JPH08194701A - Character processor - Google Patents

Abstract

(57) [Summary] [Purpose] Performs word conversion that is not affected by frequency information. [Structure] User dictionary (YKBU) that allows users to register
Two types of word dictionaries are prepared: F) and an ordinary word dictionary that performs word searches based on frequency information. KB when converting kanji
If the user dictionary (YDBUF) is specified from 5,
The user dictionary is referenced to obtain the first conversion candidate, and the normal word dictionary is used to obtain the second conversion candidate of the homophone.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character processing apparatus for inputting characters while performing kana-kanji conversion to create and edit documents such as Japanese documents.

[0002]

2. Description of the Related Art In a character processing device for inputting Japanese sentences, it is common to input characters using kana-kanji conversion. Conventionally, in addition to frequency information stored in advance in a dictionary stored in association with word reading and notation and frequency information (learning) obtained from the input result of the device operator,
There was no way to determine the order in which words appeared. Therefore, it is not possible to convert the desired word as the first candidate only when the device operator desires, and as a result, the undesired word is converted into the first candidate.

[0003]

[Problems to be Solved by the Invention] The frequency information stored in advance in a dictionary in which word readings are associated with notations and the frequency information (learning) obtained from the input result of the apparatus operator alone yields 100%. It cannot be said that the conversion result that the device operator desires can be obtained. For example, you want to alternately use the word "fit" that is stored in the dictionary with the highest frequency in advance and the word "encounter" that is stored with the lowest frequency in the dictionary that stores the readings of words in association with notations. And At that time, it is assumed that the word "fit" is selected first. When the learning function is working, if you select "fit" and then type "aou", the word "fit" will be converted as the first candidate again. Therefore, the "encounter" desired by the device operator is not converted. Also, even when the learning function is not working, if you select "fit" and enter "Au" again (because the conversion depends on the frequency information stored in the dictionary in advance at this time), enter "Au" again. The word “match” with the highest frequency information is converted, and a word different from the word desired by the device operator is also converted.

The present invention has been made in view of the above-mentioned conventional example, and cannot be converted as a first candidate either from a dictionary in which the reading of a word provided in advance is stored in association with a notation or from learning. An object of the present invention is to provide a document processing capable of converting a word desired by a device operator as a first candidate when the device operator likes and a device thereof.

[0005]

In order to achieve such an object, the invention of claim 1 provides an input means for inputting a reading of a word and a plurality of kinds of words stored in association with the notation of the reading of the word. A plurality of types of dictionaries, including a word dictionary for performing a word search based on frequency-of-use information and a user dictionary in which words can be registered, an instruction means for instructing a word dictionary to be referred to, and a reading input by the input means. Is provided with conversion means for converting into a word notation by referring to the instructed word dictionary, and output means for outputting the conversion result converted by the conversion means.

Further, in the invention of claim 2, the conversion means refers to the word dictionary designated by the designating means, outputs the conversion result as the first conversion candidate, and outputs it as the second and subsequent conversion candidates. It is characterized in that a conversion result for reading a word is obtained by referring to another word dictionary.

[0007]

According to the invention of claim 1, regardless of the frequency of use, the user only has to register the words he / she wants to use in the user dictionary. By specifying the user dictionary by the instruction means as necessary, the frequency of use can be changed. Word conversion that does not depend on it is possible.

According to the second aspect of the present invention, since the designated dictionary is used only for obtaining the first conversion candidate, various conversion candidates can be obtained, and the user's instruction operation of the dictionary can be simplified.

[0009]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 shows the system configuration of a character processing apparatus to which the present invention is applied.

In FIG. 1, reference numeral 1 is a microprocessor (CPU) that controls the entire apparatus and executes calculations and logical judgments for document processing, and 2 is a control program executed by the CPU 1, a document processing program, and frequency information. The ROM 3 for storing the word dictionary for performing the word search is used as a temporary storage of various data sent from the device components and a work area when the CPU executes various programs.
A 16-bit word RAM. The RAM 3 is provided with a buffer area described below. That is, a text buffer (TBUF) which is an area for storing document data to be created and edited, a DBIND which is an area (area) for storing an index of a homophone buffer area,
Homophone buffer area (DBBUY), keyboard (K
B) A keyboard buffer (KBBUF) for temporarily storing the reading sequence input from 5 and a user dictionary activation buffer (YKBUF) for storing information on whether to activate the user dictionary are provided. Further, the user dictionary buffer (YDBUF) to be accessed when the information to start is stored in the user dictionary startup buffer (YKBUF).
In the BUF), "reading" (20 characters), which is the reading string of the word that the user wants to convert as the first candidate, "notation" (20 characters) that is the notation of the "reading", and its " The data of "grammar information" (10 characters) indicated by the grammar information of "notation" is stored. The reading and the notation input via the KB 5 are registered in the user dictionary as words by the CPU 1.

Reference numeral 4 is a text buffer (TBUF)
External storage device (DIS, such as a hard disk or a floppy disk) for storing the created document of
K) and 5 are keyboards (KB) equipped with alphabetic keys, hiragana keys, katakana keys, and other character / symbol input keys, conversion keys for instructing start / stop of various functions of the apparatus, and various function keys. Reference numeral 6 is a cursor register (CR), and 7 is a display buffer memory (DBUF) for storing a dot pattern of data to be displayed. 8 CR the information stored in CR6 and DBUF7
A CRT controller (CRTC) that performs control for displaying at T9, and 9 is a display device (CRT) that uses a cathode ray tube or the like, and a CRT that displays a dot pattern or cursor controlled by the CRTC 8 on the screen. 10 is CR
It is a character generator (CG) that stores patterns of characters and symbols displayed at T9. And 11-13
Are an address bus (AB), a control bus (CB), and a data bus (D) that connect the above-mentioned components, respectively.
B). The AB 13 transfers an address signal designating a component to be controlled by the CPU 1. CB11 is C
A control signal is transferred to each component that is the control target of PU1. The DB 12 transfers data between the respective constituent elements. For example, keyboard (KB) 5 to CPU 1
A signal corresponding to the input character is transferred to the DB 12 via.

The word dictionary stored in the ROM 2 stores information on reading, its notation, and its grammar in association with each other, and is referred to by kana-kanji conversion or the like. Figure 2
FIG. 6 is a diagram showing the structure of the word dictionary. As shown in FIG. 2, each of the words stored in the word dictionary includes “yomi” (20 characters), which is a reading sequence of words, and the “yomi”.
The fixed data length of "notation" (20 characters), "grammar" (10 characters) indicating the grammatical information of the "notation", and "frequency information" (10 characters) of the "notation". It has three pieces of information.

FIG. 3 shows a keyboard buffer (KBBUF)
Shows the configuration of. The buffer length of KBBUF is 1024 bytes. The reading string information input from the KB 5 is accumulated in this buffer until the conditions for executing the kana-kanji conversion are satisfied. Then, when the conversion conditions are satisfied, the reading string is converted into Kanji and the contents of the buffer are cleared. In the case of the present embodiment, the input character string is stored in 2 bytes per character using the JIS X 0208 code.

FIG. 4 shows a user dictionary activation buffer (YKB).
UF) configuration is shown. This buffer stores information on whether to activate the user dictionary. If the value stored in that buffer is "0", the user dictionary will not start,
"1" indicates that the user dictionary is activated.

FIG. 5 shows a dictionary created by the device operator (YDBU).
The structure of F) is shown. As shown in FIG. 5, each of the words stored in the word dictionary is a reading string of the word, "reading" (20 characters), and a "notation" (20 characters) that is the notation of the reading. ) Has three pieces of information of a fixed data length of “grammar” (10 characters) indicating the grammatical information of the “notation”.

FIG. 6 shows the structure of a text buffer (TBUF) for storing the contents of document data being created and edited.
Reference numeral (a) in FIG. 6 shows the overall structure of (TBUF), and reference numeral (b) in FIG. 6 shows the buffer structure for each character.

As shown in the overall structure (a) of FIG. 6, a document is stored as a list of characters with a structure of one character and two bytes.

In the character buffer structure (b) of FIG. 6, one character is composed of 2 bytes. MSB (2 ¹⁵ bits)
Is the homophonic accent level, and the rest (2 ¹⁴ ~
2 ⁰ bit) is a code part. The homophone accent level indicates whether the character is a homophone or a normal character, and when it is "0", it is a normal character, and when it is "1", it is a homophone. For normal characters, the character code is J
It is stored in the IS X 0208 code. For a homophone, the homophone number is stored in the chord part. The homophone number becomes an index of the index part of the homophone buffer area described later, and specifies a specific homophone buffer.

The character processing device having the above configuration is KB5.
It operates according to various input instructions from. That is, when an input instruction is given to the KB5, an interrupt signal is sent to the CPU1, and the CPU1 responds to the RO signal in response to this.
Various control / processing programs stored in M2 are read out and executed to perform various controls and processing. For example, KB
According to the instruction from 5, the document stored in DISK4 is read into the text buffer (TBUF), and further KB is read.
In response to an instruction from 5, the pattern of the TBUF data is expanded into DBUF and the content of the document data is displayed.

Next, a kana-kanji conversion process for an input character string using the character processing device having the above configuration will be described with reference to a screen display example of the CRT 9 shown in FIG. In the case of the present embodiment, kana-kanji conversion is executed when the key "." Is input. Further, in the key input in the process described below, the key input related to the function of reading the document from the DISK 4 and the document editing will not be handled.

In FIG. 7, CM represents a cursor indicating the position where a character is displayed when the next key input is performed, and TS represents a text screen on which the contents of the document stored in the text buffer (TBUF) are displayed. In the figure, MS represents a monitor line that displays the input key data one by one.
When the kana-kanji conversion described below is executed, the reading line is once displayed on the monitor line MS, and after the kana-kanji conversion,
The conversion result is displayed in the text buffer TS.

With "match" already converted,
The Kanji conversion process of the character processing device will be explained by taking up the case where the device operator newly inputs the same character "Au" from KB5 and converts it into Kana-Kanji. In this embodiment, the following three types of Kanji conversion processing are possible.

(1) When conversion is performed depending on the word dictionary information stored in the ROM (FIG. 7 (a)) In the word dictionary of the ROM 2, the most frequent notation for "Au" is "fit". Therefore, even if “match” is already selected in (1) and then “match” is input again, “match” becomes the first candidate.

(2) When conversion is performed depending on the conversion result (learning) of the device operator (FIG. 7 (b)) The notation "Au" is read from the conversion result (learning) of the device operator. If you have already learned that “fit”, then even if you input “read” next time, “fit” will be the first candidate.

(3) When the notation desired by the device operator is converted as the first candidate by activating the user dictionary (FIG. 7 (c)) When it is already converted to "fit" or under any environment Even if there is, when the user dictionary is activated, the notation "encounter" for the reading "Au" stored in the user dictionary is converted.

The processing procedure of the CPU 1 for executing the above processing is shown in FIGS.

FIG. 8 shows a schematic processing procedure of a part which takes in a key input and performs a kanji conversion process.

First, in step S81, the data input from the keyboard (KB) 5 is loaded into the keyboard buffer (KBBUF). In the following step S82,
Check the type of input key stored in KBBUF. Here, if the user dictionary activation key is input, the processing procedure proceeds to S83, where the user dictionary activation buffer (YDBU
Information for activating the user dictionary (YDBUF) is stored in F), and based on the reading character string stored in the keyboard buffer (KBBUF) in S84, in S85, kana-kanji conversion A (code (c) in FIG. 7) Kanji conversion process) is executed. In the subsequent step S88, the result of the kana-kanji conversion is displayed on the CRT 9.

If the input key is a character input key, that is, if the user dictionary activation key is not input,
The process proceeds to step S87 based on the read character string stored in the keyboard buffer (KBBUF) in S86,
Kana-Kanji conversion B (Kanji conversion process of symbols (a) and (b) in FIG. 7) is executed. In the subsequent step S88, the result of the kana-kanji conversion is displayed on the CRT 9.

If the input key is neither the user dictionary activation key nor the character input key, the process proceeds to step S89.
Then, the processing corresponding to the key input (for example, movement of the cursor, insertion / deletion of character string, etc.) is performed, and the result is displayed on the CRT 9 in the processing of the following step S88.

FIG. 9 is a detailed flowchart of the process of "Kana-Kanji conversion A" in step S85 of FIG.

First, in step S83, when the user dictionary activation information is stored in the user dictionary activation buffer (YKBUF), that is, the user dictionary for instructing the user dictionary from KB (input means and instruction means of the present invention) 5 is entered. When the activation information is input, the reference target dictionary is set to the user dictionary (YDBUF). Next, based on the "reading" of the input and extracted phrase, a conversion candidate is searched from the user dictionary (YDBUF) of S91, which of these conversion candidates is most likely, and conversion is performed in S92. Create a candidate. Then, the maximum likelihood word is determined as the first candidate. In step S93, the reference dictionary is changed to the word dictionary (ROM2), the word dictionary (ROM2) stored in the ROM is searched based on the determined first candidate, and the homophone is determined in step S94. create. Looking at the created homophone word buffer, finally, in step S95, the conversion result is output to the CRT 9. Therefore, the CPU 1 that executes this processing procedure functions as the conversion unit and the output unit of the present invention.

FIG. 10 is a detailed flowchart of the process of step S87 "Kana-Kanji conversion B" of FIG.

First, when the input key is the character input key in step S82, the process is step S1.
Proceed to 01, and the word dictionary (ROM
Search for conversion candidates from 2). Then, step S10
In 6, the most plausible word is determined as the first candidate based on the frequency information. In step S87,
A homophone is created based on the determined first candidate. The created homophone word buffer is viewed and the conversion result is output in step S104. When the learning mode is set, the first candidate is the learning word (step S10).
, S102) and then the word dictionary (ROM2) is referred to to create a homophone (steps S103 → S10).
4).

In addition to this embodiment, the following example can be carried out.

1) When there are many homonym candidates even though the number of words used by the device operator is limited, only when the user dictionary activation key is activated, only the words registered in the user dictionary It is also possible not to convert.

2) In the above description, the case where the dictionary has a fixed length has been described, but the dictionary can also have a variable length. "Yomi" which is a reading string of words, "notation" which is the notation of the "Yomi", "grammar information" which shows the grammar dictionary of the "notation", and "frequency" which the "notation" has It is a dictionary that has four pieces of information "frequency information",
The dictionary is made variable in length by inserting a code indicating the end of each data at the end of each data.

[0039]

As described above, according to the first aspect of the present invention, the user may register the word he / she wants to use in the user dictionary regardless of the frequency of use, and the user dictionary may be used by the instructing means as necessary. By specifying, the word conversion can be performed regardless of the frequency of use.

According to the second aspect of the present invention, by using the designated dictionary only for obtaining the first conversion candidate, various conversion candidates can be obtained, and the user's operation of instructing the dictionary can be simplified.

[Brief description of drawings]

FIG. 1 is a block diagram showing a system configuration of an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a configuration of a word dictionary (ROM2) according to the embodiment of the present invention.

FIG. 3 is an explanatory diagram showing a configuration of a keyboard buffer according to the embodiment of the present invention.

FIG. 4 is an explanatory diagram showing a configuration of a user dictionary activation buffer according to the embodiment of the present invention.

FIG. 5 is an explanatory diagram showing a configuration of a user dictionary buffer according to the embodiment of the present invention.

FIG. 6 is an explanatory diagram showing the configuration of a text buffer (TBUF) that stores the contents of document data being created and edited in the embodiment of the present invention.

FIG. 7 is a diagram showing a screen display example of a CRT 9 according to the present invention.

FIG. 8 is a flowchart showing an operation of the character processing device in the embodiment of the present invention.

FIG. 9 is a flowchart showing the operation of the character processing device in the embodiment of the present invention.

FIG. 10 is a flowchart showing an operation of the character processing device in the embodiment of the present invention.

[Explanation of symbols]

 1 CPU 2 ROM 3 RAM 4 External storage device (DISK) 5 Keyboard (KB) 6 Cursor register (CR) 7 DBUF 8 CRTC 9 CRT 10 Character generator (CG) 11 Control bus (CBus) 12 Data bus (DBus) 13 Address Bus (ABus)

Claims (2)

[Claims] 1. An input means for inputting a word reading, and a plural-type word dictionary in which the word reading is stored in association with a notation, the word dictionary performing a word search based on frequency-of-use information, and a user capable of registering a word. A plurality of types of dictionaries including a dictionary, an instruction means for instructing a reference word dictionary, and a conversion means for converting the reading input by the input means into a word notation by referring to the instructed word dictionary. And a character output device for outputting the conversion result converted by the conversion unit. 2. The conversion unit refers to the word dictionary designated by the designation unit, outputs the conversion result as the first conversion candidate, and refers to another word dictionary as the second and subsequent conversion candidates. The character processing device according to claim 1, wherein a conversion result for reading the word is obtained.