JPS5840650A - Japanese input system - Google Patents

Abstract

PURPOSE:To obtain a smooth procedure for finding a necessary word among homonyms, and to improve an input speed by hitting a different reading key after inputting the reading of KANJI (Chines character) or a phrase, additionally inputting different reading, and finding a word coincident with the initially inputted reading. CONSTITUTION:The output of a keyboard 1 for KANA (Japanese syllabary) or ROMAJI (Roman character) input is transferred to a processor 4. The output signal of the keyboard 1, generated by hitting a character key 9, is processed in the processor 4 differently according to which of input mode specifying keys 11-14 is pressed, and then displayed on the display screen 8 of a display device 2. In an ROM3 and an RAM5, data on a processing procedure by the processor 4 are stored mainly. A storage device 6 is stored with a KANJI and a phrase dictionary.

Description

[Detailed Description of the Invention] The present invention relates to a Japanese writing method in which the reading of a Japanese sentence is inputted using a kana or low-character keyboard, and a desired kanji is obtained from the kana character series corresponding to the reading of the kanji part. .

The so-called kana-kanji conversion method, which involves inputting a kana character sequence corresponding to the reading of a kanji and converting it into kanji by some means, can be roughly divided into two types. The first method is called automatic kana-kanji conversion, and as a general rule, when using the kana character series, it does not distinguish between the hiragana or katakana part and the kanji part, but simply inputs each clause separately. The input device analyzes the kana character series, separates the kanji part from the flat base or katakana part, and determines the part of speech of the kanji by grammatically analyzing the kana part that follows the kanji part. , and automatically selects the most likely kanji having that reading among the parts of speech. In this method, since kana-kanji conversion is performed from reading information divided into bunsetsu units, the kanji section must convert kana to kanji in idiom units. Therefore, an idiom dictionary with at least 15,000 words and a grammar analysis program are required. For this reason, the size of dictionaries and programs generally becomes extremely large, and a large capacity memory, such as a magnetic disk, is required. As an input method for a small, dedicated Japanese typewriter, rather than a Japanese information processing system using a large computer, it is necessary to use a standard floppy disk or mini-floppy disk as an external file, and this method has not been realized. It comes with difficulties.

The second method is called the paired Kana-Kanji method, in which when the user inputs the reading of an idiom that is a kanji or a series of kanji as a kana character series, the machine reads the Y'i kanji dictionary or idiom dictionary. In this method, a plurality of kanji or phrases with the corresponding pronunciation are displayed simultaneously or sequentially on the display screen, and the user looks at the display and selects and inputs the required kanji or phrase.

With this method, the user is responsible for the process of selecting the necessary kanji from candidate kanji by simply matching the input kana character sequence with the readings in a kanji or compound word dictionary, so there is no need for grammatical analysis. , the size of the dictionary and program can be smaller than the automatic kana-kanji conversion method. Now, assuming that we are preparing an idiom dictionary with 15,000 candies, the size of the dictionary itself is approximately 120KB, assuming 8 bytes per word.
The entire control program, including reading input, display selection, etc., is about 20KB, so even if you include editing and printing programs, it will fit on a double-sided double-density 2-nifloppy disk (
(300KB) can be stored sufficiently.

However, in this method, when inputting the reading, if there are many words with the same reading, these homonyms are displayed simultaneously or sequentially on the display screen, and the necessary word is selected from among them. Since the user has to make a selection, the user must repeat the passive procedure of looking at the displayed candidate words and selecting them each time, which hinders smooth input.
Gives the user a sense of discontinuity. In addition, waiting time is required to view the display and make a selection, which reduces the input speed.

Also, if you try to prepare a sufficient dictionary for compound words such as several kanji and idioms, for example, 5000011N
It became a dictionary that spanned the length of S! on a mini-floppy disk. It will not fit within 300KB.

Furthermore, if you try to input the Tonori phrases by dividing them into kanji or idioms, the manual operation becomes extremely troublesome.

The main purpose of the present invention is to input the pronunciation of a kanji or compound word and convert the kana character series into kanji by searching for a match with the pronunciation of the word in a dictionary. The aim is to streamline the procedure for finding the information and improve the input speed.

Another object of the invention is to use a relatively small number of phrase dictionaries, and
The aim is to make it possible to process compound words with extremely simple operations.

In order to achieve the above object, the present invention provides all readings of individual idioms and kanji in the dictionary, including sounds and kun, and idioms in addition to the general reading of idioms. , along with the other readings of the kanji that make up the idiom. If the kanji cannot be determined unambiguously from the input reading, press the different reading key to input another reading. If the two readings entered using the different reading keys can exclude other homophones or idioms that appear as candidates in the first reading, then the kanji with those two readings or VS words are automatically written manually.

Furthermore, in the present invention, when inputting a compound word, press the kanji key to enter the kanji mode, and then input the reading of the compound word by pressing the division key for each kanji or compound word. It has become.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 shows the configuration of an embodiment of a device realizing the Japanese input method according to the present invention, where l is an alphabetic keyboard for inputting kana characters or Roman characters, 2 is a display device, and 3riRO
M, 4 is a processing device, and 5 is a RAM.

6 is a storage device, and 7 is a pass line.

In FIG. 1, the output of the keyboard l for inputting kana characters or romaji is sent to the processing unit via the bus line /7. Transferred to f4. The keyboard has character keys 9 such as alphanumeric characters and kana, a space key 10, and a kanji key 11, as shown schematically in FIG. 2 and details of the keys in FIG. It is composed of input mode designation keys such as a hiragana key 12, a katakana key 13, and a separate reading key 14. The output signal of the keyboard 1 generated by pressing the character key 90 is processed differently within the processing device 4 depending on which of the input mode designation keys 11, 12, 13, 14 has been pressed before. As shown in the figure, it is displayed on the display surface 8 of the display device 2. The ROM 3 and RAM 5 mainly store data indicating processing procedures in the processing device t4. Furthermore, the storage device 6 is a storage device that stores kanji and phrase dictionaries.

FIG. 5 shows a processing flow in the apparatus shown in FIG. 1, which is mainly executed by the program of the processing device 14.

In FIG. 5, when inputting kanji, in step 18 the kanji key 11 is first pressed, and in step 20,
The pronunciation of the kanji or compound word to be input is entered as a kana character sequence using the character key 9. In the general input method, route I is followed, and in this case, in step 21, the space key 10 as a division key is pressed. This results in
In step 22, a dictionary search, that is, a matching of human kana character sequences and pronunciations stored in the dictionary, is performed sequentially for all kanji or compound words in the dictionary. As a result of the check in step 23, if there is only one kanji or compound word with matching readings,
In step 24, the user is prompted to enter this as a confirmed kanji, along with the leg that is automatically input. If there are multiple kanji words in the dictionary with pronunciations that match the input kana character sequence, in step 25, the first-ranked word is displayed blinking on the display device to inform the user that there are homophones. Let me know. If the first ranked word displayed in the check in step 26 is the kanji or compound word that the user wants to input, in step 27 the user presses the space key 10 as an input key.

fixed- As a result, the first ranking word is input as a sure word.

If the first JII placeword does not match the required word, press the separate reading key 14 in step 28, and step 2
Use the character key 9 to read the required kanji or compound words in 9.
Enter more information. Note that the separate reading key 14 also serves as a classification key. As a result, a dictionary search for homophones is performed in step 30, and if there is a word that is additionally input in the check in step 31 and has a different reading that matches the nine-betsu reading, this word is input as a confirmed word.

If there is no word with a different reading that matches the entered different reading, enter another reading. Press -14 to enter a different reading. Repeat this operation until the kanji or compound word is determined. If you press the kana key after determining the kanji, this is interpreted as inputting the reading of the next kanji or compound word. If you wish to input a hiragana or katakana other than a kanji character, press the respective mode key 12 or 13. In step 32, it is determined that the next character is not a kanji character, and the selection is exited from the kanji mode. If you feel that the first reading is not finalized after inputting it, you may take the route shown by ■ and enter the separate reading by pressing the separate reading key 14 from the beginning instead of pressing the classification key. The procedure that takes this route is a predictive input procedure, and is highly effective in improving the user's spontaneity and input speed.

FIG. 6 shows an example of the contents of the kanji and phrase dictionary stored in the storage device 6 of FIG. 1, and the figure shows an example of a phrase having the pronunciation "Kishiya". ■ in the diagram
is the primary reading of each idiom, and ■@ is the reading of each kanji that makes up the idiom, and includes a reading that is effective for distinguishing between idioms.

FIG. 7 shows a processing procedure in the processing device 4 for determining phrases using this dictionary. The procedure shown in the figure is for inputting the part "Yesterday at your company" in the text on the display screen 8 in FIG. 4. First, press the kanji key 11 in Figures 2 and 3 to enter the kanji mode, and read the idiom ``yesterday'' as ``
Enter "Sakujitsu". Next, the space key 10 is pressed to indicate the end of inputting the pronunciation of one compound word. in this case,
The nine space key 10, which is pressed when the kanji is not determined, serves as the division key in FIG. This starts the dictionary search. In general, there are two idioms that have the reading ``sakujiri'': ``yesterday'' and ``sakuhi,'' but since ``sakuhi'' is only used in special sentences, here we input the reading ``sakujitsu.'' Suppose then that only "yesterday" is searched and returned. Therefore, in this example, "yesterday" is determined and input by inputting the pronunciation "Sakujitsu". At the same time, a confirmation sound is emitted to notify the user that the confirmation has been made.

Once the preceding idiom or kanji has been determined, enter the next word, here the pronunciation of "Kishiya", and press the space key 10 as the classification key. This starts a dictionary search, but in this case, as shown in FIG. 6, there are a plurality of phrases with the same sound. When a homophone exists, the first I in the dictionary
[The place word, i.e., "train", which is the idiom for i = 1 in FIG. 6, is displayed blinking on the display surface 8 of the display device 2 to notify the user of the existence of the homophone. At this time, after pressing the user's intermediate reading key 14, the user inputs the reading of a kanji character constituting the idiom to be input that is thought to be able to distinguish the idiom from other homophone idioms.

In this example, we input "your company" with i = 4, so if we input "toi" or "yashiro" as separate readings,
There are no other idioms that have these different readings, so ``your company'' is confirmed.

That is, in FIG. 7, if a homophonic phrase in the dictionary is represented by Wi, then when the next D is positive, that phrase is selected as the phrase to be input.

D = ■ , otherwise 0. Also ■.

■...searches for other readings from the beginning, and as shown in FIG. 6, if an underlined character is input, ■=lfi or ■=1 is set. Characters with different readings, such as underlines, are the core of distinguishing each idiom from other idioms, and if only these characters are input, the idiom is confirmed. That is, in the above equation, D>0. In the above example, as shown in Figure 6, if only one character is input as a separate reading, that is, if "to" and maku are inputted as "ya", D) 0 and nashi "your company" are determined. When the confirmation is confirmed, a confirmation sound is sounded and the user is notified of the confirmation, so the user can skip the subsequent separate reading input and proceed to the next step.

As a result of searching for all the same-sounding idioms Wi, if the idiom is not determined, the user presses the separate reading key 14,
Enter another reading and repeat the mental process described above.

Also, in Figure 7, after inputting the first reading ``Kishiya'', directly press the separate reading key 14 to say ``To...'' and the meaning is ``Ya...''. By inputting , it becomes a predictive input procedure, and there is no need to proceed to the next step after pressing the space key 10 and watching the flashing "train" on the display screen, greatly improving the input speed.

As explained above, according to the present invention, in the Japanese-bunjin-power method, which performs kana-kanji conversion by seeking a match between the input reading of kanji or compound words and the reading of built-in kanji and compound words, without using syntactic analysis. , if there are homophones,
Instead of passively showing the user a Someme list and selecting the desired word from it as in the conventional method, the user can directly enter the alternative pronunciation of the desired word. You can search for the desired word.

In this case, look at the homonyms lined up on the display screen,
Compared to conventional methods, such as pressing the number key corresponding to the number of the required word, or sequentially displaying homophones at the end of a sentence on the display screen, the time required to determine the required word is much shorter. Time is also reduced. As shown in the example in Figure 6, in many cases, it is not necessary to enter the separate reading completely, and it can be confirmed with a - character or two characters, and if a homophone is expected, # As shown by the path ■ in Figure I1, active input procedures such as inputting the primary reading K followed by the separate reading can be carried out in advance, reducing the user's waiting time.

As described above, according to the present invention, it is possible to realize a compact and inexpensive Nihonbunjinkiri system that focuses on dictionary searches, and the practical value is that the input is done naturally and with active intention for the user. It is.

Also, when inputting a compound word, if you press the kanji key and enter the kanji code at the beginning of the training, all you have to do is press the kanji key to separate the kanji or compound words that make up the compound word. Or, since a compound word can be input as a compound word, there is no need to store the compound word itself in memory, and the compound word input operation becomes easy.

[Brief explanation of the drawing]

Figure 1 is a block diagram of an example of a device that implements the Japanese language human power method according to the present invention, Figure 2 is a schematic diagram showing the overall configuration of the keyboard in Figure 1, and Figure 3 is an arrangement of function keys on the keyboard. FIG. 4 is a schematic diagram of the display surface of the display device 1 of FIG. 1, FIG. 5 is a flowchart showing an example of a processing procedure according to the present invention, and FIG. 6 is an example of the configuration of a dictionary according to the present invention. FIG. 7 is a flowchart showing the processing procedure in the embodiment of the present invention when the dictionary shown in FIG. 6 is created. 1...Keyboard, 2...Display device, 4...Processing device, 1st mouth ¥12 Mouth fan 4 eyes yI) 6 (211

Claims (1)

[Claims] 1. A memory means for storing a kanji dictionary and a compound word dictionary is provided, and after pressing a kanji designation key on the keyboard, the pronunciation of the kanji or compound word can be read using the kana keys or alphanumeric keys on the keyboard. In the Nihonbunjinriki method of inputting and exchanging kana character sequences to kanji by searching for a match with the pronunciations of kanji or compound words stored in the memory means, the memory means stores alternative pronunciations of the kanji and compound words, and A separate reading key is provided on the above keyboard, and after inputting the reading of the kanji or compound word, press the separate reading key,
A feature is that the alternate reading of a kanji or compound word is additionally input using the kana key or the alphanumeric key on the above-mentioned keyboard, and a Go is found that matches both the initially input reading and the additionally input alternate reading. Japanese literature and human power method. 2. Provide a category key on the above keyboard, and when inputting a compound word, press the kanji designation key above, and then press the category key for each kanji or compound word that makes up the compound word to read the compound word. 1. The Japanese human resource system according to claim 1, wherein the input data is input by disassembling the information.