JPH02144661A - Kana/kanji converter - Google Patents

Abstract

PURPOSE:To reduce the storage capacity of a dictionary by registering only a 'KANJI' (Chinese character) notation having the most 'OKURIGANA' [declensional KANA (Japanese syllabary) ending] in the case of words expressed by synonyms having different 'OKURIGANA' and deleting an unnecessary 'OKURIGANA' by an internal flag. CONSTITUTION:In the case of words 'TORIKESHI' (erase) and 'TORI-KESHI' (erase) e.g. which are synonyms and have different 'OKURIGANA', 'KANJI' notation having the much 'OKURIGANA', e.g. only the KANJI 'TORIKESHI' is registered and unnecessary 'OKURIGANA' is deleted by internal flags 6, 6'' formed in a dictionary body part to compress the storage capacity of the dictionary. In the case of a word having different 'KANJI' notations due to the differences of 'OKURIGANA' in synonyms, a 2nd and subsequent 'KANJI' notations in the dictionary body part are canceled and a flag for 'OKURIGANA' data consisting of one byte is provided in stead of the canceled contents. Thereby, storage capacity for storing the 2nd and subsequent 'KANJI' notation can be reduced and the storage contents can be simply converted into the 'KANJI' notation of 'OKURIGANA' other than the one registered in a word dictionary.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a kana-kanji conversion device such as a Japanese word processor that converts and outputs kanji or kanji-kana mixed sentences.

<Prior Art> Generally, a kana-kanji conversion device is equipped with a dictionary device that converts an input kana character string into a kanji or a kanji-kana mixed sentence.

For this reason, as a dictionary device 2, we memorize all the pronunciations of words and the corresponding kanji or Fi kanji kana admixtures (independent words and conjugated forms, endings, etc. are added), and store such words. It is equipped with a complete word dictionary that permanently stores tens of thousands of relatively frequently used words.

That is, the main body of a conventional word dictionary has a structure as shown in Figure 4, and even if they are homophones and have the same part of speech, words with different numbers of characters etc. are registered separately.

For example, in the case of the kana ``torikeshi'' (pronounced as ``f'' shown in Figure 5), the okurigana is ``cancellation'', ``cancellation'', ``.
There are three kanji notations for ``cancel'', and different part-of-speech numbers are assigned to them.

<Problem to be solved by the invention> However, even though they are homophones, if they are assigned different part of speech numbers and sounds just by changing the Japanese letter, for example, as shown in Figure 5, the kanji notation for the first month is In the case of ``,'', 1 byte is used for ``duplicate data'' and ``word length data'', and the reading kana ``torikeshi'' is 4 characters, and ``one byte is required for the data of one hiragana character.'' The above reading uses 4 bytes, 2 bytes are used for the part-of-speech number data, and 1 kanji is used.
It requires 2 by I to write the characters, so “tori υr?”
Niji uses 6 bytes for 2 kanji and 2 kana characters, for a total of 13 bytes. Generally, the storage capacity of the dictionary device for ``reading kana data'' and ``kanji notation data'' is fully compressed, so the data capacity is not fixed and can be changed. In the case of the second kanji notation “cancellation”,
The reading is the same as the first "Cancellation 1", so it is omitted. However, 3 bytes are used for "duplicate data", "word length data", and "part of speech zero data", and "kanji notation data" is used.
requires 5 bytes, so a total of 8 bytes are required. The third kanji notation "cancel" requires 4 bytes for the ``kanji notation data'', so the above 3 commonly required
Including the part-time job, 7 bytes are required. Therefore, in the entire word dictionary, the number of words written in kanji and the number of words written in okurikana differ depending on the word, so specific numbers cannot be displayed. However, [number of okurikana notations × ′kanji notation data′]
A storage capacity of ゛〈number of relevant words〉 is required. In the dictionary of the above-mentioned single imperial edict, the representative senryoku was written in kango (kango), where only the kanji notation of ``na'' was clearly written, and in the correction of okuri-gana, ``-character erasure and ``--'' were used.
It requires complicated key operations such as ``insert character'' and ``no conversion,'' which is time-consuming.

The present invention was made in order to solve these problems, and it abolishes the kanji notation of the second and subsequent okigana in the main body of the dictionary, and instead uses each 4' part of speech number data 2.
A flag indicating "untransferable data" is set in the next byte. Alternatively, do not set a separate O' sending data flag for the second and subsequent ones, and use one flag for the second and subsequent sending data in the next 1 byte of the first sending multi-gana "Kanji notation data". By setting up the dictionary O
The overall purpose is to provide a kana-kanji conversion device that compresses storage capacity and simplifies the operation of correcting okurigana.

<Means for Solving the Problems> In order to achieve the above object, the present invention provides a kana-kanji conversion device that converts an input kana character string into all kanji or kanji-kanji characters, which converts words into reading data and word length data. In addition to organizing the dictionary by ``Kana'' and ``Kanji'', it also has a part-of-speech table 2 that stores part-of-speech numbers and part-of-speech information in association with each other. For words, do not register all the kanji notation of okurigana in the dictionary main body, only register the kanji notation with many incense okurigana, and delete unnecessary okurigana using the internal flag set in the dictionary main body. in a way that
It is characterized by compressing the storage capacity of the dictionary.

<Operation> According to the above structure, in the case of a word that is a homophone synonym and has several kanji representations depending on the okurigana, the second and subsequent kanji representations in the dictionary body are abolished, and 1 is replaced instead. Since a flag is set for byte's sending data 1, the storage capacity for storing the second and subsequent kanji notation is small, and it is easy to convert to the kanji notation of sending many gana other than those registered in the word dictionary. can do.

<Examples> The present invention will be described below based on drawings showing examples thereof.

Fig. 1 is a block diagram showing the circuit configuration of the Japanese word processor of this embodiment, and Fig. 2 (al) is a block diagram showing the circuit configuration of the Japanese word processor of this embodiment.
Figure 2 shows a specific example of the dictionary main body of the
a2) is a pattern diagram of the sending multi-gana correction flag of the present invention, FIG. 2(b) is a diagram showing a specific example of the dictionary main body of another embodiment (hereinafter referred to as B method), and FIG. 3(a) is a flowchart of method A, FIG. 3(b) is a flowchart of method B, and FIG. 4 is a basic configuration diagram of the dictionary main body. In FIG. 1, 1 indicates character and word information and various instructions. An input device for inputting information, such as a keyboard.

Consists of tablet devices, optical character readers, magnetic tape devices, etc. Reference numeral 2 denotes a storage device for storing character information input from the input device 1, and includes, for example, a core memory, an IC memory, a magnetic disk, and the like. Reference numeral 3 denotes an output device for outputting the information stored and edited in the storage device 2, and includes, for example, a printer, a display device, a magnetic tape, a magnetic disk device, etc. Reference numeral 4 includes a dictionary device and the like for supplying valid information in a timely manner in response to inquiries about spelling information of character and word data stored in the storage device 2. Reference numeral 5 denotes a control device which completely controls the exchange of signals between the devices of the above configuration, and is composed of, for example, a computer. The basic configuration of the main body of the word dictionary, which is the main part of the dictionary device 4, is as follows.
As shown in the figure, it is the same as before. 71d' Duplicate data contains data on how much the reading is the same as the previous data. 8 is 'word length data 2' and 'word length data.

Next to “-Yomi kana data 1 part of speech number data.

Contains data on how many bytes there are in ``Kanji notation data.'' 9 is ``Yomi kana data,'' which includes a reading or l that is not duplicated with the previous data. 10 is ``Part of speech @sign data.''
It contains the frequency of this kanji notation and the part-of-speech number that indicates what the part of speech (noun, verb, etc.) corresponds to. 11 is ``Kanji notation data'' which contains the notation data of this kanji. In the case of registering a word with homophone synonyms and multiple usages of okurigana (in the case of registering a word with multiple usages of okurigana), register only the one kanji notation with the most common kanji notation for both A and B types in the above word dictionary.

In method A, as shown in (al) in Figure 2, 2
The 2nd kanji notation "cancellation 2'" and the 3rd kanji notation "cancellation" have been abolished, and instead a flag has been set in 1 byte of the part corresponding to ``kanji notation data'', and a control that has not been used so far. (a2) in Figure 2 shows an example of the above flag pattern.The first bit bl is ON.
If so, it is determined that it is the conventional kanji notation data, and when it is OFF, it is a flag for correction of okurigana. b2-b of the 4 bits in the first half
Since 4 is already used as 7 as another control code, the latter 4 bits are used. For example, the 6th 7.8th bit of flag 6 in the second transmission data ■ in Fig. 2 (al) is oeo, that is, if it is the same pattern as 42 in Figure 2 (a2), the second to last stroke will be removed and it will be "cancelled", and similarly the third stroke in Figure 2 (al) will be the data of l.
The 6th 7.8th bit of flag 6' is O-ken, that is, the second
Figure (a 2) (If the pattern is the same as D IFx 3, remove the first and second transfer keys from the end and "cancel"
becomes.

Next, to explain method B, in method B, the flag 6' for ``transferring data'' is not set up for each different kanji notation of okurigana, but is registered as a representative, Provide next.

In this case, there is only one flag 6', so if b s to b s are harmonized with O-ken, as in Ii3 in Fig. 2 (a2), it will automatically be "Cancel" - "Cancel". “(b
g~b8 are O・O) → “Cancel” (b6~b8 are O・・・)
In this way, the amount of money (transfer) remains constant.

According to the flowchart shown in FIG.
Enter the kana character string gold from the input device. Next step S
2 indicates kanji conversion. Proceeding to step S3, the entire dictionary device is searched, and the first notation candidate is extracted and displayed. Step S
If there is no conversion instruction in step 4, the first notation candidate kanji notation is displayed and the process ends. If there is a conversion instruction, step S
When the program goes to Step 5 to check the next control code of the part-of-speech number data', if the first bit is fist (ON), the program goes to Step S6 and extracts all the next expressions as before, and returns to Step S. If the first bit is 0 (OFF) in step S5, it is a correction flag, so the process proceeds to step S7, and by key operation, the second and subsequent feeds are corrected and displayed in step S8, and step S is performed.

Return to

Next, the operation of adjusting the feed angle of method B will be explained according to the flowchart 1 shown in FIG. 3(b).
, / are the same as 81 to S5 of the above A method, and will therefore be omitted. When looking at the next data 1 of "Kanji notation data" in step 85, if the first bit is ・(ON), the process advances to step S6 and the next notation is extracted as before, and step S4
Return to

If the first pin 1- is 0 (OFF) in Ss', it is a correction flag, so proceed to step S7, follow the correction flag 6', and proceed to step 88', where the reading is "take [7"].
In the case of ``Cancel 1~'' → ``Cancel'' → ``Cancel'', all the kanji notations with modified kanji characters are displayed in order by operating one specific key, and the process returns to step S, /. In addition, if you wish to output and display which of the above selections is given priority, the flag shown in FIG. 2 (a2) is also placed in the learning buffer being used. For example, by providing a control code such as 000 for "sales", 0 for "receivables", etc., output can be made with one conversion key from the second time onwards.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the circuit configuration of the Japanese word processor of this embodiment, Fig. 2 (al) is a diagram showing a specific example of the dictionary main body of this embodiment (method A), and Fig. 2 (a2) FIG. 2(b) is a pattern diagram of the feed correction flag of the present invention.
3(a) is a diagram showing a concrete example of the dictionary human body part of another embodiment (B method), FIG. 3(a) is a noro chart diagram of A method, and FIG.
b) A flowchart of the ViB method, FIG. 4 is a basic configuration diagram of a dictionary main body, and FIG. 5 is a diagram showing a specific example of a conventional dictionary main body. 1... Input device 2... Storage device 3... Output device 4... Dictionary device 5... Control device 6
, 6', 6'... Since they are flagged, 4 bytes are used in method A for one word written in kanji with a difference in sending υ, and less than 1 byte is used in method B. The number of storage bags required is small, and the feed gun can be adjusted with a simple key operation. Agent: Patent Attorney Takeshi Sugiyama (and 1 other person) A-way minri flow 佽-L゛Bz弐゛guaσ-Chii Fig. 1 (a) (b) f; 3

Claims (1)

[Claims] 1. In the kana-kanji conversion device that converts the input kana character string into kanji or kanji-kana mixed sentences, words are organized by kana reading data and word length data, and the kana reading and kanji notation are stored in correspondence. It is equipped with a dictionary main body and a part-of-speech table that stores part-of-speech numbers and part-of-speech information in correspondence, and for words that are homophones and synonyms but differ only in okigana, the kanji notation of all okigana is registered in the dictionary main unit. Kana-kanji conversion is characterized in that the memory capacity of the dictionary is compressed by registering only the kanji notation with the most okigana, and deleting unnecessary okigana using an internal flag set in the dictionary main body. Device.