances to step S14. In step S9, the contents of buffer B holding the result of the dictionary lookup are saved to another buffer. Based on the dictionary lookup result saved in another buffer in step SIO, the order of the words before the hyphen and after the hyphen is replaced, translation processing is performed using only these two words, and the translation result is translated into the compound word in step SIO. It is held in buffer B in correspondence with . Further step 512
Returns the dictionary lookup results that were previously saved to the buffer IB.

On the other hand, in step S2, a normal dictionary lookup is performed, and in step Sta, the results are associated with words and stored in buffer 1.
Hold at B.

In step S14, it is determined whether the above-mentioned processing has been completed for all words in the input sentence, and if it has not been completed, the process returns to step S1, and if it has been completed, the processing of the dictionary lookup morphological analysis unit 11 is ended. . And translated sentence generation part 1
When generating a translation of a hyphen compound word in step 4, the translation is output based on the translation result information held in the buffer 7B corresponding to the compound word, and a translated sentence is generated.

FIG. 11 shows an example of a sentence including a hyphen compound word inputted to the machine translation device of this embodiment and a Japanese sentence outputted from this machine translation device.

The input hyphen compound word [stem-mounted
J is divided before and after the hyphen and looked up in the dictionary,
By transposing the order and translating only these two words, we obtain the translated word "equipped with a stem", and then convert the translated word into the compound word "s
It is held in buffer B in correspondence with tem-mounted J.

The translated sentence generation unit 14 generates a translated sentence of the input sentence according to the translation result information, and as a result, obtains the Japanese sentence "device equipped with a stem."

<Effects of the Invention> In a machine translation device to which the hyphen compound word processing method of the present invention is applied, the input sentence contains a compound word that contains ) and a hyphen, and a noun and a present participle or a noun and a past participle are compounded by a hyphen. Correct translation results can be obtained and efficient translation can be achieved without registering the compound word in the user dictionary.

[Brief explanation of drawings]

Figure 1 is a block diagram of an embodiment of a machine translation device to which this invention is applied, Figure 2 is an explanatory diagram of translation levels in automatic translation, Figure 3 is a block diagram of a translation module in the above embodiment, and Figure 5. 9 to 9 are diagrams showing examples of the contents of each buffer in FIG. 4, FIG. 10 is a flowchart showing the operation of processing hyphen compound words, and FIG. 11 is an example of input words to the machine translation device in the present invention and their It is a figure which shows the example of an output word (translation word) for. 1... Main CPU 2... Main memory 3
...CRT 4...Keyboard 5...Translation module 6...Memory 11...Dictionary lookup morphological analysis unit 12...Syntax analysis unit 13 Conversion unit 14...Translated sentence generation unit 15...・Translation CPU 16... Translation program A to E... Buffer 4 string sentence Th1s S en ・ , (-/) Finished A -1-,!文'ba'tsubu ryo

Claims (1)

[Scope of Claims] 1. A dictionary lookup morphological analysis unit that looks up each word in an input sentence in a dictionary and obtains a morpheme sequence of the input sentence from the dictionary lookup information; and morphemes obtained by the morphological analysis unit. a syntax analysis unit that analyzes the syntax of the string using a dictionary and grammar rules; a conversion unit that converts the syntax obtained by the syntax analysis unit into a syntactic structure in a target language; and a target language obtained by the conversion unit. a translated sentence generation unit that generates a translated sentence from the dictionary lookup information according to the syntactic structure of For example, the compound word is divided into words before and after the hyphen, a dictionary lookup is performed for each of the two divided words, the obtained dictionary lookup information is held in correspondence with the compound word, and the dictionary As a result of the lookup, if the word before the hyphen that is divided into the compound word is a noun, and the word after the hyphen is a present participle or a past participle, based on the dictionary lookup information held corresponding to the compound word. to translate only the words before and after the hyphen, and store the translation result in correspondence with the compound word, and the translated sentence generation unit outputs the translation result as a translation of the compound word. Hyphen compound word processing method in machine translation equipment.