JPS6022226A - Kana/kanji converter - Google Patents

Abstract

PURPOSE:To decrease the KANA(Japanese syllabary)/KANJI(Chinese character) conversion processing time by comparing the number of KANA characters at and after a word retrieving position within an input KANA train with the reading length of the word stored in a word information temporary storage part and displaying the input/ output information after replacement based on the result of comparison. CONSTITUTION:A keyboard 1 which decides an input key is provided together with a memory part 2 for input KANA train information, a replacement part 3 for said train information, a word dictionary 4 and a word retrieving part 5. Then the word with which the number of KANA characters is detected at and after a word retrieving position is retrieved 5 among the words having the reading length smaller than the number of KANA characters within an input KANA train every time an input KANA is added or deleted. Then the reading lengths of the words with which said number of KANA characters is detected are compared 7 with each other. Based on the result of comparison 7, the information supplied selectively and the output KANJI information are replaced 9. In addition, the input KANA train information and the output KANJI train information are replaced 9 and displayed 10 when the fixture is indicated for the KANA/KANJI conversion result.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a kana-kanji conversion device used for creating and editing Japanese sentences.

Conventional configurations and their problems In recent years, kana-kanji conversion devices have been adopted as a method that allows the operator to create Japanese sentences relatively quickly without requiring special skill. Among these, kana-kanji conversion devices that use solid writing input are attracting attention because they do not require the operator to have grammatical knowledge of phrases, etc., and require only a small number of manual keystrokes, so they place less burden on the operator.

However, in conventional kana-kanji conversion devices for solid input, the method is to start kana-kanji conversion by inputting punctuation marks, etc., after inputting a kana string, and convert the entire kana string input so far into a kanji string at once. Because of this, there was a problem in that it took a long time from the final key input to the display of the conversion result.

On the other hand, there is a kana-kanji conversion device that converts the kana string input up to that point into one word every time kana is input, and displays the conversion result. However, expanding the unit of the Kana sequence to solid writing increases the number of conversion errors, and to correct it, for example, the input Kana sequence must be divided into shorter units and re-inputted. This has not been realized due to problems such as operational difficulties such as necessity.

Purpose of the Invention In view of the above-mentioned conventional problems, the present invention provides a method that reduces the time required from the final key manually to the display of the conversion result, and prevents erroneous conversion even when kana is input in solid writing. To provide a kana-kanji conversion device that can correct the errors extremely easily and quickly even when the errors occur.

Structure of the Invention The present invention provides a keyboard for determining inputted keys, which includes keys for at least inputting kana, deleting kana, and instructing confirmation of the result of kana-kanji conversion; and a word search position as input kana string information; means for updating the input kana string information; a word dictionary having at least a reading and a kanji string for each word as word information; Activated by inputting and deleting kana, the word with the longest matching pronunciation with the kana string after the word search position among the words with the pronunciation length of the kana number taken after the word search position in the input kana string is created in the word dictionary. means for temporarily storing at least the reading length, kanji string, and number of kanji of the detected word as word information; and means for temporarily storing at least the reading length of the detected word and the word search position in the input kana string. means for comparing the number with a kana number, means for storing at least the number of words, the number of kanji for each word, and the kanji of the output kanji string as output kanji string information, means for updating the output kanji string information, and an output kanji string. It is a kana-kanji conversion device equipped with a means for displaying a word, and it performs a word search every time a kana is input or deleted, and displays the pronunciation length of the detected word and the number of kana after the word search position in the input kana string. If different, the word search position in the input kana string is shifted by the reading length of the detected word, and 1 is added to the number of words in the output kanji string, and each time a word is detected, the kanji string of the word is By displaying this in addition to the output kanji string, adding 1 to the number of words in the output kanji string when instructed to finalize the kana-kanji conversion result, and initializing the input kana string information, the final The purpose is to shorten the time required from keystrokes to display the results of kana-kanji conversion, and to make it possible to correct erroneous conversions extremely easily and quickly.

DESCRIPTION OF EMBODIMENTS FIG. 1 is a block diagram showing the configuration of a kana-kanji conversion device in an embodiment of the present invention.

In the figure, reference numeral 1 denotes a keyboard, which accepts and judges kana input, deletion of kana, instruction to confirm the result of kana-kanji conversion, and instruction to end kana-kanji conversion. Reference numeral 2 denotes an input kana string information storage section, which stores the number of kana in the input kana string and the position at which a word search should be performed.

Reference numeral 3 denotes an input kana string information updating section, which updates the contents of the input kana string information storage section 2 according to the output of the keyboard 1 and the output of a kana number word reading length comparison section 7, which will be described later. 4 is a word dictionary, which stores the pronunciation and kanji string of each word as word information. Reference numeral 6 is a word search unit, which selects words that have a pronunciation that matches the longest kana string after the word search position among the words that have the pronunciation length of the kana number after the word search position in the input kana string. Search within 4. Reference numeral 6 denotes a word information temporary storage section, which temporarily stores the pronunciation length, kanji string, and number of kanji characters of the word detected by the word search section 5 as word information. 7 is a kana number word reading length comparison unit, which compares the kana imaginary numbers after the word search position in the input kana string with the reading length of the word stored in the word information temporary storage unit 6, and outputs the comparison result. do. 8 is an output kanji string information storage unit;
The number of words forming the output kanji string, the number of kanji characters for each word, and the kanji string are stored as output kanji string information. Reference numeral 9 denotes an output kanji string information update section, which updates the word information temporary storage section 6 according to the output of the kana number word reading length comparison section 7 and the output of the keyboard 1.
After adding or replacing the Kanji string of the word stored in the output Kanji string information storage section 8 as the Kanji string of the final word, the number of Kanji characters of the corresponding word is updated or the number of words is updated.

1° is a display section, which displays the output kanji string in the output kanji string information storage section 8.

The operation of the kana-kanji conversion apparatus of this embodiment configured as described above will be described below.

FIG. 2 is a flow diagram showing the operation of the kana-kanji conversion device of this embodiment. In the figure (branch), the output kanji string information update section 9
initializes the number of words in the output Kanji string information storage section 8 to one. Then move on to (a). In (a), the input kana string information updating section 3 initializes the number of kana in the input kana string information storage section 2 to Q, and initializes the word search position to the value ○ indicating the beginning of the kana string. Then move on to (tsu).

(1) is the state of waiting for a key to be pressed on the keyboard 1, and when a key is pressed manually, the state moves to (-I:). (to),
(9).

(In GA, keyboard 1 determines whether the pressed keys are the Kana key, the Kana deletion key, or the Kana-Kanji conversion result confirmation key.If it is the Kana key, If it is a kana-kanji conversion result finalization instruction key, go to YES), and if neither of these is the case, it is assumed that it is a kana-kanji conversion end instruction key, and the process ends. (In Wari, the input kana column information update unit 3 updates the contents of the input kana column information storage unit 2 in response to the kana input. More specifically, the input kana column information is updated based on the kana number of the input kana column. It is added to the end and 1 is added to the imaginary number.Then, the process moves to (ri).In (shi), the input kana string information updating section 3 updates the contents of the input kana string information storage section 2 in response to the deletion of kana. More specifically, 1 is subtracted from the kana number. After that, the process moves to the (column). In (k) and (g), the word search section 5 searches the word dictionary 4 based on the contents of the input kana column information storage section 2.
Perform a word search within.

”: More specifically, among the words that have a pronunciation length of kana number taken after the word search position in the input kana sequence, the word that has the longest matching pronunciation with the kana sequence after the word search position in the input kana sequence Search for words. After that, if a word could not be detected, the process moves to (1), and if a word was detected, the reading length, number of kanji characters, and kanji string of the detected word are stored as word information in the word information temporary storage unit 6, and then θ ). (B) After the stupid number word reading length comparison unit 7 calculates the kana number after the word search position from the input kana string information storage unit 2, the word search position and the kana number in the input kana string, Compare this calculation result with the reading length of the word stored in the word information temporary storage unit 6. If they are equal, go to (a); if they are not equal, go to (e).
Move to. (In a, the output kanji string information update section 9 transfers the kanji string of the word stored in the word information temporary storage section 6 to the output kanji string information storage section 8, and updates the number of kanji characters.To be more specific, , based on the number of words in the output Kanji string, transfer the Kanji string of the corresponding word as the Kanji string of the final word, and set the number of Kanji characters of the transferred Kanji string as the number of Kanji characters of the final word. Then move to ((B)). In ), the display unit 1o displays the output kanji string in the output kanji string information storage unit 8 as a kana-kanji conversion result.Then, the process moves to (t).(In (1), the input kana string information update unit 3 displays the input kana string The pronunciation length of the word stored in the word information temporary storage unit 6 is added to the value indicating the word search position in the input kana string in the information storage unit 2. Then, the process moves to (7). The string information update unit 9 adds 1 to the number of words in the output kanji string information storage unit 8. Then, the process moves to OA. Add 1 to . Then move to (a).

Hereinafter, the operation of the Kana-Kanji conversion device of this embodiment will be specifically explained using an input example.

FIG. 3 is a diagram showing the contents of the input kana string information storage section 2 and the contents of the output kanji string information storage section 8, for the purpose of explaining the operation of this embodiment in the case of the first input example. In the figure, the A-0 column shows information on the input kana string, and the D and H columns show information on the output kanji string. A is a kana number, and B is a word search position, where the beginning of the input kana string is ○. C is a kana column. Moreover, D is the number of words. E is the number of kanji characters in the first word, F is the second word, and G is the number of kanji characters in the third word. Therefore, an area for three words is secured. H is a string of kanji, and kana may be stored as a kanji in a broader sense. Furthermore, a-'At in the figure indicates each point in time that will appear in the explanation below.

First, the information on the input kana string and output kanji string is initialized in FIGS. 2 (sub) and (a) (FIG. 3 a).

Here, when kana "ke" is input from the keyboard 1, the input kana column information update section 3 updates the input kana column information storage section 2 after going through the steps in FIG. 2 (-I:). Store the kana ``ke'' at the end of the input kana string, in this case at the beginning, and add 1 to the kana number to make it 1 (A-C in Figure 3).In Figure 2 □A , the word search unit 5 searches the word dictionary 4 for the longest matching word within the range of the kana number 1 from the word search position 0 in the input kana string, that is, the beginning of the kana string.The word "hair" is the longest matching word. If detected, the word information temporary storage unit 6
, the reading length 1, the number of kanji characters 1, and the kanji string "mo" are stored as word information of the word "mo". After going through the steps in FIG. 2 (k), the dekana number word reading length comparison unit 7 in FIG. It compares the pronunciation length 1 of the word "hair" and informs the output kanji string information update unit 9 that they are equal, and moves on to Fig. 2 (-9. In Fig. 2 (-2, the output kanji string information The update unit 9 uses the kanji string 1 马 of the word stored in the word information temporary storage unit 6 as the final word in the output kanji string information storage unit 8, in this case the first word, of the output kanji string. Transfer to the beginning and store 1 as the number of kanji in the first word (
Figure 3 (D-H). In FIG. 2(S), the display section 1o displays the kanji string in the output kanji string information storage section 8, and the process moves to FIG. 2(T). I wonder if it will continue? The input is "Yes".

[If J, r<J, the contents of the information storage units 2 and 8 of the input kana string and output kanji string at each point in time are as shown in FIGS. 3c and d by the same flow of operations.

It becomes as shown in e. If the kana "wo" is subsequently input, the operation from FIG. 2 (b) onwards will be different from the previous one. The contents of the input kana string information storage section 2 at this point are A in Fig. 3 f.
-C, and as a result of the word search, the word information temporary storage unit 6 contains the reading length 4 of the word "plan" and the kanji "
It is assumed that "Plan" and the number of kanji characters 2 are stored. In the second section, the Kakasakana word reading length comparison unit 7 compares the kana number 5 after the word search position in the input kana string and the reading length 4 of the word "plan", and determines that the two are different. is notified to the input kana string information updating section 3 and the output kanji string information updating section 9, and the information shown in Fig. 2 (-@
Move to. In FIG. 2 (e), the input kana column information update unit 3
By adding the reading length 4 of the word 1 plan stored in the word information temporary storage unit 6 to the word search position 0 in the input kana string information storage unit 2, 4, that is, the position of kana ``wo'' is determined. Make it point.

Next, in FIG. 2 (n), the output kanji string information updating section 9 adds 1 to the number of words in the output kanji string information storage section 8 to make it 2 (see qL in FIG. Figure 2 (For the sword, the word search unit 6 performs a word search on the kana string after the position of “wo” indicated by 4 in the input kana string, detects the word “wo”, and stores the word information temporarily. The reading length 1 is stored in the storage unit 6.
, the kanji string "wo", and the number of kanji (1) are stored as word information. After that, move on to (shi) through Figure 2 (k) and (to).
In Figure (a), the output kanji string information update unit 9 outputs the kanji string “wo” of the word stored in the word information temporary storage unit 6, the last word in the kanji string information storage unit 8, in this case the second As a word, it is transferred from position 3, which is the next position after the number of Kanji characters 2 in the first word of the output Kanji string, and 1 is stored as the number of Kanji characters in the second word (Fig. 3h).

If the kana "su" is subsequently input, the same operation as when the kana "wo" is input is performed, and the information of the word "su" is added as the third word to the output kanji string information storage section 8. (Figure 3 i). Furthermore, Kana “IJ, rL, J, r”
are input in sequence, the comparison results of the kana number word reading length comparison unit 7 in Figure 2 (A) are equal, so the operation moves to Figure 2 (Si), and the conversion process for Kana "N" is completed. At this point, the state shown in Fig. 3j is reached.

As explained above using the input example of cylinder 1, according to this embodiment, a word search is performed every time Gana is input for solid writing input, and when a word is detected, the input If the number of kana after the word search position in the kana string and the reading length of the detected word are different, shift the word search position in the input kana string by the reading length of the detected word, and change the output kanji string. By adding 1 to the number of words, the input string can be divided into words and displayed sequentially.

In the final kana 2-person manual example, after inputting the kana "n", the conversion time is short because only the kana string after the kana "su" is converted.

FIG. 4 is a diagram showing the contents of the input kana string information storage section 2 and the contents of the output kanji string information storage section 8, for explaining the operation of this embodiment according to the second input example. The meanings of column -H and rows 2 to 0 are the same as in the case of FIG.

When the kana ``sa'', ``n'', ``ka'', and ``i'' are input in sequence, Figure 4 shows ~ in the same way that Figure 3 and B-6 were created.
0 is created. In this case, it is assumed that the operator desires a conversion result of "participating committee member" for the kana input "sankaiin". However, after inputting kana ``i'' next to kana ``ka'', ``adjournment'' is displayed based on the contents of Figure 4 0. However, when the operator realized that the conversion was incorrect and pressed the Kana delete key from Keyboard 1, the result in Figure 2 (Tsu)
The process moves from (2) to (3) to (a). In FIG. 2(a), the input kana string information updating unit 2 subtracts 1 from 4 to set the number of kana in the input kana string information storage unit 2 to 3, thereby changing the number of kana at the end of the input kana string. ” to be deleted.

Thereafter, the contents of the output kanji string information storage section 8 become as shown in Figure 4 p, and ``participation'' is displayed, and the process is as shown in Figure 2, using the same process flow as when kana ``KA'' is input. Move to (tsu). Here, when the operator confirms the display and inputs the conversion result confirmation instruction key from the keyboard 1, the process starts as shown in FIG.

Move to (through force). In FIG. 2), the output kanji string information updating section 9 adds the number of words in the output kanji string information storage section 8 to 1 and sets it to 2. Thereafter, the process moves to (Figure 4) via Figure 2 (A) (Figure 4 (Q)). Kana continues to be ``yes''.

When "i" and "n" are input in sequence, the contents of the information storage units 2 and 8 of the input kana string and output kanji string change to r % t in Figure 4 by the same operation as before, and the final kana string is changed to r % t in Figure 4. When the processing for "nj" is completed, "participating member" is displayed, and the operator can obtain the desired conversion result "participating member".

As explained above using the input example of cylinder 2, according to this embodiment, a word search is performed every time a kana is input or deleted for solid writing input, and each time a word is detected, a word search is performed. By displaying the output kanji string containing the word, adding 1 to the number of words in the output kanji string according to the instruction to confirm the kana-kanji conversion result, and initializing the input kana string information,
For example, when the operator sees the display ``Sankai,'' he can correct the incorrect conversion by simply adding the simple steps of pressing the Kana delete key, the Kana-Kanji conversion result confirmation key, and the Kana ``I'' key in sequence. can be modified to the desired conversion result.

Note that in this embodiment, for the sake of simplicity, the test of whether or not words can be connected is omitted, but the present invention can also be applied to cases where a connection test is performed based on the parts of speech of words, etc. It is possible. In addition, in this embodiment, one word dictionary is used.
However, the present invention is not limited to this. For example, independent words and attached words may be stored in separate dictionaries. Furthermore, in the present embodiment, for the sake of simplicity, the explanation of the case where there are homophones is omitted, but the present invention is also applicable to the case where there are homophones.

Effects of the Invention The kana-kanji conversion device of the present invention includes a keyboard for determining an input key, which is equipped with keys capable of at least inputting kana, deleting kana, and instructing confirmation of the result of kana-kanji conversion; means for storing kana and kana numbers and word search positions as input kana string information; means for updating the input kana string information; a word dictionary having at least a reading and a kanji string for each word as word information; (- triggered by the input and deletion of kana from the board, and the longest matching reading of the kana string after the word search position among words whose reading length is less than or equal to the number of kana after the word search position in the input kana string) means for searching a word in a word dictionary for a word having , and means for temporarily storing at least the reading length, kanji string, and number of kanji of the detected word as word information;
A means for comparing the reading length of the detected word with the number of kana after the word search position in the input kana string, and storing at least the number of words, the number of kanji for each word, and the kanji in the output kanji string as output kanji string information. By providing means for updating the output kanji string information, and means for displaying the output kanji string, the time required from the final input of the solid input kana string to the display of the kana kanji conversion result is reduced. It also has great practical effects, such as being able to quickly correct conversion errors with extremely simple operations by sequentially displaying the kana-kanji conversion results, deleting kana, and instructing to confirm the conversion results.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a kana-kanji conversion device according to an embodiment of the present invention, FIG. 2 is a flow diagram showing the operation of the same embodiment, and FIGS. 3 and 4 respectively show a first input example. FIG. 7 is a diagram showing the contents of the information storage unit of the input kana string and the output kanji string in the same embodiment in the case of the second input example. 1...Keyboard, 2...Input Kana column information storage unit, 3...Input Kana column information update unit, 4
...Word dictionary, 5...--Word search section, 6
...Word information temporary storage unit, 7...Kana number word reading length comparison unit, 8...Output Kanji string information storage unit, 9...Output Kanji string information update section, 1o・
...Display section.

Claims (1)

[Claims] A keyboard that is equipped with keys that can input at least kana, delete tables, and confirm the result of kana-kanji conversion, and that can determine the input key, and input at least the number of kana and kana in the input kana column and the word search position. means for storing as kana string information; means for updating the input kana string information; a word dictionary having at least a reading and a kanji string for each word as word information; and activation by inputting and deleting kana from a keyboard. means for searching a word dictionary for a word having a pronunciation that matches the longest kana string after the word search position among words having a pronunciation length of the kana number taken after the word search position in the input kana string; Means for temporarily storing at least the reading length, kanji string, and number of kanji of the detected word as word information, and comparing the reading length of the detected word with the number to determine whether it is input or after the word search position in the outer string. means for storing at least the number of words, the number of kanji for each word, and the kanji string of the output kanji string as output kanji string information; means for updating the output kanji string information; and means for displaying the output kanji string. , and each time an input kana is added or deleted, the longest matching pronunciation with the kana string after the word search position among the words with a pronunciation length of kana number removed after the word search position in the input kana string. The number of kana after the word search position in the input kana string is compared with the reading length of the detected word, and input kana string information and output kanji string information are selectively generated based on the comparison results. , and further updates input kana string information and output kanji string information when instructed to finalize the kana-kanji conversion result.