JP2002007392A - Machine translation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a machine translation device capable of efficiently retrieving a dictionary while reducing the capacity of the dictionary. SOLUTION: A pattern dictionary 12 shows variables, original language patterns, translation directions and target language patterns. When any translation direction is selected by a selecting means 10 and an original sentence is inputted from an input means 11, a converting means 13 repeatedly performs processing for replacing a relevant original language pattern with a variable while referring to the pattern dictionary 12 for the input sentence on the basis of the translation direction selected by the selecting means 10, and retrieving whether the input sentence containing this variable is relevant to any original language pattern in the pattern dictionary 12 or not and converts the input sentence to a sentence completely replaced with the variable. An expanding means 14 provides the translated result in the target language by expanding the converted original language pattern into target language pattern corresponding to combination with the converted original language pattern.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a machine translator for translating a source language into a target language.

[0002]

2. Description of the Related Art In a machine translation apparatus, it is important not only to translate correctly, but also to translate as desired by a user. In the conventional machine translation technology, terms that can be registered in the dictionary are terms of part of speech and unique translation patterns. It is difficult to register phrases and expressions in parts of speech. In some cases, the translation pattern can replace a numerical value or a specific term in the original sentence as a variable, but basically, the original sentence and the translated sentence correspond one-to-one. In order to cover a wide range of expressions, a large number of patterns must be prepared, and it takes a lot of time to create and search for a translation pattern, which is not efficient. For example, Japanese Patent Application Laid-Open No. 5-290082 discloses a translation pattern that can be easily created by a user and that can be efficiently searched by providing a translation pattern that can be created using variables and a nested structure. We propose a device that can do it.

[0003]

However, in the above-mentioned prior art, it is described that English and German can be translated into Japanese. However, it is desired to translate a combination of a source language pattern and a target language pattern. There is a problem in that it is redundant to create each interval, and if the source language pattern and the target language pattern do not correspond one-to-one, they cannot be simply created.

For example, to translate Japanese into English,
To translate "will improve" as "will improve", the source language pattern is Japanese, the target language pattern is English, and "improve" and "to be""will...""Will be improved" and translated as "will improve".
Suppose you also created “will…”. As described above, in order to translate English into Japanese, it is redundant to recreate the source language pattern in English and the target language pattern in Japanese, and "will ..."""~right"
Must be deleted because it corresponds to both.

[0005]

The present invention employs the following structure to solve the above-mentioned problems. <Structure 1> A source language pattern, a target language pattern corresponding to the source language pattern, a bilingual dictionary indicating translation directions of the source language pattern and the target language pattern, selection means for selecting a translation direction of the input sentence, and selection A translation means for translating an input sentence by referring to a bilingual dictionary based on the translation direction selected by the means.

<Structure 2> A source language pattern, variables corresponding to the source language pattern, a target language pattern corresponding to the source language pattern, translation directions of the source language pattern and the target language pattern, and At least one of the language patterns is selected by the pattern dictionary, which is a nested source language pattern incorporating variables obtained by replacing other source language patterns, a selection unit for selecting a translation direction of an input sentence, and a selection unit. Based on the translation direction, the input sentence is referred to the pattern dictionary, the corresponding source language pattern is replaced with a variable, and the input sentence including the variable is searched for the source language pattern of the pattern dictionary. And a conversion means for converting the input sentence into a combination of nested source language patterns in which all input statements are replaced by variables, and a combination of the source language patterns converted by the conversion means. Hazuki, expand target language pattern with reference to the pattern dictionary, machine translation apparatus characterized by comprising a deployment means for the translation result of the input sentence.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to specific examples. << Specific Example >><Configuration> FIG. 1 is a configuration diagram showing a specific example of the machine translation apparatus of the present invention. The device shown in FIG.
1, pattern dictionary 12, conversion means 13, expansion means 14,
It comprises output means 15.

[0008] The machine translation device is realized by, for example, a personal computer. The selection means 10 is a means for selecting a source language and a target language to be translated, and has a function as a user interface for displaying to a user and receiving a designation of a translation direction from the user. And hardware such as a processor and a memory for executing the same. Input means 1
1 is a means for inputting the original text, and is a CD-ROM
It is composed of an interface for taking in the original text from a recording medium such as a disk or a floppy (registered trademark) disk, a network interface, a keyboard for directly inputting the original text, and the like. Pattern dictionary 12
Is a bilingual dictionary indicating a source language pattern, a target language pattern corresponding to the source language pattern, translation directions of the source language pattern and the target language pattern, and variables, and is provided in a magnetic disk device or the like.

FIG. 2 is an explanatory diagram showing an example of the pattern dictionary 12. As shown in the figure, the pattern dictionary 12 is a dictionary composed of variables, source language patterns, translation directions, and target language patterns. In this specific example, the source language is English and the target language is Japanese. At least one of the language patterns has a nested structure that can be included as a corresponding variable in another language pattern. For example, an English pattern (it) can be included in another English pattern (improve OBJ) as a variable (OBJ). The translation direction is the direction for converting the language pattern,
=> Is left-to-right (English to Japanese; this pattern is omitted in the illustrated example), <= is right-to-left (Japanese to English), and = is bidirectional.

The conversion means 13 refers to the pattern dictionary 12 and replaces the corresponding source language pattern with a variable with respect to the original sentence input from the input means 11 based on the translation direction selected by the selection means 10; The input sentence including the replaced variable is searched for a source language pattern in the pattern dictionary 12, and this process is repeated to obtain a nested source language pattern in which the input sentence is entirely replaced with variables. It is a means to convert. The expansion unit 14 expands the sentence replaced by the variable by the conversion unit 13 using the target language pattern of the pattern dictionary 12, and obtains a translation result of the target language. In addition, these conversion means 1
The translation means 16 is constituted by 3 and the development means 14. The output unit 15 is a functional unit for outputting the translation result obtained by the expanding unit 14. The conversion means 13 and the development means 14 are realized by software corresponding to the function of each means and hardware such as a processor and a memory for executing the software. The output unit 15 includes a display for displaying translation results, a printer for printing, or an output interface for outputting translation data.

<Operation> The operation of this embodiment will be described below by taking an actual input sentence as an example. FIG. 3 is a flowchart showing the translation processing of this specific example. First, a language to be translated is selected by the selection means 10 (step S1). In this specific example, it is selected to translate English into Japanese.

Next, the original text is input to the input means 11 (step S2). Here, “I will imp
The conversion means 13 searches the pattern dictionary 12 for an English pattern corresponding to a part or the whole of the original sentence (step S3), and converts the original sentence into an English pattern so that no variable remains. (Step S4) Here, in the pattern dictionary 12, only the pattern whose translation direction is from English to Japanese => or = is searched and converted.In the pattern dictionary 12 shown in FIG. (2) (4) (5)
(6) is applicable. Then, the conversion means 13 converts the original sentence into an English pattern so that no variables remain. “I will improve it.” → (5) SUBJ will improve (6) OBJ. → (5) SUBJ will (4) VERB. → (5) SUBJ (2) VERB. → (1) SENTENCE In the case of “improve it.” → (5) SUBJ will improve (6) SUBJ., for example, a part that cannot be converted to a variable such as “improve SUBJ” remains, so this combination is not adopted. As a result, the conversion result is [[I (5)] [will [improve [it
(6)] (4)] (2)]. (1)].

Next, the converted English pattern is converted into a Japanese pattern to develop the pattern (step S).
5). [[I (5)] [will [improve [it (6)] (4)] (2)]. (1)] → [[I (5)] [will [improve [it] (4)] (2)]. (1)] → [[I (5)] [will [improve it] (2)]. (1)] → [[I (5)] [We plan to improve it]. (1)] → [I] [I plan to improve it]. (1)] → [I plan to improve it. The output means 15 says, "I plan to improve it."
Is output as a translation result (step S6).

Next, a case where the direction to be translated is reversed will be described. The selection means 10 selects to translate Japanese into English (step S1), and inputs "I plan to improve it" to the input means 11. The conversion unit 13 searches the pattern dictionary 12 for a Japanese pattern whose translation direction is <= or = from Japanese to English, and converts the original sentence to a Japanese pattern so that no variable remains ( Steps S3 and S4). In the illustrated pattern dictionary 12, (1)
(2) Since (4), (5), and (6) apply, the same conversion as described above is performed as shown below. "I plan to improve it." → (5) SUBJ plans to improve (6) OBJ. → (5) SUBJ is scheduled for (4) VERB. → (5) SUBJ is (2) VERB. → (1) SENTENCE “I plan to improve it.” → (5) SUBJ plans to improve (6) SUBJ. In this case, for example, a part that cannot be converted into a variable such as [improve SUBJ] remains, so this combination is not adopted. As a result, the conversion result is [[I (5)]
Will be [[Improve it [6]] (4)] (2)].
(1)].

The Japanese pattern converted in the same manner as in the above example is converted into an English pattern and the pattern is developed (step S5). [[I (5)] will be [[[it (6)] improve (4)] [2]]. (1)] → [[I (5)] is planning to improve [[it] (4)] (2)]. (1)] → [[I (5)] is [[improve it] is scheduled (2)]. (1)] → [[I (5)] is [will improve it]. (1)] → [I is [will improve it]. (1)] → [I will improve it.] The output unit 15 outputs “I will improve it.” As a translation result (step S6).

When the input sentence is "I will improve it." Conversion means 13
However, referring to the pattern dictionary 12, a search is made for a Japanese pattern whose translation direction is <= or = from Japanese to English, and the original sentence is converted to a Japanese pattern so that no variables remain (step S3, S4). Illustrated pattern dictionary 12
Then, since (1), (3), (4), (5), and (6) correspond, the following conversion is performed. "I will improve it." → (5) SUBJ will improve (6) OBJ. → (5) SUBJ would be (4) VERB. → (5) SUBJ is (3) VERB. → (1) SENTENCE As a result, the conversion result is [[I (5)] is [[[It (6)]
(4)] would be improved (3)]. (1)].

The Japanese pattern converted in the same manner as in the above example is converted into an English pattern and the pattern is developed (step S5). [[I (5)] would be [[Improve (6)] (4)] [3]]. (1)] → [[I (5)] would be [[improve [it] (4)] (3)]. (1)] → [[I (5)] would be [[improve it] (3)]. (1)] → [[I (5)] is [will improve it]. (1)] → [I is [will improve it]. (1)] → [I will improve it.] The output unit 15 outputs “I will improve it.” As a translation result (step S6).

<Effects> As described above, according to the specific example, since the direction to be translated is added to the pattern in the pattern dictionary 12, it is not necessary to create a pattern for each language to be translated. Has the effect that it can be converted into a dictionary. As a result, the capacity of the dictionary as a translation dictionary can be reduced.

In the above example, the translation direction information is
Although applied to a pattern dictionary with variables, a bilingual dictionary indicating the target language pattern corresponding to the source language pattern is provided,
Any machine translation device that translates using the bilingual dictionary by the translation means can be applied in the same manner as in the specific example.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a specific example of a machine translation device of the present invention.

FIG. 2 is an explanatory diagram showing an example of a pattern dictionary of the machine translation device of the present invention.

FIG. 3 is a flowchart showing the operation of the machine translation device of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Selection means 12 Pattern dictionary (bilingual dictionary) 13 Conversion means 14 Expansion means 16 Translation means

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Mihoko Kitamura 1-7-112 Toranomon, Minato-ku, Tokyo Oki Electric Industry Co., Ltd. (72) Saori Shimohata 1-7-112 Toranomon, Minato-ku, Tokyo Inside the Electric Industry Co., Ltd. (72) Inventor Takeshi Fukui 1-7-12 Toranomon, Minato-ku, Tokyo Oki Electric Industry Co., Ltd. (72) Inventor Masamichi Fuchigami 1-7-12 Toranomon, Minato-ku, Tokyo Oki Electric F term in Industrial Co., Ltd. (reference) 5B091 AA05 AA06 BA05 CA21 CC01 CC14 CC15

Claims (2)

[Claims] 1. A source language pattern, a target language pattern corresponding to the source language pattern, a bilingual dictionary indicating translation directions of the source language pattern and the target language pattern, and a selection unit for selecting a translation direction of an input sentence. And a translation unit for translating the input sentence with reference to the bilingual dictionary based on the translation direction selected by the selection unit. 2. A source language pattern, variables corresponding to the source language pattern, a target language pattern corresponding to the source language pattern, translation directions of the source language pattern and the target language pattern, and At least one of the source language patterns is a pattern dictionary that is a nested source language pattern incorporating a variable obtained by replacing another source language pattern; a selection unit that selects a translation direction of an input sentence; On the basis of the translation direction selected in step 2, the input sentence is referred to the pattern dictionary, the corresponding source language pattern is replaced with a variable, and for the input sentence including the variable, the source language of the pattern dictionary is further changed. A conversion unit for searching whether the input sentence corresponds to a pattern, and converting the input sentence into a combination of nested source language patterns in which all the input sentences are replaced with variables; And a developing means for developing the target language pattern by referring to the pattern dictionary based on the combination of the source language patterns converted in step (a) and obtaining a translation result of the input sentence.