JPH0552542B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an improvement in a kana-kanji conversion device that converts kana readings, which are the subject of kana-kanji conversion, into character strings containing kanji and kana.

[Conventional technology]

Conventionally, as a method for converting solid input or character strings into kana-kanji, the method described in Hitachi Review, May 1980, Vol. ), a method that combines longest match and back track, as described in the Showa era.
Information Processing Society of Japan 19th National Conference Proceedings 5E-
4 Two-clause longest match method described in "Kana-Kanji Conversion System for Solid Text" (hereinafter referred to as Document (2)), 1981 Information Processing Society of Japan Computational Linguistics Study Group Materials
The method of minimizing the number of clauses described in 25-6 ``About a clause structure analysis algorithm using a table method and its efficiency'' (hereinafter referred to as reference (3)) is well known.

The method in Reference (1) extracts an independent word from the left end of the input (the beginning of the input kana character string) using the longest match, and performs bunsetsu conversion processing that assigns an attached word that can be grammatically connected to the next independent word. This process is repeated until the end of the kana character string is reached; if the right end is not reached, the process backtracks, attempts another phrase conversion, and moves on.

The method in document (2) extracts all possible conversion candidates for two clauses from the left end of the input in a brute force manner, selects the one with the longest sum of the lengths of the two clauses, and converts it into a single sentence. This is a method of conversion while determining the break point for each clause by repeating the same process using that point as the starting point.

The method in document (3) is to extract phrases starting from the left end in a brute-force manner, and select a combination that minimizes the number of phrases from among them to use as the conversion result.

However, when performing solid input, generally a large amount of Yomi-Kana data is input at once, and if some of the input Yomi-Kana data is not the reading that the operator intended, many input errors occur. There is a problem.

In the past, when an incorrect input like the one described above occurs, the correcting method is to delete one character at a time from the end of the entered character string, delete the incorrectly entered character, or insert new kana data, and then ,
There is a way to re-enter the original reading.

Other correction methods include masking the end of the input character string one character at a time, and using masking to delete incorrectly input characters other than the input kana that are excluded from kanji conversion, or adding new kana data to the reading. After inserting the mask, there is a method of canceling the masking described above.

[Problem that the invention seeks to solve]

However, according to the former method, after correcting the erroneously input character string, the character string that has been input once must be re-entered, making the operation cumbersome.

Also, according to the latter method, if there is an incorrectly input character at or near the beginning of a character string that has been input, you must mask it one character at a time up to the corresponding position at or near the beginning. Although this method is easier to operate than the first modification method described above, there is still room for improvement.

The present invention has been made in consideration of the above points, and its purpose is to provide a system where part of the reading data input to create kanji-kana mixed characters is not the reading intended by the operator. Even if the incorrectly entered character is at the beginning or near the beginning of a character string that has already been entered, the incorrectly entered character can be corrected faster than before. Fix it more quickly than
It can reduce the time required to create documents.
This paper attempts to provide an improved kana-kanji conversion device.

[Means for solving problems]

The purpose is to provide an input unit for inputting reading-kana data to be subjected to kana-kanji conversion, and means for storing reading-kana data input from the input unit in a storage unit;
A conversion processing unit that converts the input kana reading into a string containing kanji and kana, a control unit that controls the execution of kana-kanji conversion, a display unit that displays the conversion result, and the readings that are excluded from conversion and the conversion target readings. A kana-kanji conversion device comprising a mask control processing means for identifying kana and kanji on a display section, and a conversion target reading kana data input and retrieval processing means for extracting reading kana data to be converted from a storage section,
A first instruction means for masking all the kana readings inputted from the input section with one touch and excluding them from being converted into kanji at once, and converting by unmasking the input kana characters that have been specified as not to be converted one by one character by character. a second instruction means for changing the designation to the target of conversion, a third instruction means for masking the kana reading one character at a time and specifying it as not to be converted, and a third instruction means for specifying the reading of the part to be excluded from conversion. and a fourth instruction means for unmasking all kana characters to be converted at once with a single touch, and the input unit is configured to input readings other than the input readings excluded from the kanji conversion target by the masking process. This is achieved by providing a correction means that allows deletion of kana or insertion of new kana reading.

[Effect]

As described above, the kana-kanji conversion device of the present invention includes a first instruction means that excludes all the readings inputted from the input section from being converted into kanji at once, and inputs that are specified as not being subject to kanji conversion. A second instruction means for specifying reading kana as a conversion target character by character, a third instruction means for specifying reading kana character by character as being excluded from conversion target, and a third instruction means for specifying reading kana character by character to be excluded from conversion target. A fourth instruction means for converting all the kana readings of a part at once, and for identifying on the display unit the readings specified by the instruction means that are to be excluded from the conversion and the readings to be converted. Since the mask control processing means is provided with a mask control processing means and a conversion target pronunciation kana data input retrieval processing means for extracting pronunciation kana data to be converted from the storage unit, the above-mentioned first to fourth instruction means can be combined as follows. By doing so, it is possible to quickly correct the reading/kana data that has been input incorrectly. In other words, if there is an incorrect input in a character string that has been entered once, and the incorrectly input character is located at or near the end of the string, the characters from the end of the input string will be masked one by one as before. Then, after deleting erroneously input characters other than the input kana readings that are excluded from the kanji conversion target by masking, or inserting new kana reading data, the masking described above can be canceled one by one. On the other hand, if an incorrectly input character is located at or near the beginning of a character string that has been input, all the relevant characters can be masked at once with a single touch, and the masking process will remove them from kanji conversion. After deleting erroneously input characters other than the input reading (Kana) or inserting new reading (Kana) data, the aforementioned masking can be canceled all at once with a single touch.

〔Example〕

Hereinafter, the present invention will be explained based on one embodiment of the drawings.

FIG. 2 is a block diagram of the kanji conversion device according to the present invention, which includes a kana input section 1, a display section 4, a control section 2, a conversion processing section 5, and a storage section 3.

The input unit 1 is, for example, a keyboard for inputting kana input, designation of the kana reading to be converted, and inputting conversion, and the display unit 4 is comprised of, for example, a CRT or the like for displaying the input kana reading and the conversion results.
The control unit 2 is responsible for controlling the execution of kana-kanji conversion, and is configured using a CPU. The storage unit 3 is
Temporarily holds input Kana strings and data used in conversion. The conversion processing unit 5 converts the reading kana data to be converted into a sentence containing kanji and kana.

FIG. 3 shows an example of the configuration of the input section 1.

Note that the input device shown in FIG. 3 is not particularly limited to this example with respect to the key arrangement and the number of keys. Further, the reading/kana input key is generally composed of a plurality of keys such as kana, alpha, etc.

FIG. 1 is a flowchart showing the execution operation of the kana-kanji conversion process by the apparatus of the present invention, and the execution operation of the apparatus of the present invention will be explained below based on FIG.

The operator operates any of the keys shown in FIG. 3 from the input unit 1 to input processing commands or character data. The control unit 2 first determines in processing step 10 whether the conversion candidate after the conversion key is pressed is in an unestablished state (hereinafter referred to as an unestablished state) or in a non-established state (hereinafter referred to as an established state). Here, the initial settings of the establishment flag and the start pointer and end pointer, which will be explained later, are used for document creation, etc.
Set when entering a state that requires kana-kanji conversion. In this embodiment, an establishment flag "0", a start point pointer "1", and an end pointer "0" are stored as initial values. If it is in the established state, the input key is determined in processing steps 12 and 14, and in the case of the conversion target reading Kana data input key shown in FIG. The reading kana data inputted from the section 1 is stored in the storage section 3. Also, step 1
4, the first instruction means 91 shown in FIG.
Second instruction means 92, third instruction means 93, fourth instruction means
In the case of the instruction means 94, the process moves to mask control processing step 30, where the input character string is shaded or removed. Step 12
In the case where the conversion key 96 shown in FIG. 3 is pressed, the process moves to processing step 40, and the pronunciation of the conversion target, which is the part not shaded in the mask control processing step 30, is input into the conversion target pronunciation of the kana data input buffer. The conversion processing unit 5 (step 50) converts the reading kana data in the reading kana data input buffer to be converted into a sentence containing kanji and kana (step 60). At this time, in order to change the established state to an unestablished state, set the establishment flag to "1".
is set (step 62), and the conversion result is output to the display section 4.

When the establishment flag is "1", that is, in an unestablished state, it is determined in step 16 whether the input key is a non-conversion key, and if it is a non-conversion key, it is determined in step 70 that the Perform the conversion process,
The conversion result displayed on the display section 4 is erased. If it is not a non-conversion key, a candidate confirmation process is performed in step 80, and in order to change the unestablished state to an established state,
Set the establishment flag to “0” (step 8)
2).

The operations of the above-mentioned input reading kana data storage processing step 20, mask control processing step 30, and conversion target reading kana data input retrieval processing step 40 will be explained in more detail.

FIG. 4 is a flowchart showing the operation of the reading kana data input storage processing step 20 of FIG.

First, in processing steps 201 and 202, the key pressed by the operator is determined. If the input key is the conversion target pronunciation kana data input key shown in FIG. 3, in processing step 203, it is determined whether or not there are any already input pronunciation kana character strings that are shaded. Note that the start pointer (S), which will be described later, manages the start end of the input kana character string, and the end pointer (E) manages the end of the input kana character string.
The mask pointer (M) manages the beginning of the shaded kana character string in the input kana character string. For example, as shown in FIG. 5, if you input "He called the pheasant," the start pointer (S) stores "1" indicating the position of "ka",
The end pointer (E) is "9" indicating the position of "da", and since the mask pointer (M) has only a shaded reading, the value "10" of the end pointer +1 is stored. "Yonda" as shown in Figure 6
If there is shading, the mask pointer (M) is
``7'' indicating the position of ``yo'' is memorized. In processing step 203, if the input reading kana character string has no shading, that is, if a new reading kana is to be input after the last input reading kana, in processing step 208, the input reading kana memory buffer mask pointer ( A new kana reading is stored at the position indicated by M), and in processing steps 209 and 210, the values of the end pointer (E) and mask pointer (M) are incremented by "1". Processing step 20
If there is shading in step 3, a new kana reading is inserted at the position indicated by the mask pointer (M) of the input reading kana storage buffer in step 208. Then, in processing steps 209 and 210, the values of the end pointer (E) and mask pointer (M) are increased by "1". Processing step 20
If it is determined in step 1 that the input key is a backward key, it is determined in processing step 211 whether or not the input reading kana exists. If it exists, in processing step 212 it is determined whether the input reading kana has shading. judge. If there is no shading, the values of the end pointer (E) and mask pointer (M) are decreased by "1". If there is shading, processing steps 214, 215, and 216 move the yomi indicated by the mask pointer (M) of the input yomi kana memory buffer to the last input yomi by one character to the left, and remove the shading. Delete the input kana character before the beginning of a certain input character. Then, the values of the end pointer (E) and mask pointer (M) are decreased by "1". For example, if you input the Kana data input key "A" to be converted into the output state shown in Figure 6, the output state will be as shown in Figure 7, and if you input the backward key in Figure 7, the output state will change to the output state shown in Figure 6. Become.

Next, FIG. 8 shows the operational flow of the mask control process shown in step 30 in FIG. In processing steps 301, 305, 306, and 310, the first instruction means 91, second instruction means 92, third instruction means 93, and fourth instruction means 94 shown in FIG.
Determine which key is being input. In the case of input from the first instruction means, processing steps 302 and 30 are performed.
3, 304, the input kana indicated by the mask pointer (M) is shaded until the value of the mask pointer (M) becomes "1". In other words, all of the input kana readings are shaded. In the case of the fourth instruction means input, processing steps 314, 315, 3
At step 16, the shading of the input kana indicated by the mask pointer (M) is canceled until the value of the mask pointer (M) becomes larger than the value of the end pointer (E). That is, all the shaded input kana characters are unshaded. In addition, in the case of third instruction means input, in processing step 307, it is determined whether an input reading (kana) exists at the position indicated by the value decreased by "1" from the value of the mask pointer (M), and if it exists, In processing steps 308 and 309, the input pronunciation is shaded. In the case of second instruction means input, it is determined in processing step 311 whether an input reading kana exists at the position indicated by the mask pointer (M), and if it exists, processing steps 312,
In step 313, the shading of the input reading kana is removed. For example, if the first instruction means is input in the state shown in FIG. 5, the state will be as shown in FIG. 9, and if the fourth instruction means is input in the state shown in FIG. 9, the state will be in the state shown in FIG. Further, if the third instruction means is inputted in the state shown in FIG. 5, the state will become the state shown in FIG. 10, and if the second instruction means is further inputted, the state will return to the state shown in FIG. 5.

Therefore, by appropriately combining the first to fourth instruction means, incorrectly input reading and kana data can be quickly corrected. In other words, if there is an incorrect input in a character string that has been entered once, and the incorrectly input character is located at or near the end of the string, the characters from the end of the input string will be masked one by one as before. Then, after deleting erroneously input characters other than the input kana readings that are excluded from the kanji conversion target by masking, or inserting new kana reading data, the masking described above can be canceled one by one. On the other hand, if an incorrectly input character is located at or near the beginning of a character string that has been input, all the relevant characters can be masked at once with a single touch, and the masking process will remove them from kanji conversion. After deleting erroneously input characters other than the input kana reading, or inserting new kana reading data, the above-mentioned masking can be canceled all at once with a single touch, providing excellent operability.

If the input key is determined to be a conversion key in the processing step 12 of FIG. It is stored in a conversion buffer and converted into a sentence containing kanji and kana in conversion processing step 50. For example, if the conversion key is input in the output state shown in FIG.
A sentence containing kanji and kana is created, ``He wrote an article,'' and the result is output to the display unit 4 as shown in FIG. 11.

Candidate confirmation processing step 8 for the output result
The operation flow of 0 is shown in FIG.

At processing step 801, the initial value "0" of J is stored, and at step 802, it is determined whether there is shading in the input reading. If it doesn't exist,
Store "0" in the terminal (E) and "1" in (M),
Perform initial settings. If there is shading, move the input yomi kana with the shading to the beginning of the input yomi kana and proceed to processing steps 804, 805, 806, 8.
At step 07, the shaded character count processing step 808 is stored in the end pointer (E), and the processing step 80 is stored.
At step 9, the number of shaded characters +1 is stored in the mask pointer (M), and initial settings are performed. For example, the 11th
In the output state of the figure, the conversion target reading Kana input key "ga"
When input, the output state shown in FIG. 13 is obtained.

Incidentally, Japanese Patent Application Laid-Open No. 60-252983 discloses a technique in which reading kana data in the range to the right including the character at which the cursor is positioned is automatically converted into kana-kanji on a sentence-by-sentence basis.

However, according to the technology described in the above publication, the reading kana data in the range to the right of the cursor is automatically converted into kana-kanji on a sentence-by-sentence basis. If the data is entered incorrectly again, there is a risk that the reading/kana data in the range to the right of the cursor will be converted into a sentence containing kanji/kana that the operator did not intend.In such a case, all the erroneously converted kana/kanji It is undeniable that this involves the hassle of having to make corrections again.

In contrast, the input section includes a first instruction means for masking all the kana readings inputted from the input section and excluding them from the kanji conversion target at once, and the input section has a first instruction means for masking all the kana readings inputted from the input section, and the input section has a first instruction means for masking all the kana readings input from the input section, and excludes them from the kanji conversion target by the masking process. According to the present invention, which is equipped with a correction means that can delete Kana readings other than the input reading Kana that has been inputted, or insert new Kana readings, the input Kana data that has been subjected to the mask processing is temporarily subject to Kana-Kanji conversion. Since it is in a fixed state that cannot be changed, it does not cause any problems as mentioned above, and only the minimum necessary character corrections are required.
The time required for document creation can be reduced.

〔Effect of the invention〕

The present invention is as described above, and as is clear from the explanation of the illustrated embodiment, according to the present invention, part of the reading kana data input to create kanji-kana mixed characters is changed to the reading intended by the operator. If the character is not in kana, the incorrectly entered character can be corrected faster than before, especially at the beginning of the character string once entered.
Or to obtain an improved kana-kanji conversion device that can correct the erroneously inputted character more quickly than before even if there is an erroneously inputted character near the beginning, thereby shortening the time required to create a document. be able to.

[Brief explanation of the drawing]

The drawings show an embodiment of the present invention; FIG. 1 is a flowchart showing the execution operation of the kana-kanji conversion process by the device of the present invention, FIG. 2 is a block diagram of the kana-kanji conversion device according to the present invention, and FIG. 3 is an input FIG. 4 is a flowchart showing the operation of reading kana data input and storage processing; FIGS. 5 to 7, 9, and 10 are diagrams showing screen display information; FIG.
11 is a flowchart showing the operation of the mask control process, FIG. 11 is a diagram showing the screen display information of the conversion result in FIG. 6, FIG. 12 is a flowchart showing the operation of the confirmation process after conversion, and FIG. The figure is a diagram showing an example of screen display information after confirmation processing. 1...Input section, 2...Control section, 3...Storage section,
4...display unit, 5...conversion processing unit.

Claims (1)

[Scope of Claims] 1. An input section for inputting reading-kana data to be subjected to kana-kanji conversion, means for storing reading-kana data inputted from the input section in a storage section, and converting the input reading-kana into kanji-kana mixed characters. A conversion processing unit that converts into columns, a control unit that controls the execution of kana-kanji conversion, a display unit that displays the conversion results, and a display unit that identifies kana to be excluded from conversion and kana to be converted. In the kana-kanji conversion device, the kana-kanji conversion device has a mask control processing means, and a conversion target reading kana data input retrieval processing means for extracting the reading kana data to be converted from the storage unit, all of the reading kana input from the input unit can be inputted with one touch. A first instruction means for masking and excluding the kanji from being converted at once; a second instruction means for unmasking the input kana characters specified as not for conversion one by one and changing them to conversion targets; A third instruction method for masking kana characters one by one to exclude them from conversion, and a one-touch method to unmask all kana readings in the part specified to be excluded from conversion and convert them all at once. and a fourth instruction means for making the kanji conversion target, and the input unit is capable of deleting kana readings other than the input readings excluded from the kanji conversion target by masking, or inserting new kana readings. A kana-kanji conversion device characterized by comprising a correction means for.