JPS63155262A - Japanese processing system - Google Patents

Abstract

PURPOSE:To easily correct punctuation which is performed automatically by a device by providing an editing function for rendering which is as long as a predetermined number of characters. CONSTITUTION:When an instruction 11 for corrective reconversion is inputted over a look at a conversion result displayed on a CRT, plural candidates of 'KANJI' (Chinese character) conversion 12 are generated. The generated variable table candidates are displayed 13 according to priority. When an operator makes an instruction 14 of a selection of a variable table candidate string, the punctuation is varied according to the rendering length of the selected candidates. The remaining rendering is reconverted 15 automatically according to the varied new punctuation and the result is displayed 16 again.

Description

[Detailed Description of the Invention] Technical Field The present invention relates to a Japanese language processing method, and in particular to a Japanese language processing device (hereinafter also simply referred to as "device") that uses a solid writing input method, to easily correct the divisions of words, nine phrases, etc. This paper relates to a Japanese language processing method that can be used to process Japanese characters.

The solid writing input method in the conventional kana-kanji conversion method means that the operator does not have to specify the boundaries between phrases or words.
This is a method that the device automatically separates.

The problem here is that when the delimitation made by the device is different from what the operator intended (incorrect conversion), the operation for the operator to correct it is surprisingly troublesome. It is.

One of the reasons for this is that the conversion candidate matrix that the operator refers to when correcting the delimiter at the time of the above-mentioned erroneous conversion is usually displayed in a corner such as the lower right corner of the screen. That is, the problem is that the amount of movement of the operator's line of sight is large.

Purpose The present invention was made in view of the above circumstances, and its purpose is to solve the above-mentioned problems in conventional Japanese language processing methods, and to easily correct the delimiters automatically performed by the device. The purpose of this invention is to provide a Japanese language processing method that can be used to process Japanese characters.

Configuration The above object of the present invention is to provide a Japanese language processing device having a kana-kanji conversion means, a means for instructing correction re-conversion, and a variable length candidate creation means for creating variable length candidates of pronunciation in addition to homophones as conversion candidates. and when a reconversion instruction is given by the correction reconversion instruction means, the variable length candidate creation means displays a variable length candidate in the vicinity of the line where the correction is located, and also displays a variable length candidate for a predetermined number of characters. This is achieved by a Japanese language processing method that has an editing function for the reading of the character string and converts the remaining character strings according to the candidate selected by the operator from among the variable length candidates.

Hereinafter, the configuration of the present invention will be explained in more detail based on examples.

FIG. 2 is a hardware configuration diagram of a Japanese language processing device showing an embodiment of the present invention. In FIG. 0, 1 is a keyboard;
2 is a keyboard controller, 3 is a key person hippo sofa,
4 is a key manual buffer controller, 5 is a candidate buffer, 6 is a candidate buffer 2 controller, 7 is a CRT controller, 8itCRT, 9 is a fleeting Kanji conversion module,
10 indicates a CPU.

The outline of the operation of the Japanese language processing device of this embodiment is as follows. That is, a solid character string input from the keyboard 1 is temporarily stored in the keypad 3 under the control of the keypad buffer controller 4. Then, the words are automatically divided into two phrases using punctuation marks and the like as triggers, and the words are converted into kana-mixed sentences by the kana-kanji conversion module 9. Each converted candidate is sent to the candidate buffer controller 6
are stored in the candidate buffer 5 under the control of the CRT controller 7, and the
Displayed on RT8.

Hereinafter, the display operation in the reconversion operation, which is a characteristic operation of this embodiment, will be explained based on the flowchart shown in FIG. 3 and the specific examples shown in FIGS. 4(a) to (e).

The operator looks at the conversion result displayed on the CRT 8, and if the result is different from the intended one, instructs correction and re-conversion (step 11). When instructed to perform corrective re-conversion, the device creates plurality of Kanji conversion candidates within the Kana-Kanji conversion module 9 (step 12). In addition, shorter length candidates and longer length candidates are also included (hereinafter referred to as "variable length candidates").

Next, among the created variable length candidates, candidates with the same "yomi" length as before reconversion are displayed with the highest priority, and after displaying homophones, the "yomi" length is changed from the longest to the shortest. The information is displayed (step 13).

When the operator selects the desired one from the displayed variable length candidate string (step 14), the delimiter is changed according to the "reading" length of the selected candidate. Then, the remaining "yomi" are automatically reconverted according to the new delimiter (step 15), and the results are redisplayed (
Step 16).

Specific examples of the display in the above operation are shown in Fig. 4 (a) to (e).
It was shown to. Note that 21.22 in FIGS. 4(Q) and (d) show examples of variable length candidate sequences.

FIG. 4(a) shows a state in which the input sentence ``It's nice weather now.'' is converted into ``It's nice weather now.'' Here, 23 indicates a notation, and 24 indicates a first cursor. In addition, FIG. 4(b) shows that the first cursor 24 is the target of change in order to change "Ima wa" in the conversion sentence "It's nice weather now." to "Today wa". This shows the state where it has been moved to the beginning of the phrase "nowa".

Below, we will explain the display situation when changing ``Ima wa'' in the above conversion sentence ``It's nice weather now.'' to ``Today wa.'' As shown in FIG. 4(b), when the operator moves the first cursor 24 to the beginning of the phrase "nowa" to be changed, the device recognizes this as a request to reconvert this phrase. Accordingly, as shown in FIG. 4 (Q), the candidate matrix 21 is placed two lines below and to the right of the converted sentence "It's nice weather now" (the position shown in the figure), and its "reading" is ”25 is read and displayed in the display area 26.

Note that the above-mentioned "reading" 25 always displays a fixed number (for example, 16 characters) from the position of the notation corresponding to the first cursor 24. Further, candidates are displayed in the candidate matrix 21 with variable length.

At this time, the second cursor 27 is also displayed in the reading display area 26, and editing within this area 26 becomes possible. If the display area 26 exceeds the above 16 characters due to editing work, the end of the sentence within the area is returned to the buffer. Furthermore, if there is a shortage, it is supplemented from the buffer so that the predetermined 16 characters are always displayed. This yields variable-length optimal candidates.

Here, "Ima wa" is changed to "Today wa", so the above reading display area 26 displays "Ima hai Itenki desu.", and the part after "Desu." is blank (No. 4
(See figure (C)).

In the reading display area 26, the second cursor 27
When "Imaha" is corrected and converted to "Kiyoha" using normal editing techniques such as deletion, insertion, and replacement, variable length conversion candidates are displayed in the candidate matrix 21 (the fourth (See figure (d)). The notation is corrected by selecting the one desired by the operator from among these (see FIG. 4(e)).

This makes it easy to change word divisions, which previously required considerable effort. In particular, reading and conversion candidates are displayed near the correction location, improving operator visibility and operation. This has the effect of making it easier.

FIG. 5 shows a detailed flowchart of the automatic reconversion (step 15).

When the operator selects the desired variable length candidate string from the displayed variable length candidate string (step 14), the reading length of the currently displayed word and the reading length of the selected candidate are compared (step 15a), and based on the result, 3. Branch into the street. That is, if the reading length of the currently displayed word is shorter than the reading length of the selected candidate, the current word and the next (adjacent to the right) word are connected as one word (step tsb) and the process returns to step 15a. .

In addition, if the reading length of the currently displayed word is longer than the reading length of the selected candidate, after replacing the currently displayed word with the word selected above (step 15c), The remaining pronunciation is set as the "pronunciation of the selected candidate" (step 15d), and this is set as the current word to be converted (step 15e). Then, create a reading sequence of up to 16 characters to the right from the current conversion target position (step 1).
5f), regarding this, the above-mentioned Kana-Kanji conversion module 9
(step 15g).

The candidate string output above is sorted in order of "reading length" and "learning value" (step tsh), and the first candidate is automatically selected from this candidate string (step 15j). The same determination process as in step 15a is performed on the selected candidate (step 15j), and if the reading length of the currently displayed word is longer than the reading length of the selected candidate, the process returns to step 15c and the word currently displayed is returned to step 15c. If the reading length of the word is shorter than the reading length of the selected candidate.

The current word and the next (adjacent to the right) word are connected as one word (step 15k) and the process returns to step 15j.

Further, in step 15a and step 15j, if the reading length of the currently displayed word is equal to the reading length of the selected candidate, the displayed word is replaced with the selected word (step 15Q); It is displayed as a reconversion result (step 16).

FIG. 1 is an operational flowchart of display control in the reading display area 26 in the above embodiment. FIG. 1 shows that in the modified reconversion operation (step 32),
By moving the first cursor to the phrase or word that you want to correct (step 33), the pronunciation and candidates corresponding to the phrase are displayed (step 34), and at this time, control is transferred to the reading display area ( Step 35) Edit within the region featuring points (Step 36)
However, at this time, the pronunciation is controlled so that it matches the predetermined number of characters (steps 37 to 39), and candidates for the corrected pronunciation are displayed (step 41). The operation after the operator selects from among the above candidates is as explained in detail with reference to FIG.

According to the above embodiment, when the operator is shown the delimitation made by the device and it is different from the intended one, he can simply select a candidate and the conversion result corresponding to the new delimitation will be displayed. This improves the operability of the installation. Therefore, a key for changing the delimiter is not required on the device side, and the structure becomes simpler.

Effects As described above, according to the present invention, in a Japanese language processing device having a kana-kanji conversion means, there is a means for instructing correction and re-conversion, and a means for creating conversion candidates with variable length readings in addition to homophones. variable length candidate creation means is provided, and when a reconversion instruction is given by the modified reconversion instruction means, the variable length candidate creation means displays a variable length candidate in an area adjacent to the line where the modified portion is located; It has an editing function for the readings of a predetermined number of characters, and the remaining character strings are converted according to the candidate selected by the operator from among the variable length candidates, so the delimiter automatically corrected by the device can be corrected. This method has the remarkable effect of realizing a Japanese language processing method that can easily perform the following steps.

[Brief explanation of the drawing]

Fig. 1 is a flowchart showing details of the operation in an embodiment of the present invention, Fig. 2 is a hardware configuration diagram of a Japanese language processing device which is an embodiment, and Fig. 3 is a flowchart showing an overview of the operation of the embodiment. , FIGS. 4(a) to 4(e) are diagrams showing specific examples of operations, and FIG. 5 is a flowchart showing details of automatic reconversion. 1: Keyboard, 2: Keyboard controller, 3: Key person cover buffer, 4: Key person cover sofa controller,
5: Candidate buffer, 6: Candidate buffer controller, 7
: CRT controller, 8: CRT, 9; Kana-Kanji conversion module, to: cpU, 11 to 16: Processing steps, 21.22: Variable length candidate string, 23: Notation, 24: First cursor, 25: Reading. 26: Reading display area, 27: Second cursor, 31-4
6: Processing step. Figure 3 Figure 4 Figure 4

Claims (1)

[Claims] (1) In a Japanese language processing device having a kana-kanji conversion means, a means for instructing correction re-conversion and a variable length candidate creation means for creating variable length candidates of pronunciation in addition to homophones as conversion candidates are provided, When a reconversion instruction is given by the correction reconversion instruction means, the variable length candidate creation means displays a variable length candidate in the vicinity of the line where the correction is located, and also has an editing function for reading a predetermined number of characters. 1. A Japanese language processing method, characterized in that the remaining character strings are converted according to a candidate selected by an operator from among the variable length candidates.