JPH02139662A - Kana/kanji converting device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention] (a) Industrial application field This invention is mainly used in document processing devices such as Japanese word processors, and converts input kana character strings into kanji, kana, mixed sentences, and hiragana. This article relates to a kana-kanji conversion device that converts into mixed sentences.

(b) Conventional technology In the kana-kanji conversion process of the Japanese word processor mentioned above, conventionally, when a kana character string is input from the keyboard to instruct kana-kanji conversion, the input kana character string is stored in a dictionary. Words written in kanji or kanji and kana are sequentially converted into the words stored in the kanji or kanji/kana mixture.The user continues to give instructions for conversion, and the word written in the desired notation is selected by the user. is selected from the keyboard.

In addition, in the one-sided conversion process, as a first method,
There is a method in which each input kana character is converted to a katakana character by corresponding I to l, and a second method is to convert the input kana character string into a katakana character and a hira character. The proper distribution of the katakana part and the hiragana part is determined by dividing the hiragana part into a katakana part and examining the hiragana part using an attached word dictionary (Japanese Patent Application Laid-Open No. 120-198).
(Refer to Publication No. 276), and as a third method, the input kana character string is converted into a single character string in the same way as kanji using a dictionary (Japanese Patent Laid-Open No. 160469/1983).
(see both Japanese Patent Application Laid-open No. 63-49965).

(C) Problems to be Solved by the Invention However, in the above-mentioned kana-kanji conversion process, ■ conversion instructions must be given many times until the desired word is displayed, and the number of conversion instructions becomes unnecessary. many.

■Even if a word similar to the desired word is displayed, if you instruct further conversion, it may end up being converted into a completely different word.

■There are problems such as the fact that it is not clear until the end that the desired word is not in the dictionary.

In addition, in the first method of katakana conversion processing, the input kana character string is simply converted into a katana character string without analyzing its constituent elements, so attached words (particles, auxiliary verbs, etc.) are also katakana. There is a problem with converting to kana characters.

For example, if we take the expression "kujiraga" as an example, if you input "kujiraga" (one-sided input, or
(batch conversion to katakana after inputting hiragana), it is converted to ``kujiraga'' and cannot be converted to the original ``kujiraga''. In order to correct this, you need to correct each letter one by one, or change ``Kujiraga'' to ``Kujira J'' and ``GA''.
One possible method is to input the information separately, but it would take too much time and effort.

Furthermore, in the second method, there is a problem in that it takes time to determine the appropriate distribution of Katakana and Hiragana parts using the ancillary word dictionary.

In the third method, taking "Kujira ga" as an example, if both "Kujira" and "Kujira" are registered in the dictionary, "Kujira ga" can be easily changed by the conversion instruction.
However, in this case, the dictionary memory capacity required for the - word increases, and there is a problem that many words cannot be registered within the limited capacity.

To give another example, if the character string ``Shinkankakuha'' is input, the last character ``Ha is J
There is also the problem of not knowing to what extent it should be converted into katakana, as it is not clear whether it is an adjunct or a suffix. When converting to kanji, when ``ha'' is an adjunct, it is ``shinsensakuha'', and when it is a suffix, it is ``shinsensakuha'', and in the case of katakana conversion, it is ``shinsensakuha'', respectively. It must be converted to ``Shinkankafwa'' or ``Shinkankafha,'' but it is not possible to determine such adjuncts and suffixes.

This invention was made in consideration of such circumstances, and in the kana-kanji conversion process, a word is converted into words with multiple homophone synonyms for one reading, thereby reducing the number of words in the selection process. To provide a kana-kanji conversion device that converts only words into katakana characters by giving layers and taking into consideration the constituent elements of the word and the adjunct following it in the katakana conversion process. be.

(d) Means for Solving the Problems FIG. 1 is a block diagram showing the configuration of the invention according to claim 1 of the present application. As shown in the figure, the invention according to claim 1 includes:
A storage means 102 for storing a kana character string inputted from the input means 101, and a storage means 102 for storing a kana character string input from the input means 101, and a kana character string that can be analyzed and converted into a word as an adjunct word consisting of a kana character string, a particle, an auxiliary verb, etc. Kana-Kanji conversion means 10 that divides Kana character strings into Kana character strings and converts Kana character strings that can be converted into words into Kanji or words written with Kanji and Kana mixed.
3, a homophone synonym conversion means 104 that converts the kana character string that can be converted into a word into a plurality of types of homophone synonym notation words that have the same pronunciation and the same meaning as the word converted by the kana-kanji conversion means 103; This is a kana-kanji conversion device comprising a display means 105 for displaying the contents inputted from the input means 101 and the contents converted by the kana-kanji conversion means 103 and the homophone-synonym conversion means 104.

Further, FIG. 2 is a block diagram showing the configuration of the invention according to claim 2 of the present application, and as shown in the figure, the invention according to claim 2 stores the kana character string input from the input means 101. The kana character strings stored in the storage means +02 and the storage means 102 are analyzed and divided into kana character strings that can be converted into words and kana character strings as attached words consisting of particles, auxiliary verbs, etc., and can be converted into words. A kana-kanji conversion means +03 that converts a kana character string into a word written in kanji or a combination of kanji and kana; and a kana character string that converts a kana character string that can be converted into a word into a kana character string that corresponds to each kana character in l to l ratio. Kana conversion means +06 and input means! Display means 10 for displaying the content input from 01 and the content converted by the kana-kanji conversion means 103 and the katakana conversion means 106
This is a kana-kanji conversion device equipped with 7.

In addition, in this invention, a word with homophone synonym notation is also called "output fluctuation" because there are multiple notations for one word, and means a word with the same reading and meaning but different notations. .

In other words, as an example is shown in Figure 4, for the word tsu, there are seven hiragana characters, normal kanji notation, okurigana different from the normal kanji notation, appropriate characters, variant characters, etc. means each conversion.

Further, the kana-kanji conversion means 103 in the present invention is a conventionally known ordinary kana-kanji conversion device, and includes a dictionary for kana-kanji conversion.

Further, the input means 101 in the present invention may be any device that can input a kana character string, such as a keyboard device, a tablet device, an OCR, or an external storage device such as a magnetic tape or floppy disk.

The display means 105 and 107 may be any device that can display the input kana character string, the converted kanji-kana mixed sentence, or the katakana-hiragana mixed sentence, for example,
A CRT display device, an LC (liquid crystal) display device, an EL display device, or the like is used.

Further, the homophone and synonym conversion means 104 includes a dictionary for conversion, but this dictionary may be shared with the dictionary included in the kana-kanji conversion means 103.

In addition, a kana-kanji conversion method! 03, the homophone synonym conversion means 104 and the unilateral conversion means 106 are generally C
It is convenient to use a microcomputer consisting of PU, ROM, RAM, and I10 board, and the storage means 102
As such, RAM is usually used therein.

(E) Effect According to the invention of claim 1 of the present application shown in FIG. 1, the input means! The kana character string input from 01 and stored in the storage means 102 is divided by the kana-kanji conversion means +03 into a kana character string that can be converted into a word and a kana character string as an attached word. The character string is converted into a word written in Kanji or a mixture of Kanji and Kana.

Furthermore, a kana character string that can be converted into a word is converted into words expressed by a plurality of types of homophone and synonyms by the homophone and synonym conversion means 104, and the contents thereof are displayed on the display means 105.

Therefore, a kana character string that can be converted into a word is hiragana 7
Katakana/normal kanji notation/words with different kanji from normal kanji notation/aji/variant characters are converted into words with the same reading and the same meaning, so create a hierarchy in the selection of words. , and it is possible to efficiently select words with the desired notation.

Furthermore, according to the invention of claim 2 of the present application shown in FIG.
The kana character string input from the input means +01 and stored in the storage means +02 is divided by the kana-kanji conversion means 103 into a kana character string that can be converted into a word and a kana character string as an attached word, and can be converted as a word. Nakakana character strings are converted into words written in Kanji or a combination of Kanji and Kana.

Furthermore, a kana character string that can be converted into a word is converted by a katakana conversion means 106 into a katakana character string that corresponds to each kana character on a ratio of 1 to 1, and the contents thereof are displayed on a display means 107.

Therefore, it is possible to save much more dictionary memory space than registering single words in a dictionary, and since information on adjunct words has been analyzed, it is easier to use than just conventional single word conversion. It is possible to easily obtain a simple notation that is close to the wishes of the user.

(f) Examples Hereinafter, the present invention will be described in detail based on examples shown in the drawings. Note that this invention is not limited to this.

FIG. 3 is a block diagram showing the configuration of an embodiment of the present invention.

In this figure,! is a keyboard as an input means. The keyboard l has various keys for document input such as kana keys and alphabet keys, a conversion key for instructing kana-kanji conversion, and sub-conversion keys for homophone synonym conversion and temporary kana conversion, which will be described later. , and inputs information on kana character strings and various instruction information to the control unit 2 (conversion keys and subconversion keys are not shown).

The control unit 2 is composed of a microcomputer consisting of a CPU, ROM, RAM, and I10 board, and performs data processing according to a control program written in the ROM.

The following description assumes a virtual circuit block having this data processing function.

Reference numeral 3 denotes storage means consisting of a RAM for storing kana character strings input from the keyboard l.

4 is a kana-kanji conversion means and is equipped with a word dictionary for kana-kanji conversion.

The kana-kanji conversion means 4 first analyzes the kana character string stored in the storage means 3 when the conversion key on the keyboard l is pressed down. This analysis is performed based on a commonly used kana-kanji conversion algorithm, using linguistic information such as word dictionaries and grammar, and probabilistic information such as word frequency. Through this analysis, the kana character string stored in the storage means 3 is converted into a kana character string that can be converted into a word consisting of a prefix, a word stem, a suffix, etc., and a kana character string as an adjunct consisting of a particle, an auxiliary verb, etc. Divide into.

Then, a kana character string that can be converted into a word is converted into a word written in kanji or a combination of kanji and kana.

5 is a homophone synonym conversion means, and when the subconversion key is pressed after the conversion key on the keyboard l, the kana character string that has been divided by the kana-kanji conversion means 4 and can be converted into words is converted into kana-kanji. The word converted by means 4 is converted into a plurality of types of words with the same pronunciation and the same meaning as the converted word.

Here, words with homophone and synonym notation are also called "output fluctuations" because there are multiple notations for one word, and are words that have the same pronunciation and meaning but different notations.

What is shown in Figure 4 is an explanatory diagram showing an example of the content of homophone synonym conversion. The kanji notation is different from the kanji written in the kanji.
It is treated as a group of words, and conversion is performed in units of these groups.

6 is a one-sided conversion means, and after a kana character string is input from the keyboard l, if the conversion key is not pressed down but the direct conversion key is pressed down, the word that is analyzed by the kana-kanji conversion means 4 is Converts a kana character string that can be converted into a katana character string that corresponds one-to-one to each kana character.

If the kana character string stored in the storage means 3 has already been converted into a kanji or a word containing kanji and kana by the kana-kanji conversion means 4, this conversion can be performed by pressing down the sub-conversion key 7. A word is first converted into a hiragana character string, and then converted into a hiragana character string that corresponds one-to-one to the hiragana character string.

Reference numeral 7 denotes a display section consisting of a CRT display or LC (liquid crystal) display, which displays the contents entered from the keyboard 1 and stored in the storage means 3, the kana-kanji conversion means 4, the homophone synonym conversion means 5, and the katakana conversion. The contents converted by means 6 are displayed.

5 and 6 are diagrams for explaining the analysis process in unilateral conversion by the unilateral conversion means 6. FIG.

In the conversion by the above-mentioned kana-kanji conversion means 4,
In addition to the word to be selected, that is, the notation character string candidate, the constituent elements are taken into consideration when performing kana-kanji conversion. For example, assume that the character string ``Shinkankakuha'' is input from the keyboard 1 and stored in the storage means 3, and is converted by the kana-kanji conversion means 4 as a candidate for the kanji-mixed phrase ``Shinkankakuha''. In this case, information as shown in FIG. 5 is obtained.

This means that if the conversion candidate to kanji is the notation ``Shinkan wa'', the number of written data (characters) is 1 character for ``Shin'' as the prefix and 2 characters for ``sensation'' as the stem. , there is no suffix, and the adjunct is a single character "wa", and the reading is "shinkankakuha", so the number of reading data (characters) is "shinkankakuha" as a prefix. '', the root word is ``kankaku'' (4 letters), there is no suffix, and the attached word is ``wa'' (one letter).

After getting such information, keyboard! When the sub-conversion key is pressed down q times, the notation character string converted to "Shinkan wa" is once converted to the input character string "Shinkankakuha" expressed in hiragana, and the fifth The information shown in the figure is retained as is.

Then, when the sub-conversion key on keyboard I is pressed down one more time, for example, among the information shown in Figure 5, if only the stem of the word is set in advance to be changed to katakana. To do this, remove the stem part of the notation data, that is, the 2nd and 3rd from the beginning of the notation string, and extract the stem part of the reading data, that is, the 3rd to 6th from the beginning of the reading data.
The notation is done by changing the number to Katakana. As a result, "New Kankaku wa" is obtained.

Furthermore, if the prefix and suffix are set in advance to be katakana, only the attached word is left and converted to katakana, resulting in ``shinkankafu wa''.

Further, for example, when the notation ``Shinsensakuha'' becomes a candidate through conversion by the kana-kanji conversion means 4, information as shown in FIG. 6 is obtained.

This means that if the candidate for conversion to kanji is the notation ``Shinkan-ha'', the number of written data (characters) is 1 character for ``shin'' as the prefix and 2 characters for ``sensation'' as the stem. , indicates that the suffix is one character of "ha" and there is no attached word, and the reading is "Shinkankakuha", so the number of reading data (characters) is the same as that of "Shin" as a prefix. It has 2 characters, 4 characters for the root word ``gankaku'', 1 character for the suffix ``ha'', and no adjuncts.

If the sub-conversion key is pressed after manually inputting such information, all written character strings are converted to hiragana, which is the input string, as described above, and then the sub-conversion key is pressed again. If it is pressed down, as described above,
Obtain "New Kankakuha" or "Shinkankafuha".

The operation in such a configuration will be explained below based on the flowchart shown in FIG.

First, in a state where a kana character string is input from the keyboard 1 and stored in the storage means 3 (step 2
01), when the conversion key on keyboard 1 is pressed down (
Step 202), in the kana-kanji conversion means 4, the number of presses of the conversion key is counted by the counter memory N (step 203), and a known learning table, which is probabilistic frequency information, is searched (step 204).
, the presence or absence of learning candidates is determined (step 205).

Then, if there are learning candidates, the learning candidates are output (
Step 206), if there are no learning candidates, a dictionary search is performed (Step 207), and the display order candidates stored in the counter memory N are output (Step 2).
08), return to step 202.

Also, when the secondary conversion key is pressed down (step 209
), the homophone synonym conversion means 5 and the katakana conversion means 6 store the number of presses of the sub-conversion key in a counter memory N,
(step 21O), and if the count is 1 (step 211), the input character string is converted into hiragana (step 212).
), the hiragana character string candidate is output to the display section 7 (step 213), and if the count number is 2 (step 2
+4), one-sided conversion is performed (step 215), one-sided candidates are output to the display unit 7 (step 216), and the process returns to step 202.

Next, it is determined whether the count number in the counter memory N1, which is the number of times the conversion key has been pressed, is 0 (step 217), and if it is 0, that is, if the conversion key has not been pressed even a month. , the count of the counter memory N is set to several months so that the number of times the sub-conversion key is pressed is regarded as one time (step 21B).
), Step 21! Return to

If the count number of counter memory Nt is not 0, that is, if the conversion key is pressed several times and kana-kanji conversion is performed, and if the count number of counter memory N is 3. (Step 2
19), the okurikana conversion process is performed (step 220).
), the display section 7 displays the forward-gana conversion candidates as shown in Fig. 4.
(step 221), counter memory N,
If the count number is 4 (step 222), the matching character conversion process is performed (step 223), and
The correct character conversion candidates as shown in the figure are output to the display unit 7 (step 224), and if the count number in the counter memory N is 5 (step 225), variant character conversion processing is performed ( Step 226), variant character candidates as shown in FIG. 4 are output (Step 22).
7), respectively return to step 202.

Then, when the count number of counter memory N reaches 6 (when the sub-conversion key is pressed down 6 times),
The count number of the counter memory N is set to l (step 22B), and the process returns to step 211, where the hiragana candidates are outputted to the display unit 7 again.

Then, when the pressing of the sub-conversion key is finished and a key for confirming a conversion candidate, such as a key for inputting the next character string, is pressed (step 2
The display candidate data displayed on the 29L display section 7 is registered in the learning table (step 230), and the display candidate data is stored in a text memory (not shown) (step 231).

Further, in step 229, if no key operation is performed to confirm the conversion candidate, the process returns to step 202, and the depressing operation of each key is determined again.

In this way, the kana string that can be converted into a word is
As shown in Figure 4, 7 pieces of hiragana / normal kanji notation / different okuri-gana from normal kanji notation / guess characters /
As variant characters, they are sequentially converted into words with the same reading and meaning.

Therefore, when a word similar to the desired word is displayed using the conversion key, the homophone synonyms of that word can be sequentially displayed using the sub-conversion key, and the word can be selected hierarchically. You can select words efficiently.

In addition, in katakana conversion, the hiragana character string to be converted into a word is converted into a katakana character string that corresponds l to 1 to each year-gana character string, and the adjunct words remain in hiragana. Since the words are retained, the capacity of the dictionary memory can be saved compared to registering single words in a dictionary, and it is possible to easily obtain a simple notation that is close to the user's wishes.

(g) Effects of the Invention According to this invention, in kana-kanji conversion, by performing homophone synonym conversion and adding hierarchy to word selection, it becomes possible to efficiently select desired words. In addition, by grouping words, it is possible to create ri created from hiragana, or words with different okurigana or variant characters from normal basic kanji, so each word can be dictionary independently. This makes it possible to save the capacity of the dictionary memory.

In addition, in katakana conversion, information on words and adjuncts is analyzed, so it is easier to obtain katakana notation that is closer to the user's wishes than simply converting to a katakana character string. .

[Brief explanation of the drawing]

Figure 1 is a block diagram showing the configuration of the invention according to claim 1 of the present application, Figure 2 is a block diagram showing the configuration of the invention according to claim 2 of the present application, and Figure 3 is the configuration of an embodiment of the invention. FIG. 4 is an explanatory diagram showing an example of the contents of homophone and synonym conversion, FIGS. 5 and 6 are diagrams explaining the analysis process of unilateral conversion, and FIG. FIG. 6... One-sided conversion means, 7... Display section. l...keyboard, 2...control unit, 3...storage means, 4...kana-kanji conversion means, 5...homophone synonym conversion means, 2': 2nd: go

Claims (1)

[Scope of Claims] 1. A storage means for storing a kana character string input from an input means, and a kana character string that can be analyzed and converted into a word by analyzing the kana character string stored in the storage means, particles, auxiliary verbs, etc. a kana character string as an attached word, and a kana character string that can be converted into a word, and a kana character string that can be converted into a word, and a kana character string that can be converted into a word, and a kana character string that can be converted into a word, and a kana character string that can be converted as a word, and a kana character string that can be converted as a word, and a kana character string that can be converted as a word. Homophone synonym conversion means for converting into a plurality of types of homophone synonym notation words having the same reading and the same meaning as the word converted by the means, contents input from the input means, kana-kanji conversion means, and homophone synonym conversion means A kana-kanji conversion device comprising display means for displaying the converted content. 2. A storage means for storing the kana character string input from the input means, and an auxiliary word consisting of the kana character string, particles, auxiliary verbs, etc. that can be analyzed and converted into words by analyzing the kana character string stored in the storage means. a kana-kanji conversion means that divides a kana character string that can be converted into a word into a word written in kanji or a combination of kanji and kana; and a kana character string that can be converted as a word for each kana character. A kana-kanji conversion device comprising: a katakana conversion means for converting into a katakana character string corresponding to the katakana character string, and a display means for displaying the content input from the input means and the content converted by the kana-kanji conversion means and the katakana conversion means. .