JPH0585055B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a Roman-to-Kanji conversion device for converting Roman character input into Japanese for computer processing.

(Prior art) Conventionally, IEICE Transactions Vol.J67-DNo.7
(pp. 784-791), Japanese word processors and the like are widely used to input Japanese by converting Roman characters to Kanji or Kana-Kanji.

(Problem to be solved by the invention) However, in today's Japanese documents,
Not only Japanese words but also words from other languages such as English are often used in texts mixed with Japanese words, not written in katakana but in the original spelling of the language. In particular, depending on the field of the document being written, words from other languages often appear frequently. Conventionally, in order to deal with this, the user consciously changed the state of the device called "input mode", for example from "conversion mode" to "non-conversion input mode", and then changed It was necessary to enter words in the language of the user and then return from the "non-conversion input mode" to the "conversion mode" after inputting the words. Moreover, this increases the mental burden on the user by forcing the user to perform many operations, leading to a decrease in input efficiency and concentration on input contents.

The purpose of the present invention is to use the characteristics of the adjacency relationships of characters appearing in the input character string to
An object of the present invention is to provide a Romanji-Kanji conversion device capable of Japanese input with less burden on the user.

(Means for Solving the Problem) The present invention provides a character string permissibility determining means for determining whether or not an ordinal co-occurrence relationship of characters in an input character string is permissible in Japanese; A Japanese conversion means converts the input character example into Japanese if it is permitted by the character string admissibility determination means; It is characterized by having another language conversion means that performs a process of converting into a language.

(Operation) The Roman alphabet notation normally used for Japanese input only uses a certain set of characters.

Since Roman characters are only used in combinations of one to two characters to represent one hiragana character, each character can only be used before and after a specific character.

There is a characteristic that.

For example, the letters that can appear after the letter "s" are usually vowels "a", "i",
There are only "u", "e", "o", and "s" to represent a consonant. Therefore, as an example, "schema"
Assuming that the character string ``c'' is input, it can be determined that a word with a Roman alphabet notation that is not allowed in Japanese has been input or is about to be input when the character "c" is input.

(Embodiment) FIG. 1 is a diagram showing an embodiment of a Roman-to-Kanji conversion device according to the present invention. The user's Romaji input is stored in the input buffer 101. The character string feature detection unit 102 uses the input buffer 10
1, and determines whether the currently input character string is a character string allowed in the main language. The conversion control unit 103 selectively activates the main language conversion processing unit 104 or the sub-language conversion processing unit 105 based on the language type information input from the character string feature detection unit 102, and uses the returned result as the conversion result. Output. The main language conversion processing unit 104 uses the main language conversion dictionary 106 to search for words in the main language from the Roman alphabet input, and outputs the results to the conversion control unit 103. The sub-language conversion processing unit 105 uses the sub-language conversion dictionary 107 to search for words in the sub-language from the Roman alphabet input, and outputs the results to the conversion control unit 103.

Next, the operation of the embodiment will be explained using the example shown in FIG. For simplicity, the target of conversion is a single clause. FIG. 2 shows a state in which a Roman character string input by the user to generate the expression "Fortran" is held in the input buffer 101.
In the figure, the halftone dotted portion has been determined by the character string feature detection unit 102 to be acceptable as a Roman character notation for representing Japanese.
The position of the arrow (↑) is a character position determined by the character string feature detection unit 102 to be unacceptable as representing Japanese. When a conversion command is input to the conversion control unit 103 at the point when the Roman characters have been input to the position of the arrow (↓), the conversion control unit 1
03 determines that the input character string represents a non-Japanese word based on the determination result of the character string feature detection unit 102, and starts the sub-language conversion processing unit 105.
The sub-language conversion processing unit 105 searches the sub-language conversion dictionary 107 using the character string in the input buffer 101 as a key, and when the word "fortran" is found, converts the notation and information 302 of the end position in the input buffer 101. It is returned to section 103. Conversion control unit 1
03, since there are still character strings after the end position of the word of the sub language input from the sub language conversion processing unit 105, the main language conversion processing unit is activated to obtain the adjunct word "wo". As a result of the above processing, the string "fortranwo" in the input buffer is changed to the output "fortranwo".
is converted to ``.

(Effects of the Invention) As described above, the present invention uses the restrictive conditions regarding the appearance of characters in the Roman alphabet, which are commonly used in Japanese input, to automatically change the currently input word to the Japanese notation. Automatically determines whether the language has a notation in another language or not. Therefore,
According to the device of the present invention, when inputting Japanese text in which words of other languages frequently appear, the user does not have to worry about changing the mode, etc., which is efficient and reduces mental burden. It becomes possible to input a small amount of Japanese.

[Brief explanation of the drawing]

FIG. 1 is a diagram illustrating an embodiment of the Romaji-Kanji conversion device according to the present invention. FIG. 2 is a diagram showing the contents of an input buffer in which the user's Roman alphabet input is held. FIG. 3 is a diagram showing word notation and end position information outputted to the conversion control unit 103 by the sub-language conversion processing unit 105. In the figure, 101 is an input buffer, 102 is a character string feature detection unit, 103 is a conversion control unit, and 104
105 is a main language conversion processing unit, 105 is a sub-language conversion processing unit, 106 is a main language conversion dictionary, 107 is a sub-language conversion dictionary, 301 is an input buffer, and 302 is information on word notation and end position, respectively.

Claims (1)

[Claims] 1. A character string admissibility determining means for determining whether or not the ordinal co-occurrence relationship of characters in an input character string is permissible in Japanese, and if the character string admissibility determining means Japanese language conversion means that performs a process of converting the input character string into Japanese;
A Roman-to-Kanji conversion device comprising: other language conversion means for converting the input character string into another language if the input character string is not accepted by the character string admissibility determination means.