JPS61235978A - Character string correction system - Google Patents

Abstract

PURPOSE:To output sure Kanji(Chinese character)-Kana(Japanese syllabary) character string candidate by providing a character string output means deciding a required character string and outputting it when plural Kanji-Kana character strings are obtained as the result of Kana-Kanji conversion. CONSTITUTION:A Kana-Kanji converting means 6 applies Kana-Kanji conversion while repeating the retrieval of a word dictionary 5 and a grammar detection based on a reading character string under the control of a control section 1. The means 6 extracts a corresponding word and applies grammar detection, leaves proper phrases only and excludes inadequate phrases. A word output means 7 decides a desired character string and outputs it with the method that it is selected by the user while all candidates given from the means 6 are displayed.

Description

[Detailed Description of the Invention] (Industrial Application Field) The present invention relates to a Japanese input method, and more specifically, a method for obtaining a correct Kanji-Kana character string based on an input Kanji-Kana character string. This relates to a character string correction method.

(Prior art and its problems) In recent years, with the spread of Japanese input devices such as word processors, inexperienced users with little experience in Japanese input and general users who are not familiar with the usage of kanji Opportunities to use these devices are increasing.

Among the Japanese input methods, there is the Bentouch input method, which inputs a kanji-kana character string while picking up kanji on the board with a touch pen, the code input method, which inputs the code corresponding to each kanji, and the actual writing of kanji by hand. Online handwritten character recognition methods, which allow users to directly input the desired kanji, are effective, but because the input unit is characters, the input kanji character strings may not be correct as Japanese. The user must guarantee that this is the case.

For this reason, for example, even if a user mistakenly inputs the character string "go slowly" as "go slowly," he or she would not be able to tell that this writing is incorrect.

(Object of the Invention) The object of the present invention is to provide a character string correction method for outputting a probable Kanji-Kana character string candidate in place of the incorrect spelling or usage of a Kanji-Kana character string that the user wants to input. Our goal is to provide the following.

(Structure of the Invention) According to the present invention, the kana characters in the input kanji-kana character string and the kanji characters obtained by searching a kanji dictionary containing the readings of the kanji characters based on each kanji character string. A reading character string creation means for creating one or more reading character strings by combining readings, a kana-kanji conversion means for converting the reading character string into kana-kanji, and a plurality of kanji-kana characters as a result of the conversion. It is possible to obtain a character string correction method characterized by comprising character string output means for determining and outputting a required character string when a string is obtained.

(Example) Examples of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows an embodiment of the present invention. 1 is a control unit that manages the overall control, 2 is an input means for inputting a kanji-kana character string, 3 is a kanji dictionary that stores the readings of kanji, and 4 is a kanji dictionary that stores kanji dictionary 3. 5 is a reading character string creation means for creating one or more reading character strings by combining the reading of each kanji obtained by searching with the kana characters in the input character string, 6 is a word dictionary containing notation and part-of-speech information, 6 is a kana-kanji conversion means that converts reading character strings into kana-kanji while repeating searches of the word dictionary 5 and grammar tests, and 7 is a kana-kanji conversion means that converts multiple kana-kanji characters as a result of kana-kanji conversion. When Kanji/Kana character string candidates are obtained, this is a character string output means that instructs and outputs the required character string from among the word candidates. Further, using the embodiment shown in FIG. 1, FIG. 2 shows an outline of the flow of obtaining a correct kanji-kana character string from an input kanji-kana character string.

Here, we will explain the procedure for correcting the above-mentioned "to go" as an example. However, for convenience of explanation, it is assumed that the kana-kanji converting means 6 is configured to receive a kana character string separated into clauses.

First, the kanji-kana character string ``gokusuru'' is input from the input means 2, for example, by direct input character by character or by extracting phrases from the kanji-kana character string that has already been input (
101 in Figure 2).

In addition to the above-mentioned Bentouch input, you can input characters by character.
This can also be done by display selection by giving the reading of a kanji and selecting from a group of kanji displayed, by using text data that has already been input, or by inputting a character string from an external source. In addition, the extraction of phrases can be performed, for example, in Information Processing Society of Japan Computational Linguistics Study Group Material 15-2 "Certification of Japanese phrases" (1
Automatic processing using methods such as those described in 978),
This can be done by the user's instructions to specify a range. Also,
It is also possible to check in advance whether the extracted phrase strings are correct as Japanese or not using a dictionary lookup grammar test or the like, and to process only those that are not phrases.

The character string "Yuyokusuru" inputted from the input means 2 is sent to the reading character string creation means 4 under the control of the control section 1. The reading character string creation means 4 searches the kanji dictionary 3 based on the kanji ``jo'' and ``gyo'' that make up the obtained character string ``yugyo suru'' and extracts the reading of each kanji (102); These and the remaining kana characters "su" and "ru" are combined to create a possible reading character string (103). Here, if the kanji dictionary 3 contains "fjiyo" and "ji" as the pronunciations of "extra" and "kou" and "kiyo" as the pronunciations of "line", the pronunciation character string creation means 4 , From these pronunciations, the pronunciation strings ``Jiyokosuru'', ``Jiyokiyousuru'', ``Shikosuru'',
``Jigiyosuru'' is created and passed to the kana-kanji conversion means 6.

Under the control of the control unit 1, the kana-kanji conversion means 6 performs kana-kanji conversion based on the reading character string while repeatedly searching the word dictionary 5 and checking the grammar (104).

Here, in the word dictionary 5, "woman worker" is written for "jiyoko".
(noun), ``slow progress'' (su-modular noun) is ``shikou,'' while ``matters'' (noun), ``season'' (noun), and ``statute of limitations'' (noun) are ``jigiyo.''"business" (
Suppose that it contains ``terrain'' (noun) and ``topography'' (noun).

The kana-kanji conversion means 6 extracts these words and performs a grammar test, leaving only those that are correct as clauses and eliminating inappropriate ones. In this example, since ``suru'' is present at the end of the word, only ``suru'', which is the mundane noun ``suru'' connected to ``suru'', remains and is passed to the character string determining means 7.

The word output means 7 determines a desired character string by, for example, displaying all the candidates passed from the kana-kanji conversion means 6 and having the user select one (105), and outputs it (106). In this example, there is only one "go slowly", so this can be output.

Through these series of steps, we can change "to go slowly" to "to go slowly"
can be corrected.

(Effect of the invention) By using the present invention, even when a user makes a spelling or usage error in a Kanji/Kana character string, the correct character string candidate is automatically extracted and presented. It is possible to reduce the work involved in inputting characters and correcting typographical errors.

[Brief explanation of the drawing]

FIG. 1 is a block diagram for explaining an embodiment of the present invention, and FIG. 2 is a flowchart showing an outline of the process of obtaining a correct character string from an input Kanji/Kana character string. In the figure, 1... control unit, 2... input means, 3...
...Kanji dictionary, 4... Reading string creation means, 5...
Word dictionary, 6... Kana-Kanji conversion means, 7... Character string output means. Figure 2

Claims (1)

[Claims] 1. Combining the kana characters in the input kanji-kana character string and the readings of the kanji obtained by searching a kanji dictionary containing the readings of kanji based on each kanji, 1.
A reading character string creation means for creating one or more reading character strings, a kana-kanji conversion means for converting the reading character string into kana-kanji, and a case where a plurality of kanji-kana character strings are obtained as a result of the conversion. A character string correction method characterized by comprising a character string output means for determining and outputting a character string required for a character string.