JPS61118869A - Kana/kanji converting sentence editing device - Google Patents

Abstract

PURPOSE:To improve the efficiency of processing by providing a device for delimiting a sentence of KANA (Japanese syllabary) to a clause. CONSTITUTION:A sentence of KANA (Japanese syllabary) inputted from an input device 2 is edited and processed by an editing device 3, and a KANA sentence file is prepared and also displayed on a display device 4. The sentence of this sentence file is sent to an analyzing device 6 by a sentence unit, a KANA string of the sentence unit is delimited to a clause unit by utilizing a grammar information dictionary 7 and a KANA to KANJI (Chinese character) converting dictionary 8, and also a file of clause delimiting information which has combined the number of characters of the clause by a sentence unit is prepared. In a homonym selecting part C, from an input sentence file which has been sent from a clause delimiting part B, and a clause delimiting information file, a sentence of KANA is converted to a KANA and KANJI mixed sentence by referring to the KANA to KANJI converting dictionary 8, and displayed on a display device 10. A user selects a homonym by seeing the display device 10.

Description

DETAILED DESCRIPTION OF THE INVENTION Technical Field The present invention relates to a kana-kanji conversion text editing device used in Japanese word processors, Japanese telex terminals, and the like.

<Prior art> A kana-kanji conversion device converts input kana sentences into clauses (formed by words and adjuncts or particles).

Performs kana-kanji conversion in units of . For this reason, conventionally,
When inputting a sentence, the user had to divide it into phrases. Therefore, inputting text takes time and work efficiency decreases.

Furthermore, in the past, following the input of a sentence, a series of processes such as kana-kanji conversion and homophone selection had to be performed in succession, so the next input was not possible until this series of processes was completed. This was also a cause of reduced efficiency. In addition, each input device requires one dictionary stored in a storage medium such as a floppy disk or ROM, and a kana-kanji conversion program, which has the disadvantage that the kana-kanji conversion device is expensive. Was.

<Object of the invention> The present invention has been made in view of the above circumstances, and its purpose is to eliminate the need for the user to separate phrases,
To provide a kana-kanji conversion text editing device capable of independently performing each process of input, kana-kanji conversion, and homophone selection.

<Configuration of the Invention> The present invention includes a first processing means that inputs kana sentences and creates a kana input sentence amount file, and divides the kana sentences from the above-mentioned kana input sentence file into phrases, and generates phrase break information. a second processing means for creating a file, converting kana sentences from the kana input text file into sentences containing kana and kanji using information from the bunsetsu delimiter information file, and converting the kana sentences from the kana input text file into sentences containing kana and kanji from among the converted homonyms; and a third processing means for selecting the kanji of the above-mentioned first. 2nd and 3rd
It is possible to independently execute the processing of the processing means.

<Example> An embodiment of the present invention will be described below.

FIG. 1 shows the configuration of the Hakana-Kanji conversion text editing device. This kana-kanji conversion text editing device 1 is composed of an input section A1, a clause separation section B, and a homophone selection section C.

The input unit A includes an input device 21, an editing device 3, and a display device 4. In the input device 2, the user manually enters a sentence in kana. The editing device 3 edits the input kana text and creates a kana input text file. Then, the display device 4 displays this edited kana text.

This input section A only performs input and editing of kana sentences, and does not perform conversion processing into kanji. In addition, Kanato is
Contains hiragana, katakana, alphabets, numbers, and symbols.

The bunsetsu separation section B is controlled by the control device 5. The analysis device 6 consists of a grammar information dictionary 7 and a kana-kanji conversion dictionary 8. Control device 5
sends the sentences in the kana input sentence file sent from input section A to the analysis device 6 in sentence units (character strings up to commas), and selects homophones from the phrase break information file sent from the analysis device 6. Send to Department C. The analysis device 6 uses the grammar information dictionary 71 and the kana-kanji conversion dictionary 8 to divide the sentence-based kana string into phrases, and creates a phrase separation information file that combines the number of characters in the phrases for each sentence.

The homophone selection section C includes an operating device 99, a display device 10, a control device 11, and a conversion device 12, and also includes the above-mentioned kana-kanji conversion dictionary 8 in common with the clause separation section B. On the operating device 9, the user performs an operation to select a desired kanji from homonyms while looking at the display on the display device 10. The control device 11 sends the kana input sentence file and the bunsetsu break information file sent from the bunsetsu separation unit B to the conversion device 12, and also converts the homonyms sent from the conversion device 12 according to instructions from the operation device 9. Select one candidate from among the candidates. The conversion device 12 converts the kana string in phrase units sent from the control device 11 into kanji using the kana-kanji conversion dictionary 8, and finds all homophone candidates if there are any. In this homophone selection section C,
In addition to kana-kanji conversion and homophone selection, final editing processing is also performed.

Next, the processing procedure of the above-mentioned kana-kanji conversion text editing device 1 will be explained with reference to the flowchart of FIG. 2. Note that in the flowchart of FIG. 2, solid input means inputting a kana sentence without dividing it into clauses, and hereinafter, the kana input in this case will be referred to as solid input.

First, input a full kana sentence into input section A. The input unit A creates one solid input sentence file from one sentence of this solid input. Then, this solid input sentence meaning file is sent to the phrase delimiter B.

Information is transmitted from input section A to bunsetsu separation section B via LAN.
It can also be done through (internal information communication network),
The information registered on the floppy disk may be read by the bunsetsu separator B. In this input section A, when one sentence has been inputted, by pressing the input end key of the input device 2, the inputted sentence is converted into a dedicated text file, distinguished from a normal text file, and stored on a floppy disk, for example. Register. This completes the text file that can be used in the bunsetsu section B.

The clause dividing unit B reads the solid input text file sent from the input unit A according to a user's instruction or automatically, and performs a process of dividing the kana text into clauses on a sentence-by-sentence basis. This process checks whether the kana or kana string from the beginning of the sentence is registered as a word in the kana-kanji conversion dictionary 8, and then uses the grammatical information dictionary 7 to check whether the context of the word is grammatically correct. and find out. If there is no contradiction, the process of determining 0 or more clauses is repeated until the end of the sentence, and when the entire sentence has been segmented, a clause break information file is created. Then, together with the input text file input from the input section A, this phrase delimiter information file is sent to the homophone selection section C.Normally, this process in the phrase delimiter B takes the longest time.

Homophone selection section C converts kana sentences into sentences containing kana-kanji by referring to kana-kanji conversion dictionary 8 from the solid input sentence file and bunsetsu separation information file sent from bunsetsu separation section B, and converts the kana sentences into sentences containing kana-kanji. A sentence containing simple kanji is displayed on a display device 10.

Then, the user selects a homophone while looking at the display device 10. After that, editing processes such as changing the final format and text layout are performed to create a file containing kana and kanji.

The processes of the input section A1, the clause separation section B, and the homophone selection section C described above can be executed independently.

Then, the processing of the bunsetsu delimiter B starts at the input part A.
Or it is instructed by the homophone selection section C. The control device 5 of the bunsetsu separation section B and the control device 11 of the homophone selection section C
It is also possible to combine them into one, so that when the homophone selection section C does not perform the processing, the clause separation section B performs the processing. In this case, the processing speed in the homophone selection section C is the same as in the conventional case, so if the processing in the phrase separation section B is performed using the free time, the processing efficiency will be improved. Furthermore, since the bunsetsu separation unit B and the homophone selection unit C share a control device, it is possible to reduce the cost of the kana-kanji conversion text editing device.

In addition, the configuration of the input section A is simple, and multiple inputs are possible.
) The force part A can also be connected to the bunsetsu separation part B.

In this case, a dictionary and a kana-kanji conversion program can be commonly used for a plurality of input devices, so a Japanese input device with good cost performance can be realized.

Hereinafter, the process of dividing phrases will be explained based on a specific example.

Figure 3 (al, (bl, C) shows the solid input text file created in the input section A, the phrase separation information file created in the sentence separation section B, and the kana-kanji created in the homophone selection section C. An example of the contents of each mixed text file is shown below.If you divide the sentence ``Kyo no Tenki wa Haretzu.'' in the solid input text file into clauses, you will get ``Kyo no'', ``Tenki wa'', and ``Haretzu.''. ", and the combination of the number of characters in each clause is 445. Using this 445 as clause delimiter information, a clause delimiter information file is created.

Now, in the bunsetsu separator B, the kana sentence ``Today's tenki wa harezu.'' shown in Figure 3(a) is extracted from the solid input text file sent from the input section A. Suppose that FIG. 4 shows the extracted contents of the kana-kanji conversion dictionary 8 for the kana that make up this sentence. Further, FIG. 5 shows a part-of-speech connection table of the grammar information dictionary 7.

First, find the word in the dictionary that has the largest number of characters from the beginning of the sentence. "Ki" is in the dictionary as a word, and
Both "Kyo" and "Kiyou" are in the dictionary. ``Kyouno'' is not a word in the dictionary. Furthermore, in the dictionary, “Today’s...
Since there is no word with 5 or more characters such as ``--'', ``kyo'' is a candidate.

Since ``kyo'' is a noun, the 110 parts of speech of the next word include nouns, suffixes, and particles. "No" is in the dictionary as a noun. There is no word with more than two letters such as ``note-''. There is no ``no'' as a suffix, and ``no'' is used as a particle.
There is. What about “today” and “field”? Since Jt does not form a compound word, "no" as a particle becomes a candidate after all.

The word that follows the particle "no" is a noun.

``te'', ``ten'', and ``tenki'' are found in the dictionary as nouns. There is no noun with four or more letters such as ``Tenkiha-'' in the dictionary. Therefore, the noun "weather" is a candidate.

If the longest word in the dictionary does not continue, start over with the next longest word.

"Haha" is in the dictionary, but the next word is "re-・-" and is not followed by "hahare." Since ``weather'' and ``ha'' do not follow each other, the adverb ``wa'' becomes a candidate.

From here on, "Haru" and "Desu" are obtained in the same way.

In this way, for a sentence input in solid text, the kana string is expressed as the word in the dictionary and the part-of-speech relationship between the characters before and after the word as part of the word #i.
The sentences are divided into clauses by referring to the connection table and comparing them, and repeating this process up to the punctuation mark of each sentence.

<Effects of the Invention> As explained above, the present invention is equipped with a device for dividing kana sentences into phrases, thereby eliminating the need for the user to perform the troublesome task of dividing kana sentences into phrases, thereby improving processing efficiency. can be increased. In addition, it is possible to independently execute each process of inputting sentences, segmenting each phrase, and selecting homophones, and by connecting one or more input devices, multiple users can share the dictionary and kana-kanji conversion program. , it is possible to improve processing efficiency and reduce the cost of the kana-kanji conversion text editing device. In addition, there is no need to divide into clauses during input, and the time-consuming clause division process can be performed independently of input processing.
It is also possible to input kana sentences by voice using a tape recorder, etc., and for example, it is possible to complete the input and phrase segmentation processes at night and only select homophones the next day.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, FIG. 2 is a flowchart showing the processing procedure of the embodiment of the present invention, and FIG.
Figure 4 shows the structure of the file; Figure 4 shows the structure of the Baka Kanji conversion dictionary; and Figure 5 shows the structure of the grammar information dictionary. A-・Input section B-・Bunsetsu separation section C--Homophone selection section 311th,,,,A branch blue 7-rA))/%5.
w+) +1” 1 (7 eyes, register dragon tilt °-!) 1 eyes)
7

Claims (1)

[Claims] (1) A first processing means that inputs kana sentences and creates a kana input sentence file, and a second process that divides the kana sentences from the above kana input sentence file into clauses and creates a clause separation information file. and a third method for converting a kana sentence from the kana input text file into a kana-kanji mixed sentence using information from the bunsetsu delimiter information file, and selecting a required kanji from the converted homonyms. A kana-kanji conversion text editing device comprising a processing means, and capable of independently executing the processing of the first, second and third processing means.