JPH04309171A - Device and method for machine translation - Google Patents

Abstract

PURPOSE:To obtain the machine translation device which can translate character strings mixing an objective language into a source language and can mix the source language and the objective language into the character strings as the translated results. CONSTITUTION:A partial character string designation part 103 designates the character strings not requiring translation among inputted character strings. Next, a partial character string retrieval part 104 retrieves the designated character strings. The retrieved character strings are turned to special characters at a special character generation part 105, and the pairs of the retrieved character strings and the special characters are stored in a special character part partial character string correspondence storage part 106. Next, the inputted character strings are turned to character strings replacing the character strings, which are designated at the partial character string designation part 103, with the special characters by a special character transformation part 107. At a transformation part 108, the character string in the objective language mixing the special characters is outputted, recovered at a special character recovery part 110 and outputted from a character string output part 111.

Description

[Detailed description of the invention]

0001

[Industrial Field of Application] When the languages composing a first character string are the first language and the second language, the present invention provides a method for converting a substring of the first character string in the second language into a first character string. The present invention relates to a machine translation device and method for converting a first character string into a second character string in a second language without conversion. Further, in the present invention, when the first language is the only language that constitutes the first character string, and there is a partial character string that does not need to be converted into the second language, the first character string can be converted into the first language. The present invention relates to a machine translation device and method for converting a second character string into a second character string in which a second language and a second language are mixed.

0002

BACKGROUND OF THE INVENTION Examples of conventional machine translation devices and methods include those described on pages 7 to 10 of BIT Separate Volume Machine Translation (edited by Hirosato Nomura and Hozumi Tanaka, Kyoritsu Shuppan).

FIG. 8 is a schematic block diagram of such an example machine translation system. As shown in FIG. 8, the character string input section 3
The first character string in the first language input from 01 is
The conversion unit 302 converts it into a second character string in the second language, and outputs it from the character string output unit 303.

[0004] In FIG. 8, the converter refers to a unit that converts a first character string in a first language into a second character string in a second language.

[0005]

[Problems to be Solved by the Invention] However, in the device having the above configuration, all of the character strings input to the converting section 302 are translated, resulting in a character string in which the first language and the second language are mixed. There was a problem that the translation could not be done when inputting . There is also the problem that partial character strings in the first character string that do not need to be converted into the second language are also converted into the second language.

[0006] In view of the above-mentioned problems, the present invention provides that when the languages composing the first character string are the first language and the second language, the partial character string in the second language is not converted. An object of the present invention is to provide a machine translation device and method for converting a first character string into a second character string in a second language.

[0007] Furthermore, if the language constituting the first character string is the first language, and there is a substring that does not need to be converted into the second language, the first character string can be converted into the first language. An object of the present invention is to provide a machine translation device and method for converting a second character string into a second character string in which a second language is mixed.

[0008]

[Means for solving the problem] The invention according to claim 1 includes:
a character string input section for inputting a first character string; a character string temporary storage section for temporarily storing the first character string input to the character string input section; a substring specification section that specifies a substring of the first string; a special character generation section that generates a special character corresponding to the substring specified by the substring specification section; a special character partial string correspondence storage section that stores and stores a partial string generated in the generation section and a set of special characters corresponding to the partial string; Read the special characters, convert the partial string specified by the partial string specification section into the special string, and convert the string stored in the temporary string storage section into a first string containing the special characters. and a first character string containing the special character converted by the special character converting unit temporarily stored in the character string temporary storage unit into a second character string containing the special character. a conversion section that performs conversion; a conversion result temporary storage section that temporarily stores the result converted by the conversion section; and a conversion result temporary storage section that temporarily stores a result converted by the conversion section; a special character recovery unit that searches the first character string from the special character partial string correspondence storage unit and converts the special character into a corresponding first character string; A machine translation device characterized in that it includes a character string output unit that outputs a second character string that includes a partial character string of a character string.

The invention according to claim 2 includes an input step of inputting a first character string, a partial character string designation step of designating a partial character string of the input character string, and a step of specifying a partial character string corresponding to the partial character string. A special character generation stage in which a special character is generated, a temporary storage stage in which the partial character string and the special character are temporarily stored, and a partial character string specified in the first character string input in the input stage. a special character conversion step of converting a partial string of characters into the special character generated in the special character generation step; and a conversion step of converting a first character string including the special character into a second character string including the special character. and a special character for converting a special character included in the second string obtained in the conversion step into a substring of the first string corresponding to the special character temporarily stored in the temporary storage step. A machine translation method comprising a recovery step and a character string output step of outputting a second character string that includes a partial string of the first character string obtained in the special character recovery step.

0010

[Operation] As described above, the present invention converts a partial character string that does not need to be converted into a special character, prevents the special character from being converted by the conversion section, and converts the special character into a special character string in the special character recovery section. Recovered to string. Therefore, by specifying a character string in the first character string that does not need to be converted into the second language, you can convert the first character string that is a mixture of the first and second languages into the second language. It is possible to convert a first character string in a first language into a second character string in a mixture of the first language and the second language.

[0011]

Embodiments Hereinafter, embodiments of the present invention will be explained in detail with reference to the drawings.

FIG. 1 is a block diagram of a machine translation device showing an embodiment of the present invention, FIG. 2 is a machine translation flowchart showing an embodiment of the invention, and FIGS. 3, 4, 5, and 6 are explanations of the flowcharts. FIG.

In FIG. 1, 101 is a character string input section;
02 is an input character string and a special character conversion unit 10 which will be described later.
103 is a partial string specification section for specifying a partial string that does not need to be converted; A substring search unit 105 searches for a substring that does not need to be converted from the output, and generates a special character corresponding to the substring searched by the substring search unit 104, and distinguishes between the substring and the special character. A special character generating unit stores the set in a special character substring corresponding unit to be described later, and 106 is a special character substring that stores the set of the substring generated by the special character generating unit 105 and its corresponding special character. A correspondence storage unit 107 is a partial character string specification unit 103 in the input character string temporarily stored in the character string temporary storage unit 102.
108 is a special character conversion unit that converts a character string specified by , into a special character, and outputs the character string to a character string temporary storage unit; a conversion unit that converts the character string into a second character string including the special characters; 109 is a conversion result temporary storage unit that temporarily stores the result converted by the conversion unit 108; 110 is included in the second character string; A special character recovery unit 111 converts a special character into a corresponding partial string of the first character string, and a character string output unit 111 outputs a character string output from the special character recovery unit.

In FIG. 3, example sentence 1 converts the first character string in the first language into the first language by specifying a character string in the first character string that does not need to be converted into the second language. An example of converting to a second string that is a mixture of language and a second language.
In example sentence 2, by specifying a character string in the first character string that does not need to be converted to the second language, the first character string in which the first language and the second language are mixed can be converted into the second language. This is an example of converting to a second character string that is a language.

Next, the operation of the same embodiment will be explained with reference to FIGS. 1 and 2.
, will be explained with reference to FIGS. 3, 4, 5, 6, and 7. 201 is a stage for inputting a character string. For example, the character string shown in FIG. 3 is input from the character string input section 101. 202 stores the above step 2 in the character string temporary storage unit 102.
This is the stage of storing the character string input in step 01. 203
This is the step of specifying a partial string that does not need to be converted among the strings stored in step 202 above. For example, the character string shown in FIG.
The character string is specified. Step 204 is a step of searching for a partial character string from among the input character strings that include the partial character string specified in step 203 above. Step 205 is a step of generating special characters corresponding to the partial string searched in step 204 above. For example, the special character corresponding to the character string in FIG. 4 in the special character generation unit 105 becomes the special character in FIG. Step 206 is a step of storing the set of the partial character string searched in step 204 and the special character generated in step 206 in the special character partial string correspondence storage section 106 . 207 first inputs the input character string in which the partial string output from step 205 is specified, and then stores the special character corresponding to the partial string searched in step 204 in the special character partial string correspondence storage section. 106, the designated character string is converted into a corresponding special character, and the result is then stored in the character string temporary storage section 109. 2
Step 08 is a step of converting the character string stored in step 207 into a second character string that includes a special character string. For example, the character string shown in FIG. 3 is outputted from the conversion unit 108 as a character string shown in FIG. 209 is step 20 above
This is the stage where the conversion result obtained in step 8 is stored in the conversion result temporary storage unit 109. 210 first searches for a special character in the character string stored in step 209 above, and then retrieves a substring of the first character string corresponding to the searched special character from the special character substring correspondence storage unit 106. This is the stage of searching and converting the special string into a corresponding substring of the first string. Step 211 is a step of outputting a second character string including the substring of the first character string recovered in step 210 above. For example, the input character string shown in FIG. 3 is outputted from the character string output unit 111 as a character string shown in FIG.

As a result of these steps, by specifying character strings that do not need to be converted into the second language, the first character string in the first language and the second language can be converted into the second language. It becomes possible to convert the first string in the first language into a second string in the first language and the second string in the first language. .

It should be noted that the present invention is not limited to the above embodiments, and various modifications can be made based on the gist of the present invention, and these are not excluded from the scope of the present invention.

[0018]

[Effects of the Invention] As explained above in detail, according to the machine translation device and method of the present invention, it is possible to specify a character string in a first character string that does not need to be converted into a second language. Converting a first character string in a mixture of the first language and the second language into a second character string in the second language,
A first character string in a first language can be converted into a second character string in a mixture of the first language and the second language.

[Brief explanation of drawings]

[Fig. 1] Block diagram of a machine translation device showing an embodiment of the present invention

[Figure 2] Machine translation flowchart showing the operation of the same embodiment

[Figure 3] Explanatory diagram of the machine translation flowchart of the same example

[
Figure 4: Explanatory diagram of the machine translation flowchart of the same example

[Figure 5] Explanatory diagram of a machine translation flowchart of the same example

[Figure 6
]Explanatory diagram of the machine translation flowchart of the same example

[Figure 7]

[Figure 8] Schematic block diagram of a conventional machine translation system

[Explanation of symbols]

101 Character string input section 102 Character string temporary storage section 103 Partial string specification section 104 Partial string search section 105 Special character generation section 106 Special character partial string correspondence storage section 107 Special character conversion section 108 Conversion section 109 Conversion result temporary storage Section 110 Special character recovery section 111 Character string output section 301 Character string input section 302 Conversion section 303 Character string output section

Claims (2)

[Claims] [Claim 1] A character string input section for inputting a first character string;
a character string temporary storage section that temporarily stores the first character string input to the character string input section; and a partial character that specifies a partial string of the first character string temporarily stored in the character string temporary storage section. a column specification section, a special character generation section that generates a special character corresponding to a substring specified by the substring specification section, and a substring generated by the special character generation section and the special character corresponding thereto. A special character partial string correspondence storage section stores and stores a set of special character partial string correspondence storage sections, and special characters stored in the special character partial string correspondence storage section are read from the special character partial string correspondence storage section, and the special characters specified by the partial character string specification section are read. a special character conversion unit that converts a partial character string into the special character string and converts the character string stored in the character string temporary storage unit into a first character string including special characters; and the character string temporary storage unit. a conversion unit that converts a first character string containing the special character converted by the special character conversion unit, which is temporarily stored in the special character conversion unit, into a second character string containing the special character; A first character string corresponding to a special character included in a second character string temporarily stored in the temporary conversion result storage part and the second character string temporarily stored in the temporary conversion result storage part from the special character partial string correspondence storage part. a special character recovery unit that searches and converts the special character into a corresponding first character string; and a second character string that includes a substring of the first character string obtained by the special character recovery unit. A machine translation device characterized by comprising a character string output unit that outputs. 2. An input step for inputting a first character string; a partial string specification step for specifying a partial string of the input character string; and a special character string generating step for generating a special character corresponding to the partial string. a character generation stage; a temporary storage stage for temporarily storing the partial character string and the special character; and a partial character string specified in the partial character string specification stage included in the first character string input in the input stage. a special character conversion step of converting a first character string including the special character into a second character string including the special character; a special character recovery step of converting the special characters included in the second character string obtained in the conversion step into a substring of the first character string corresponding to the special characters temporarily stored in the temporary storage step; A machine translation method comprising a character string output step of outputting a second character string that includes a partial string of the first character string obtained in the special character recovery step.