JPH05334339A - Machine translating system - Google Patents

Abstract

PURPOSE:To reduce the burden of postediting on a user by providing an electronic dictionary with a specific conjugation/inflection field. CONSTITUTION:The electronic dictionary 15 is provided with the conjugation/ inflection field wherein information indicating whether or not respective words as the constituent elements of a compound word are conjugated or inflected when an index word is the compound word can optionally be set and stored. Namely, when the input index word is the compound word, the user optionally sets and stores the information showing whether or not the respective words as the constituent elements are conjugated or inflected in the conjugation/ inflection field of the electronic dictionary 15. Then, the information is referred to during translation execution, all the words to be conjugated or inflected among the words constituting the index word are correctly conjugated or inflected. Therefore, the user can easily input the information for conjugation or inflection even unless the user have technical knowledge and the burden of postediting on the user is reducible.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a machine translation device,
In particular, the present invention relates to dictionary editing means when a user constructs an electronic dictionary for a machine translation device.

[0002]

2. Description of the Related Art Conventionally, electronic dictionaries used in systems such as machine translation are not only basic dictionaries built in advance in the system, but also two user dictionaries for registering words that the user needs at any time during translation work. It is configured. A user dictionary suitable for the situation is indispensable for the machine translation.
However, unlike the basic dictionary, the user dictionary is created by a user who does not have specialized knowledge about machine translation. Therefore, in order to minimize the user's time and effort, the headword and parallel translation of the source word, or the existing learning It was often the case that the user inputs only the minimum necessary information that can be understood by looking at the user dictionary and then constructs a simple user dictionary in which the values are supplemented by default. Therefore, for example, when constructing a user dictionary for translating from Japanese to English, each linguistic feature is used as a default value. It can be said that, when the translated word used by the user is a compound word composed of a plurality of words, it can be determined to some extent depending on the part of speech from the characteristics of normal English. For example,
If it is an English noun, "PC" is "personal computer"
(s) "is the last word, and if it is a verb," share "is"
A method in which the word to be used is determined from the characteristics of the part of speech, such as the first word such as "use (d) in common", and the user inputs the type of usage such as rule change or irregular change if necessary. Was taken.

[0003]

However, in practice, it may not work if the value is arbitrarily decided in the system. For example, in a Japanese-English machine translation system in which the source word is Japanese and the translated word is English, if the English word "bokeh break" in Japanese is "fray and break", then naturally the headword of that translated word is used. The past tense is "f
rayed and broke ", past participle is" frayed and broken "
However, the verb uses only "fray" because there is a method of using the first word. for that reason,
The past tense is "frayed and break", the past participle is "frayed an"
Unnatural verbs such as "d break" can be used. There is a problem that the user has to manually correct these after the translation is completed and the English sentence is generated.

The present invention has been made in order to solve the above-mentioned problems, and when the entry word input by the user is a compound word, all the words to be utilized among the words constituting the entry word are used. It is an object of the present invention to provide a machine translation device that allows a user to specify information for correct utilization and reduces the addition to the user such as post-editing.

[0005]

In order to achieve this object, a machine translation device of the present invention, when a headword and a headword are compound words composed of a plurality of words,
The electronic dictionary is provided with a word that is a component of the word and a field that stores a part of speech of each word, and in the electronic dictionary, when the entry word is a compound word, each word of the component is Arbitrarily set information on whether to utilize or not
There is a utilization field that can be stored.

[0006]

In the machine translation apparatus of the present invention having the above configuration, when the entry word entered in the utilization field of the electronic dictionary is a compound word, whether or not each of the constituent words is utilized This information is arbitrarily set / stored by the user. This information is referred to when the translation is executed, so that all the words to be used among the words forming the headword are correctly used.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram schematically showing the hardware configuration of a machine translation device according to the present invention. As shown in FIG. 1, the Japanese-English machine translation device 10 includes a data processing unit 11 including a computer, a display unit 12 including a CRT, an input unit 13 including a keyboard, and the like. The printing unit 14 includes a scanning printer or a laser printer, and an electronic dictionary 15 stored in a hard disk, an optical disk, or the like.

The electronic dictionary 15 includes a basic dictionary that the system internally has and a user dictionary that allows the user to register necessary items, and is a Japanese word necessary for analyzing Japanese as a source word. Morphological analysis dictionary unit 31, Japanese syntax analysis dictionary unit 32, conversion dictionary unit 33 necessary for converting from Japanese to English which is a translation word, and English necessary for generating English which is a translation word Syntax generation dictionary unit 3
4 and an English morpheme generation dictionary unit 35.

In the data processing unit 11, a translation unit 16 for performing translation processing realized by software using the function of a computer, an input from the input unit 12, editing before and after translation, and dictionary registration. And input and output sentences on CRT1
3, an editing unit 17 to be sent to the data processing unit 3, a data storage unit 18 for temporarily storing input / output data, and a user
The compound word editing unit 19 processes the dictionary information input from the input unit 13 according to the dictionary registration menu, and registers or deletes the dictionary information in the electronic dictionary 15.

The compound word editing section 19 includes a space detecting section 41 for detecting spaces in an English compound word, a word dividing processing section 42 for dividing each word constituting the compound word for editing. The storage unit 43 that temporarily stores the information that is input by each of the words, that is, whether the word is used or not, is temporarily stored after being replaced with the information indicating which word is used, and the information that is used in the field is stored. For each word, the English morpheme generation dictionary unit 35 in the electronic dictionary 15
And a dictionary search unit 44 for checking whether or not there is utilization information in the English morpheme generation dictionary unit 35.

In the English morpheme generation dictionary unit 35, as shown in FIG. 2, a headword field 61, a part-of-speech 62 of the headword, and a constituent word field 63 if the headword is a compound word. Part-of-speech field 64 of constituent words
And a compound word information field 60 including a utilization field 65 indicating that the constituent word is utilized.
Further, there is an inflectional type information field 66 in which a value indicating whether the inflection word is used regularly or irregularly is entered when the headword is composed of one word. Information is described in the past participle field 67 for describing the past participle information and the past participle form field 68 for describing the past participle form information only when the value indicating the irregular part utilization is the irregular use. Further, there is a present participle field 69 in which a value indicating whether a consonant at the end of a sentence is overlaid or a rule change, which is necessary when forming the present participle, is entered.

Next, the basic operation of this machine translation system will be schematically described.

The input sentence of the source word input from the input unit 13 is edited by the editing unit 17 into an appropriate form as needed, and then sent to the translation unit 16. The translation unit 1
Reference numeral 6 translates an input sentence into a translated word using the electronic dictionary 15, and temporarily stores the resulting output sentence in the data storage unit 18. When the user activates the editing unit 17 from the input unit 13 and specifies a necessary output sentence, the editing unit 17 retrieves the output sentence from the data storage unit 18 and displays it on the display unit 12, or Printing department 1
Output from 4. If a word with incorrect dictionary information or an unregistered word is found in the output sentence, the user activates the editing unit 17 at any time to create a user dictionary to improve the translation accuracy in the next time. Register in 15 or modify existing dictionary contents.

In the electronic dictionary used in the Japanese-English machine translation system, if the source word is a group of words, the translated words are often handled as a group of words. Therefore, translated words often take the form of compound words. FIG. 3 shows a flow of processing of such an entry word, in which an unregistered word “out of focus” is registered in the electronic dictionary as an example.

First, in order to register the verb "out of focus" in the electronic dictionary 15, the user inputs the dictionary registration menu of the editing unit 17 to the input unit 13 as shown in FIG.
The input word "fray and break", which is the headword to be registered and which is to be registered, and then its translation "fray and break"
Input from 3 (S1). Due to the linguistic characteristics of English, in many compound words the position of the main word, that is, the word to be utilized, can be determined by the part of speech. In the verb, for example, "use (d) in common" of "shared" is a word used by the leading "use", that is, a main verb. However, in the case of the translated word "fray and break" for "out of focus",
The past tense is "fraye"
"d and broke", the past participle is "frayed and broken", and it has two main verbs.

The compound word editing unit 19 receives, from the editing unit 17, the translated word data input by the user from the input unit 12. The space detection unit 41 therein determines whether or not a space is included in the translated word (S
2). If a space is included, it is determined that the word is a compound word, and the word division processing unit 42 starts compound word processing to divide the word into spaces (S3).

The compound word editing unit 19 informs the editing unit 17 to enter a menu for compound word processing, and the editing unit 17 determines which word in the translated word is used, that is, what number word the main verb is. Display the question on the display unit 12 (S
4) As shown in FIG. 5, the user inputs “1” and “3” from the input unit 13. This number is sent to the storage unit 43 in the compound word editing unit 18, where the first "f
The information "ray" and the third "break" are verbs and are words to be utilized, and the storage unit 43 stores "fray" and "an" in the word division processing unit 42.
For each word divided into "d" and "break", the word used is always the same part of speech as the original headword, so
From S5, "fray" and "break" are assumed to be verbs, and the entry word of each word, its part of speech, and information to be used are stored (S5).

The dictionary search unit 43 in the compound word editing unit 19
Is the English morpheme generation dictionary unit 35 in the electronic dictionary 15.
Is searched to see if there is individual dictionary information for "fray" and "break" (S6). When the individual dictionary information exists in the English morpheme generation dictionary unit 35, the dictionary information such as the syntax information is input (S8). When the individual dictionary information does not exist in the English morpheme generation dictionary unit 35, the editing unit 17
The display unit 1 has a menu for adding morphological information of individual dictionaries such as inflectional information and participles in case of irregular change.
2 is displayed (S7).

If the translated word does not include a space, the compound word processing is not entered and a menu for directly adding morpheme information is entered (S7).

When the necessary morpheme information is input, other dictionary information is input by the editing section (S8). When the information is collected, the editing unit 17 stores the dictionary information in the data storage unit 18 input by the user from the input unit 13 and the storage unit 43 in the compound word editing unit 19. The compound word information is collectively registered in the electronic dictionary 15 (S9).

The order in which information is actually stored in the English morpheme generation dictionary unit 35 in the electronic dictionary 15 is shown in FIG.
Will be explained. Translation of "out of focus""fray and
In the case of "break", the information is stored in the fields in the English morpheme generation dictionary unit 35 from the information stored in the storage unit 43 in S5 as follows.
“Ray and break” is described in the entry word field 61 (S11), and its part of speech “V (verb)” is described in the part of speech field 62 (S12). In the word field 63, the individual headwords “fray”, “and”, and “br” that are divided by the word division processing unit 42 and are stored in the storage unit 43.
"eak" is described (S13). In the part-of-speech field 64, "fray" stored in the storage unit 43 is described.
"Fray" and "break" from the fact that and "break" are verbs
The part-of-speech information is described in, and the part-of-speech information is unnecessary for "and" that is not used, so nothing is described (S14). In the utilization field 65, a flag "T" indicating that "fray" and "break" are utilized is set (S15).

If the information of each word is not already registered in the English morpheme generation dictionary, the information is described in each of the fields by the menu as shown in S7. The storage procedure in that case is as follows. In the case of the above-mentioned "fray", since it is not a compound word, nothing is described in the compound word information field 60. Instead, a value indicating a rule change is described in the inflection type information field 66 (S16). When a value indicating irregular change is described here, necessary values are described in the past tense field 67 and the past participle field 68 (S1).
7). In the case of "break", the inflected information field 66
And a value indicating an irregular change in the past tense field 67.
"Broke" and "broken" in the past participle field 68
Is described. After that, a value indicating a rule change is described in the present participle field 69 (S18).

The above example describes the case where the word that the user wants to register is not registered in the electronic dictionary. However, even if the user wants to change the parallel translation of a certain headword, if the English translation is a compound word, the same space detection is performed at the time when a new translation is input, and the same effect can be expected. In addition, since the headwords like this example are decimals, the user interface displays words that are candidates for inflection words by part-of-speech and asks the user for confirmation, and only the necessary items are processed as described above. It is also possible to apply to enter. If the headword contains a conjunction such as "and" or "or", there is a high possibility that multiple words will be used, as in this example. Therefore, only the relevant word has priority for this process. You may call it. INDUSTRIAL APPLICABILITY The present invention can be applied not only to a dictionary between a plurality of languages used in a machine translation device but also to the construction of other natural language processing electronic dictionaries, thesauri, knowledge databases and the like.

[0025]

As is apparent from the above description, according to the machine translation device having the compound word editing unit of the present invention, when the entry word is utilized in the electronic dictionary used for translation or the like, it is a constituent element. A user who does not have specialized knowledge can easily input information for correctly utilizing all words, and it is possible to reduce post-editing and the like to be added to the user.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of the present invention.

FIG. 2 is dictionary information stored in an electronic dictionary according to an embodiment of the present invention.

FIG. 3 is a flowchart showing a flow of a translation process according to an embodiment of the present invention.

FIG. 4 is a Japanese headword according to an embodiment of the present invention;
It is a figure of the menu screen which inputs an English translation.

FIG. 5 is a diagram of a screen for inputting the order of main verbs according to an embodiment of the present invention.

FIG. 6 is a flowchart showing a flow of an embodiment for storing information in an English morpheme generation dictionary according to the present invention.

[Explanation of symbols]

 10 Japanese-English machine translation device 15 Electronic dictionary 61 Headword field 63 Constituent word field 64 Part-of-speech field 65 Utilization field

Claims (1)

[Claims] 1. An electronic dictionary having a field for storing a headword and, when the headword is a compound word composed of a plurality of words, a word which is a constituent of the headword and a part of speech of each word. In the machine translation device provided with, in the electronic dictionary, when the headword is a compound word, a utilization field capable of arbitrarily setting and storing information on whether or not each word that is a constituent element thereof is utilized. A machine translation device characterized by being provided.