JP2542384B2 - Character processor - Google Patents

Description

TECHNICAL FIELD The present invention relates to a character processing device having a learning function for converting a character string and determining a first candidate for notation for a homonym.

[Prior Art] Conventionally, as a character processing device for converting a character string, a character processing that performs so-called kana-kanji conversion, in which the operator converts the reading into kanji by inputting the reading and giving an instruction to convert the kanji The device is realized.

Recently, in the character string conversion in the above device, a so-called learning process of preferentially deciding and displaying the first candidate according to the most recently selected word, even for conversion candidates of a plurality of notations, for example, homonyms Widely used by character processors. Therefore, a character processing device that performs such learning processing is provided with a memory that stores, as learning data, information related to Chinese characters selected by a predetermined selection criterion.

In this learning process, in addition, the kanji that are used most frequently are displayed in order, or the kanji related to the technical term for each technical field is set as the first candidate among the homonyms.

[Problems to be Solved by the Invention] However, the contents of this memory cannot be grasped by the operator as to what is the first candidate for homonyms, and if the number of homonyms is increased, the memory capacity is exceeded. There was a problem that the learning data of could not be stored.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a character processing device which can effectively use a limited capacity of memory and can easily select and input homonyms.

[Means for Solving Problems] In order to solve such problems, the present invention, when selecting a homonym in Kana-Kanji conversion, a display unit for displaying candidate Kanji for selection and a display unit. First and second selecting means for selecting a desired kanji from the candidate kanji displayed on the screen, and storage means for storing the kanji selected by the first and second selecting means. Storage means for distinguishing and storing the kanji selected by the selection means and the kanji selected by the second selection means, and the kanji stored in the storage means is prioritized over the display means when the next homophone is selected. Of the kanji stored in the storage means according to the instruction of the instruction means for deleting the kanji selected by the first selecting means among the kanji stored in the storage means First Characterized in that and means for deleting the Chinese character selected by the-option means.

[Operation] According to the present invention, the kanji selected from the homonyms by the first selecting means or the second selecting means are separately stored in the storage means according to the selection, and the next homophonic meanings are stored. When selecting a word, the kanji stored in the storage means is displayed on the display means by the control means. Since the deletion of the kanji selected by the first selecting means among the kanji stored in the storing means is executed by the deleting means in accordance with the instruction of the instructing means, it is unnecessary for the selection of homonyms. Learning data can be deleted.

[Examples] Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1 shows an example of the overall configuration of the present invention. In FIG. 1, a CPU is a microprocessor, which performs arithmetic operations for character processing, logical judgments, etc., and each component connected to those buses via an address bus AB, a control bus CB, and a data bus DB. To control.

The address bus AB transfers an address signal indicating a component to be controlled by the microprocessor CPU.
The control bus CB transfers and applies the control signals of the respective components to be controlled by the microprocessor CPU. The data bus DB transfers data between the constituent devices.

Second, ROM is a read-only fixed memory. PA is a program area in which control procedures by the microprocessor CPU, which will be described later with reference to FIGS. 7 to 12, are stored.

RAM1 is a writable random access memory having a 16-bit word structure, and is used for temporary storage of various data from each constituent element.

TBUF is a document buffer, which is a memory area for storing character information converted into Chinese characters. The DIC is a memory area for storing a reading and a dictionary in which kanji corresponding to the reading are tabulated in order to perform kana-kanji conversion.

YBUF is a memory area for storing the kana reading code input by the kana input key KANA in the keyboard KB. LRND is learning data, and is a memory area that stores a word code that means a learned word, that is, a word in which a homonym is selected.

HSTRY is a memory area for storing the learning data registration history list, and stores the storage index of the word registered in the memory area LRND. WCOD is a memory area for storing the same copy code as the word code registered in the memory area LRND.

i is a memory area for storing counter variables used by each program stored in the program area PA. IM
EMO is memory area LRN when registering words in memory area LRND
This is a memory area for storing the storage index of each word in D.

KB is a keyboard, and various functions for instructing various functions for the character processing device such as alphabet keys, hiragana keys, katakana keys and other character symbol input keys, and conversion keys, learning keys, document editing end keys, etc. It is equipped with a key.

In the figure, CON is a conversion key for instructing the selection of homonyms, that is, conversion of kanji, KA as the first selection means.
NA is a character / symbol input key for inputting the reading, LRN is a learning key for instructing learning and kanji conversion as the second selection means, and EDEND is a document edit end key for declaring that the input edit process is terminated.

DISK is a memory such as a floppy disk for storing a fixed form document, which stores the created document, and the stored document is called by the microprocessor CPU when necessary by the instruction of the keyboard. CR is a cursor register. The CPU can read and write the contents of the cursor register. The CRT controller CRTC, which will be described later, displays a cursor at a position on the display device CRT for the address stored here.

DBUF is a display buffer memory, which stores patterns such as document information stored in the document buffer TBUF.

The CRTC plays a role of displaying the contents stored in the cursor register CR and the buffer DBUF on the display CRT.

The CRT is a display device using a cathode ray tube or the like, and the display of the dot configuration pattern and the cursor on the display device CRT is controlled by the CRT controller. Furthermore, C
G is a character generator, which stores a pattern of characters and symbols to be displayed on the display device CRT.

In the character processing device composed of each of these components,
It operates in response to various inputs from the keyboard KB. When the input from the keyboard KB is supplied, first, an interrupt (interrupt) signal is sent to the microprocessor CPU, and that microprocessor CPU has a fixed memory. Various control signals stored in the ROM are read out, and various controls are performed in accordance with those control signals.

FIG. 2 shows the structure of the dictionary DIC in the character processing device of the present invention. In FIG. 2, YF is a reading unit, which stores up to 10 characters for reading a word in 2 bytes per character, and stores 0 in the remaining area. KF is a Kanji part, and stores up to 4 characters in a word notation with 2 bytes per character, and 0 is stored in the surplus area.

The GF is a grammatical information section, which is used for learning and stores grammatical information such as the part of speech of the word. LF is the learning possibility determination data, and 1 is stored for words that can be learned for homonyms, and 0 is stored for non-learnable hiragana character strings or words for kanji that have no homonyms.

Each word consists of 31 bytes, and a word code that identifies the word is assigned from the beginning of the dictionary. For example, the variable W (i) means the i-th word from the beginning of the dictionary.

FIG. 3 shows the structure of the document data buffer TBUF. Third
In the figure, document data is represented by 1 character and 2 bytes. 0 is stored as a non-input code.

FIG. 4 shows the structure of the input reading code buffer YBUF.
In FIG. 4, the kana code output when the kana input key KANA on the keyboard KB is pressed is stored in 2 bytes per character.

 FIG. 5 shows the structure of the learning data memory area LRND.

In FIG. 5, the word code of the learned word is stored in the memory area LRND in the form of a list, and the memory area LRND is composed of 2 bytes for 1 word and 4 words for 200 words.
It consists of 00 bytes. In the initial state and the remaining memory area, 0 is stored as a code meaning a word that does not exist, and -1 (FFFF) is stored as a list end code at the end of the list.

FIG. 6 shows the structure of the learning data registration history list area HSTRY.

In FIG. 6, the memory area HSTRY is the memory area LRND.
When registering a word code to, the registered memory area LRND
This is a list for storing the index of, and is composed of 2 bytes for one word and 200 bytes for 100 words. In the initial state and the remaining area, 0 is stored as a nonexistent LRND index value, and -1 (FFFF) is stored as a list end code at the end of the list.

FIG. 7 is a flow chart showing the operation of the character processing device of the present invention. This control procedure is mainly microprocessor CPU
Is executed by

In step S7-1, a key is pressed from the keyboard KB, and the microprocessor CPU waits for an interrupt to occur.

When a key is input, the microprocessor CPU discriminates the input key in step S7-2, and branches to any of steps S7-3 to S7-7 depending on the type of the key.

Step S7-3 is a process when the Kana input key KANA is pressed, and the input Kana code is stored in the memory area YB.
Store in UF.

Step S7-4 is a process when the conversion key CON is pressed, and converts the reading string stored in the memory area YBUF into a Chinese character string selected from the homonyms (detailed in FIG. 8). Statement).

Step S7-5 is a process when the learning key LRN is pressed, and the word code stored in the memory area WCOD is stored in the learning data area LRND (detailed in FIG. 11).

Step S7-6 is a process when the document edit end key EDEND is pressed, and it is considered that the document input is completed, and a process for invalidating the learning data automatically learned by the conversion process is performed (FIG. 12). In detail).

Step S7-7 is processing when a normal key (for example, a cursor movement key) other than the Kana key KANA, the conversion key CON, the learning key LRN, and the document edit end key is pressed,
This is a process that is generally performed in a character processing device of the same type, and is well known, so it will not be described in particular.

Step S7-8 is a display process for displaying the changed part as a result of the above-mentioned editing process. The data in the document stored in the document buffer TBUF is read into a pattern by reading one character by the microprocessor CPU, and is displayed in the display buffer DB.
This is a widely used processing that is usually output to UF.

FIG. 8 is a detailed flowchart of the learning data registration process by pressing the conversion key in step S7-4.

In step S8-1, microprocessor CPU
Searches the dictionary DIC according to the stored contents of the input reading sequence YBUF.

In step S8-2, the microprocessor CPU determines whether or not the reading matching the stored contents of YBUF is found as a result of the dictionary search, and if found, the process branches to S8-3, and if not found, the process branches to S8-13. Then, step S8-
In 3, the word code of the found word is stored in the memory area WCOD.

If no word is found, the contents of the buffer YBUF are left unchanged in the document buffer TBUF in step S8-13.
To step S7-4, and the process returns to step S7.

In step S8-4, the microprocessor CPU searches whether or not the same word as the word stored in the memory area WCOD exists in the memory area LRND. Step S8−
In step 5, it is determined whether the same word code as the memory area WCOD is found in the memory area LRND. If it is found, the process branches to step S8-11. If not found, the process branches to step S8-6.

In step S8-6, the memory DIC is searched for the next Kanji word for the same reading in the buffer YBUF. In step S8-7, it is determined whether or not the next word for the same reading as the buffer YBUF is found. If found, the process branches to step S8-3, and if not, the process branches to step S8-8.

In step S8-8, the microprocessor CPU determines whether or not the word stored in the memory area WCOD can be learned by referring to the learnability determination data (LF in FIG. 2) of the word. If it is determined that learning is possible, the process branches to step S8-9, and if it is determined that learning is not possible, the process branches to step S8-12.

The word determined to be learnable is registered in the memory area LRND in step S8-9, and the history data area HSTRY stores the word index in the memory area LRND in step S8-10.

In step S8-12, the notation code of the found word WCOD is set in the document buffer TBUF, and then in step S7.
Return to -8.

In step S8-11, the notation code of the word code found in the memory area LRND is set in the document buffer TBUF, and the process returns to step S7-8.

FIG. 9 is a detailed flowchart of the learning data registration processing in steps S8-9 and S11-12.

In step S9-1, the counter variable i is initialized to 0.
Set.

In step S9-2, the value represented by the variable LRND (i) using the variable i as an index is the list end code FF.
Whether or not it is FF is judged, and if the list end is detected, the program immediately returns to the program of FIG. 8; otherwise, the processing of step S9-3 and thereafter is executed.

In step S9-3, the microcomputer CPU determines whether or not the value represented by the variable LRND (i) is the unset value 0. If the variable LRND (i) has an unset value of 0, the process branches to step S9-5, and otherwise to S9-4.

In step S9-4, the counter variable i is incremented by 1, and the process loops to step S9-2.

In step S9-5, the value indicated by the variable LRND (i) is replaced with the value in the memory area WCOD.

In step S9-6, the value of the counter variable i is stored in the memory area IMEMO, and then the process returns.

FIG. 10 is a detailed flowchart of the history data registration process in step S8-10.

In step S10-1, the initial value 0 is set to the counter variable i.

In step S10-2, the value represented by the variable HST (i) with the variable i as the index is the list end code F.
Whether or not it is FFF is judged, and if the list end is detected, the process immediately returns, but if not, the learning possibility judgment processing of step S10-3 and thereafter is executed.

In step S10-3, it is determined whether the value represented by the variable HST (i) is the unset value 0, and if the variable HST (i)
Is a non-set value 0, the process branches to step S10-5, otherwise to step S10-4.

In step S10-4, the counter variable i is incremented by 1, and the process loops to step S10-2.

In step S10-5, the value indicated by the variable HST (i) is replaced with the memory area IMEMO value, and then the process returns.

FIG. 11 is a detailed flowchart of the learning data registration process by pressing the learning key in step S7-5.

In step S11-1, an initial value 0 is set to the counter variable i.

In step S11-2, the value represented by the variable LRND (i) with the variable i as the index is the list end code.
Whether or not it is FFFF is determined, and if the list end is detected, the process proceeds to step S11-12, but if not, the processes of step S11-3 and thereafter are performed.

In step S11-3, it is determined whether or not the value represented by the variable LRND (i) matches the WCOD value. If they match, the process branches to step S11-5, otherwise to step S11-4. .

In step S11-4, the counter variable i is incremented by 1, and the process loops to step S11-2.

In step S11-5, the value indicated by the variable i is stored in the memory area IMEMO.

In step S11-6, the learning data registration list HST
An initial value 0 is set to the counter value i to search RY.

In step S11-7, the value represented by the variable HST (i) with the variable i as the index is the list end code FFF.
It is determined whether or not it is F, and if the list end is detected, go to step S11-12, otherwise, go to step S11-8.
The following processing is performed.

In step S11-8, it is determined whether the value represented by the variable HST (i) matches the value of the memory area IMEMO,
If they match, the process branches to step S11-10, and if not, the process branches to step S11-9.

In step S11-9, the counter variable i is incremented by 1, and the process loops to step S11-7.

In step S11-10, the variable HST (i) is set to 0 in order to make it unused, and the learning registration history HSTRY is allocated in step S11-11.

In step S11-12, the word code shown in the memory area WCOD is set in the memory area LRND (the same as FIG. 9).

FIG. 12 is a detailed flowchart of the learning invalidation process of step S7-6 which functions as a deleting means. The code signal generated by the document edit end key EDEND is detected by the microprocessor CPU as a deletion instruction signal. Then, in step S12-1, an initial value 0 is set to the counter variable i.

In step S12-2, the value represented by the variable HST (i) with the variable i as the index is the list end code F.
It is determined whether or not it is FFF, and if the list end is detected, go to step S12-6, and if not, go to step S12-.
The following processing is performed.

In step S12-3, the value of the memory area LRND represented by using the value of the variable HST (i) as an index: LRND (HST
Replace (i)) with the unset value 0.

In step S12-4, the unset value 0 is set to the variable HST (i).

In step S12-5, the counter variable i is incremented by 1, and the process loops to step S12-2.

In step S12-6, the data in the memory area LRND is searched, and if the unset value 0 exists between the LRND data (values from 1 to FFFF), that portion is packed and allocated (rewritten), To return.

In the description of the present embodiment, the dictionary structure is assumed to be a dictionary with a fixed word length, but the same processing is performed even with a more compressed dictionary structure composed of variable word lengths. be able to.

Although the word code, which is a serial number from the beginning of the dictionary, is used as the data used as the learning data, the notation code itself of the word may be used.

Furthermore, as the timing to initialize the learning data,
I set a single key as a key to instruct the end of document editing, but in addition to this key, a key that means the end of document editing,
Similar functions may be assigned to a plurality of keys such as a print instruction key and a key for instructing an external storage device to record a created document.

Next, the operation of this embodiment will be described by taking a specific key operation as an example.

When the operator inputs a kanji reading on the keyboard KB, the reading entered on the display device CRT is displayed in the order of steps S7-1 → S7-2 → S7-3 → S7-8 in FIG.

Next, when the operator presses the conversion key on the keyboard KB, the reading input in the order of steps S7-1 → S7-2 → S7-4 (detailed procedure shown in FIG. 8) → S7-8 is Kanji. Is converted into Kanji and the converted Kanji characters are displayed on the display device CRT.

At this time, when a homonym is selected, the operator uses the next candidate key (not shown) to sequentially display desired Chinese characters on the display device CRT. When a desired kanji is found and the operator presses the conversion key CON for kanji conversion, the control procedure shown in FIG. 9 is executed in the order of steps S7-1 → S7-2 → S7-5 and selected. Kanji is automatically registered in memory LRND as learning data for homonyms.

However, this automatically registered learning data is automatically deleted after the document is edited. When it is desired to retain the learning data, the Kanji conversion is instructed by pressing a learning key described later.

Next, when the operator inputs a reading for a homonym registered in the memory LRND, step S7-1
→ S7-2 → S7-4 to the control procedure of Fig. 8
Kanji stored in LRND is automatically selected and displayed
Displayed on CRT.

Therefore, the operator does not need to select the homonyms for the Chinese characters registered from the second time by once selecting the homonyms.

The registered learning data is stored in the memory LRND, and at the same time, the memory HST stores the attribute information indicating that the learning data is stored in the memory LRND. The stored contents of the memory HST are used to erase the learning data registered by the operator himself.

When the learning key LRN is pressed during the conversion of homonyms into kanji, the learning data is stored in the learning data LRND after the processing to leave the learning data in the memory LRND in the control procedure shown in FIG. 11. After inputting, the reading of the input is converted to the selected kanji.

The reading input by such a processing procedure is converted into kanji and the document creation is completed. When the operator presses the key EDNED for instructing the end of document creation, the control procedure of FIG. 12 is executed, and only the learning data designating deletion by the attribute information stored in the memory HST is deleted from the memory LRND.

As described above, according to the present invention, if the Chinese character selected once from the homonyms is selected, the homonyms selection process is not required from the second time, and the first character that is simply displayed is displayed.
It suffices to indicate conversion of candidate kanji. To convert the selected Kanji into another Kanji, use the Next Candidate key to display another Candidate Kanji, and then press the conversion key CON. The selected Kanji will be the first candidate for the next time. It becomes kanji. Further, after the document processing, the selected kanji (learning) data that does not need to be left is automatically erased, so that the capacity of the memory for storing the learning data can be made smaller than in the conventional case.

Furthermore, when the learning data invalidation processing (deletion processing) is performed, the number of learning data stored in the memory LRND is counted, and the counting result is displayed on the display device to display the memory.
It is also possible to prevent the overflow of LRND, the operator can know the number in the learning data stored in the memory, and the learning data can be organized.

Furthermore, in the present embodiment, an example in which the learning data is erased after the completion of the document creation is shown, but in the document, the new learning data is stored by providing a memory different from the memory LRND and storing the learning data The learning data to be registered may be extracted from the memory and re-registered in the memory LRND.

[Effects of the Invention] As is clear from the above description, according to the present invention, once a homonym is selected during document creation, Kanji conversion should be performed first only by inputting the reading of this word. Since the words are automatically displayed, the operator does not need to select homonyms, which facilitates key operation and also leaves the selected kanji as learning data after the document is created. , And it becomes possible to effectively use the memory for storing the learning data. Further, since the learning data to be erased is automatically erased at the end of the document, it is possible to reduce the useless labor of the operator for erasing the data each time.

[Brief description of drawings]

FIG. 1 is a block diagram showing an example of the configuration of the present invention, FIG. 2 is an explanatory diagram showing an example of the memory configuration of a dictionary DIC of the present invention, and FIG. 3 is an example of the memory configuration of a document data buffer TBUF of the present invention. FIG. 4 is an explanatory diagram showing an example of the memory configuration of the input read buffer YBUF of the present invention, FIG. 5 is an explanatory diagram showing an example of the memory configuration of the learning data memory LRND of the present invention, and FIG. FIG. 7 is an explanatory view showing an example of the memory configuration of the learning data history list of the present invention, and FIGS. 7 to 12 are flow charts showing an example of the control procedure of the microprocessor CPU in the embodiment of the present invention. CPU: Microprocessor, RAM: Random access memory, CRT: Display device.

Claims (2)

(57) [Claims] 1. When selecting a homonym in kana-kanji conversion, a display means for displaying candidate kanji for selection and a first kanji for selecting a desired kanji from the candidate kanji displayed on the display means. And second selection means, and storage means for storing the kanji selected by the first and second selection means, the kanji selected by the first selection means and selected by the second selection means Storage means for distinguishing and storing the stored kanji, a control means for preferentially displaying the kanji stored in the storage means on the display means at the next selection of homonyms, and the storage means stored in the storage means. Instructing means for instructing deletion of the kanji selected by the first selecting means among the kanji, and selecting by the first selecting means among the kanji stored in the storage means in response to the instruction of the instructing means And a means for deleting the written kanji. 2. The deleting means deletes the kanji selected by the first selecting means from the kanji stored in the storing means when the end of the document creation is instructed. The character processing device according to claim 1, which is characterized by the above.