JPH05257967A - Machine translator - Google Patents

Machine translator

Info

Publication number
JPH05257967A
JPH05257967A JP4053124A JP5312492A JPH05257967A JP H05257967 A JPH05257967 A JP H05257967A JP 4053124 A JP4053124 A JP 4053124A JP 5312492 A JP5312492 A JP 5312492A JP H05257967 A JPH05257967 A JP H05257967A
Authority
JP
Japan
Prior art keywords
dictionary
compound word
sentence
word
translated
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
JP4053124A
Other languages
Japanese (ja)
Inventor
Junichi Ikoma
順一 生駒
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
KOBE NIPPON DENKI SOFTWARE KK
Original Assignee
KOBE NIPPON DENKI SOFTWARE KK
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by KOBE NIPPON DENKI SOFTWARE KK filed Critical KOBE NIPPON DENKI SOFTWARE KK
Priority to JP4053124A priority Critical patent/JPH05257967A/en
Publication of JPH05257967A publication Critical patent/JPH05257967A/en
Withdrawn legal-status Critical Current

Links

Landscapes

  • Machine Translation (AREA)

Abstract

PURPOSE:To improve an analytical method for composite words in a machine translator. CONSTITUTION:An original analyzing part 1-1 analyzes the grammatical structure or meaning relation of an original based upon the information of a translating dictionary 1-4 and transfers the analyzed result to a composite word processing part 1-2. The processing part 1-2 refers to the contents of a dictionary and the analyzed result, and at the time of judging a composite word, executes the processing of the composite word and transfers the processed result to a translated sentence forming part 1-3. The forming part 1-3 outputs a translated sentence matched with the grammer of output side language based upon the analytical result obtained from the processing part 1-2.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【産業上の利用分野】本発明は自然言語で書かれた文を
他の自然言語に翻訳する機械翻訳装置において、複合語
の辞書登録の方法と原文解析方法を改良して、複合語の
辞書登録による弊害をなくすことにより、より高品質な
訳文を得ることができる機械翻訳装置に関するものであ
る。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is a machine translation device for translating a sentence written in a natural language into another natural language. The present invention relates to a machine translation device that can obtain a translated text of higher quality by eliminating the harmful effects of registration.

【0002】[0002]

【従来の技術】現在の機械翻訳の技術では、複合語を辞
書に登録しない場合、訳語の選択がうまくできなかった
り、ぎこちない訳になることが多い。ところが、辞書に
複合語が登録されると、その文字列と一致する文字列が
原文に出現した場合は、常にその複合語が優先して辞書
引きされる。従って、本来別々に辞書引きするはずの語
を複合語として辞書引きしてしまうことにより入力文の
解析に失敗することがあった。例えば、「high s
peed」という形容詞と名詞からなる複合語を「高
速」と訳語で名詞として辞書に登録したとする。原文が
「high speed system」の場合は「高
速システム」となりうまく翻訳できるが、原文が「ve
ry high speed system」の場合
は、「very」という副詞が形容詞しか修飾しない性
質を持つため、うまく翻訳することができない。このよ
うに場合は、辞書からその複合語を削除すれば、「ve
ry」は形容詞「high」を修飾できるためうまく翻
訳することができるが、それでは「high spee
d system」における「高速」という訳語を得る
ことができない。
2. Description of the Related Art In the current technology of machine translation, when a compound word is not registered in a dictionary, it is often difficult to select a translated word or the translation becomes awkward. However, when a compound word is registered in the dictionary, when a character string that matches the character string appears in the original sentence, the compound word is always prioritized in the dictionary. Therefore, there are cases in which the analysis of the input sentence fails because the words that should originally be dictionary-divided separately are dictionary-divided. For example, "high s
It is assumed that a compound word composed of the adjective "peed" and a noun is registered in the dictionary as a noun with the translated word "high speed". When the original text is "high speed system", it becomes "high speed system" and can be translated well, but the original text is "ve
In the case of “ry high speed system”, the adverb “very” has the property of modifying only adjectives, and thus cannot be translated well. In this case, if the compound word is deleted from the dictionary, "ve
"ry" can be translated successfully because it can modify the adjective "high", but then "high speed"
The translated word "high speed" in "d system" cannot be obtained.

【0003】[0003]

【発明が解決しようとする課題】従来の機械翻訳装置で
は、本来別々に辞書引きされるはずの語を複合語として
辞書引きしたことにより入力文の解析に失敗した場合、
利用者は原文を修正するか、複合語辞書を削除するか、
語切り指定を行うしか、正しい訳を得る方法がなく、修
正に手間と工数がかかるという欠点があった。
In the conventional machine translation device, when the input sentence analysis fails due to the dictionary lookup of the words that should originally be dictionary lookups as compound words,
The user can either modify the source text, delete the compound dictionary,
There is a disadvantage that there is no way to get the correct translation except to specify the word cut, and it takes time and effort to correct it.

【0004】[0004]

【課題を解決するための手段】本発明の機械翻訳装置
は、ある言語を解析して別の言語に翻訳する機械翻訳装
置において、単語の訳語と複合語になるための条件およ
び前記条件が満たされた場合の前記複合語の処理方法を
記述した辞書を備え、入力文を前記辞書を用いて文法的
構造,意味等を解析する解析手段と、前記解析された文
から前記辞書における複合語の条件を参照して検出され
た前記複合語を前記処理方法に従って処理する複合語処
理部と、前記複合語処理部で処理された文から訳文を生
成する訳文生成部を備えたことを特徴とする。
A machine translation apparatus of the present invention is a machine translation apparatus that analyzes a language and translates it into another language. The condition for becoming a compound word and a compound word and the condition are satisfied. A dictionary describing the processing method of the compound word in the case of being processed, and analyzing means for analyzing the grammatical structure, meaning, etc. of the input sentence using the dictionary, and the compound word in the dictionary from the analyzed sentence. A compound word processing unit that processes the compound word detected by referring to a condition according to the processing method, and a translation sentence generation unit that generates a translated sentence from the sentence processed by the compound word processing unit. ..

【0005】[0005]

【実施例】次に本発明について図面を参照して説明す
る。
The present invention will be described below with reference to the drawings.

【0006】図1は本発明の一実施例の構成図である。
本発明は、原文解析部1−1、複合語処理部1−2、訳
文生成部1−3、翻訳用辞書1−4、とから構成され
る。
FIG. 1 is a block diagram of an embodiment of the present invention.
The present invention includes an original sentence analysis unit 1-1, a compound word processing unit 1-2, a translated sentence generation unit 1-3, and a translation dictionary 1-4.

【0007】原文解析部1−1は、翻訳用辞書1−4の
情報をもとに、原文1−01の文法的構造や、意味関係
を解析し、結果を複合語処理部1−2にわたす。複合語
処理部は、辞書内容と解析結果を参照して、複合語と判
断した場合に複合語の処理を行い、結果を訳文生成部1
−3に渡す。訳文生成部1−3は複合語処理部1−2か
ら得た解析結果から、出力側言語の文法に沿った訳文1
−25を出力する。
The original sentence analysis unit 1-1 analyzes the grammatical structure and the semantic relationship of the original sentence 1-01 based on the information of the translation dictionary 1-4, and the result is sent to the compound word processing unit 1-2. I will give you. The compound word processing unit refers to the dictionary contents and the analysis result, processes the compound word when it judges that it is a compound word, and outputs the result to the translation generation unit 1.
-3. The translation generation unit 1-3 uses the analysis result obtained from the compound word processing unit 1-2 to translate the translation 1 according to the grammar of the output language.
Output -25.

【0008】図2は本発明を使用した際の翻訳の流れを
示している。まず原文1−01が入力されて、原文解析
部1−1にわたされる(2−1)。原文解析部1−1は
翻訳用辞書1−4を参照したうえで、原文を解析して、
解析結果を複合語処理部1−2に渡す(2−2)。複合
語処理部1−2は、翻訳用辞書1−4を参照した結果、
その語に複合語情報が記述されていれば、その条件に一
致しているかどうか判断する(2−4)。一致していれ
ば、複合語として辞書に記述された処理を行う(2−
5)。その結果を訳文生成部1−3に渡し訳文を出力す
る(2−6)。一致していない場合は、そのまま通常の
語として結果を訳文生成部1−3に渡し訳文を出力する
(2−6)。
FIG. 2 shows the flow of translation when using the present invention. First, the original sentence 1-01 is input and passed to the original sentence analysis unit 1-1 (2-1). The source sentence analysis unit 1-1 analyzes the source sentence by referring to the translation dictionary 1-4,
The analysis result is passed to the compound word processing unit 1-2 (2-2). The compound word processing unit 1-2 refers to the translation dictionary 1-4,
If the compound word information is described in the word, it is judged whether or not the condition is met (2-4). If they match, the processing described in the dictionary as a compound word is performed (2-
5). The result is output as a translated sentence to the translated sentence generation unit 1-3 (2-6). If they do not match, the result is output as a normal word as it is to the translated text generation unit 1-3, and the translated text is output (2-6).

【0009】具体例を上げて説明する。仮に英語から日
本語への翻訳とし、「high speed」という複
合語の情報が「speed」の辞書に記述されていると
する。辞書には、図3のように訳語が複数記述されてい
る。最初の訳語には条件が記述されておらず、この訳語
が通常の訳語となる。2番目の訳語には複合語となるた
めの条件が記述されており、この条件に一致した場合に
のみ、原文解析部は訳語を選択し、複合語としての処理
を行う。
A specific example will be described. It is assumed that the translation is from English to Japanese and the information of the compound word "high speed" is described in the dictionary of "speed". A plurality of translated words are described in the dictionary as shown in FIG. No condition is described in the first translation, and this translation is the normal translation. The condition for becoming a compound word is described in the second translated word, and only when this condition is met, the original sentence analysis unit selects the translated word and performs the process as the compound word.

【0010】原文が「very high speed
system」である場合は「高速」の訳語の条件の
うち、1の条件は満たしているが、2の条件を満たして
いないので、訳語として「スピード」が選択され、最終
的に訳文「非常に高いスピードシステム」が得られる。
The original text is "very high speed"
In the case of “system”, the condition 1 of the translated word “high speed” is satisfied, but the condition 2 is not satisfied, so “speed” is selected as the translated word, and finally the translated sentence “very "High speed system" is obtained.

【0011】原文が「high speed syst
em」である場合には、「高速」の訳語の1、2の条件
を満たすので、訳語として「高速」が選択される。この
際、辞書に記述してある、複合語になったときの処理を
行い「high」の情報を削除しておく。かれは「高
速」という訳語に、「high」の意味が含まれるから
である。このようにして最終的に訳文「高速システム」
が得られる。最後に、複合語処理後の解析結果が訳文生
成部1−3にわたされ訳文が生成される(2−6)。
The original sentence is "high speed system".
In the case of "em", since the conditions 1 and 2 of the translated word "high speed" are satisfied, "high speed" is selected as the translated word. At this time, the processing when a compound word is described in the dictionary is performed and the information of "high" is deleted. This is because he includes the meaning of "high" in the translated word "high speed." In this way the final translation "High speed system"
Is obtained. Finally, the analysis result after the compound word processing is passed to the translated sentence generation unit 1-3 to generate a translated sentence (2-6).

【0012】[0012]

【発明の効果】以上説明したように本発明は、機械翻訳
装置における、複合語の辞書登録、およびその解析方法
を改良することにより、利用者による語切り、訳文の修
正を行うことなしに正しい訳文の生成を可能にし、修正
の手間を省き翻訳能率を向上させる効果がある。
As described above, according to the present invention, by improving the dictionary registration of a compound word and the analysis method thereof in the machine translation device, it is possible to correct the word without the user cutting the word and correcting the translated sentence. It is possible to generate a translated sentence, save the trouble of correction, and improve the translation efficiency.

【図面の簡単な説明】[Brief description of drawings]

【図1】本発明の一実施例を示す構成図である。FIG. 1 is a configuration diagram showing an embodiment of the present invention.

【図2】本発明を使用した場合の翻訳の流れ図である。FIG. 2 is a flow chart of translation when using the present invention.

【図3】本発明を使用した場合の辞書の内容例を示す図
である。
FIG. 3 is a diagram showing an example of contents of a dictionary when the present invention is used.

【符号の説明】[Explanation of symbols]

1−1 原文解析部 1−2 複合語処理部 1−3 訳文生成部 1−4 翻訳用辞書 1−01 原文 1-1 Original sentence analysis unit 1-2 Compound word processing unit 1-3 Translation sentence generation unit 1-4 Translation dictionary 1-01 Original sentence

Claims (1)

【特許請求の範囲】[Claims] 【請求項1】 ある言語を解析して別の言語に翻訳する
機械翻訳装置において、単語の訳語と複合語になるため
の条件および前記条件が満たされた場合の前記複合語の
処理方法を記述した辞書を備え、入力文を前記辞書を用
いて文法的構造,意味等を解析する解析手段と、前記解
析された文から前記辞書における複合語の条件を参照し
て検出された前記複合語を前記処理方法に従って処理す
る複合語処理部と、前記複合語処理部で処理された文か
ら訳文を生成する訳文生成部を備えたことを特徴とする
機械翻訳装置。
1. A machine translation apparatus for analyzing a language and translating it into another language, describes a condition for becoming a compound word and a compound word, and a method for processing the compound word when the condition is satisfied. And a means for analyzing the grammatical structure, meaning, etc. of the input sentence using the dictionary, and the compound word detected from the analyzed sentence by referring to the condition of the compound word in the dictionary. A machine translation device comprising: a compound word processing unit that processes according to the processing method; and a translated sentence generation unit that generates a translated sentence from the sentence processed by the compound word processing unit.
JP4053124A 1992-03-12 1992-03-12 Machine translator Withdrawn JPH05257967A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP4053124A JPH05257967A (en) 1992-03-12 1992-03-12 Machine translator

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP4053124A JPH05257967A (en) 1992-03-12 1992-03-12 Machine translator

Publications (1)

Publication Number Publication Date
JPH05257967A true JPH05257967A (en) 1993-10-08

Family

ID=12934062

Family Applications (1)

Application Number Title Priority Date Filing Date
JP4053124A Withdrawn JPH05257967A (en) 1992-03-12 1992-03-12 Machine translator

Country Status (1)

Country Link
JP (1) JPH05257967A (en)

Similar Documents

Publication Publication Date Title
JPH05257967A (en) Machine translator
JPH03260766A (en) Translating device with parenthesis generation deciding function
JP2007133905A (en) Natural language processing system and natural language processing method, and computer program
JPS61282969A (en) Language translating device
JP2770555B2 (en) Natural language analysis system
JP3244286B2 (en) Translation processing device
JP2726543B2 (en) Translation method
JP2794919B2 (en) Machine translation equipment
JP4033088B2 (en) Natural language processing system, natural language processing method, and computer program
JPH05143640A (en) Japanese-language syntax analysis system
JPH0421167A (en) Machine translation device
JP2926745B2 (en) Machine translation equipment
JP2742059B2 (en) Dictionary editor for translation
JPH04296969A (en) Mechanical translation device
JPH05151255A (en) Method for inputting japanese word
JP2770536B2 (en) Sentence analyzer
JPH041852A (en) Mechanical translation device
JPH0535785A (en) Machine translation device
JPH0589166A (en) Machine translation device
JPH05120329A (en) Machine translation device
JP2004086919A (en) Mechanical translation system
JPH05290084A (en) Unregstered word translation system
JPH0289175A (en) Syntax analyzing system in machine translation
JPH05143638A (en) Natural language analyzing method
JPH05250406A (en) Automatic translating device

Legal Events

Date Code Title Description
A300 Withdrawal of application because of no request for examination

Free format text: JAPANESE INTERMEDIATE CODE: A300

Effective date: 19990518