JPS5922130A - Converter of japanese syllabary to chinese character - Google Patents

Abstract

PURPOSE:To change the definition and analysis sequences of a syntax easily, by changing the destination of a transition table. CONSTITUTION:A KANA (Japanese syllabary) input device 2 receives a control signal from a control part 1 to send an input character string to a character string buffer 3. An analyzing device 10 refers to a transition state number in a transition table storage device 6 to retrieve the contents of dictionary storage devices 8-1-8-n. The contents of buffers 3 and 4 and a connection pointer 9 are updated successively and when KANA character strings are all analyzed, an output device 5 reads a finally obtained KANJI (Chinese character) and KANA mixed character string out of the buffer 4 and outputs it. Therefore, when definitions and analysis sequences of syntax are changed, the destination of the transition state number when it can be retrieved are changed to facilitate the changing.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention inputs a kana character string, performs character string matching on a dictionary, and then converts it into an appropriate tJ as a Japanese sentence.
This invention relates to a kana-kanji conversion device that outputs sentences containing kanji and kana.

Currently, the first method for Japanese fetal power has been proposed and is in actual use, and one of them, the kana-kanji conversion method, converts Yamabe's kanji-kana mixed sentences into their reading form. This is an input method. This method allows thousands of types of kanji to be entered using just a few dozen keys, so it is widely used in Japanese word processors and other Japanese information processing devices. It is used.

For example, the kana-kanji conversion method takes Japanese syntax as input, as shown in equation (1). ) ・・・・・・・・・・・・
...(In Equation 11 (1), [ ] is the omitted town's insect eye, and AIB means person or B.

The kana-kanji conversion method searches various dictionaries containing words corresponding to the elements of Japanese syntax using the input kana character string in the steps indicated by the Japanese syntax, and when the word is currently ascended, the kanji of the word is searched. It is a method to obtain the desired kanji-kana mixed sentence by outputting the notation. To realize this method, 1ζ conventionally performs the IJ character string analysis method using a block-forming glosser, as shown in Figure 1. Or, a means was used to perform +' by constructing a simple device. Therefore, for example, if the expression (1) is (input syntax) 2 [(prefix)] (nominal) [(nominal) )
” [(suffix)) l (published by Jfj) ・・・・・・・・・
...If you are trying to change the body's anti-ff (W codeword) to something like (2), you will have to do a lot of work, such as modifying the program or recreating the special equipment. It required a lot of work. (However, [ ] means 0 or more repetitions.) The purpose of the present invention is to eliminate such conventional drawbacks and to develop a kana-kanji kukan ha that can change the syntactic definition and analysis procedure to be more accurate. Our goal is to provide the following.

The present invention provides a transition table storage means that stores Japanese syntactic information as a transition table, and uses the 1M table to determine search procedures for various dictionaries corresponding to the syntactic information. Therefore, even when changing syntactic information (even the dictionary search procedure), it is possible to change to the desired kana-kanji conversion device by simply modifying the transition table, without modifying the program or special equipment. be able to.

The present invention will be specifically described below with reference to the drawings. FIG. 2 is a block diagram showing a specific embodiment of the present invention. In the figure, 1 is a control unit that controls the entire control, 2 is a kana character string input device for inputting a kana character string, and 3 is a kana character string input device for storing the kana character string input from the kana character string input device 2. , 4 is a kana character string buffer that has the function of indicating the unconverted kana character string part during the kana-kanji conversion process, and 4 is a kanji-kana mixed character string buffer 7 for temporarily storing the converted kanji-kana mixed character string. ．．
5 is a kanji/kana mixed character string output device for reading out the contents of the kanji/kana mixed character string buffer 74, and 6 is a transition table storage device that stores Japanese syntactic information in four tables in a transition diagram or a form similar to a transition diagram. , 7 is a transition pointer for storing a position in the transition table storage device 6 to represent the processing progress state during the kana-kanji conversion process, and 8-1 to B-n are provided with at least the pronunciation and notation of the word. dictionary storage device, 9-1
~9-n is a dictionary search device for searching the dictionary storage devices 8-1 to 8-n, and IO is a dictionary determined by the contents of the kana character string buffer 3, the transition table storage device 6, and the transition table pointer 7. One of the storage devices 8-1 to 8-n is searched by one of the dictionary search devices 9-1 to 9-H, and when the search is successful, the kanji/kana mixed character string buffer 4 and This is an analysis device for updating the kana character string buffer 3 and also updating the transition pointer 7 according to the search result.The operation of the entire device when a kana character string is actually input will be described below. Note that in this specification, words include not only independent words such as nouns and proper nouns, but also affixes (prefixes,
It is assumed that the interpretation is broad, including the suffix), etc.

Now, suppose that the kana character string is inputted into the kana character string input device 2 as "Daihatsu Gen".The zero character string input device 2 receives the control signal from the control unit l and converts the input character string into characters. Send to column buffer 3. Figure 3 (a) (b) (c)
is a diagram for explaining an example of the kana character string buffer 3 and its functions.0 In Fig. 3(a), 301 is a kana character string field for actually storing a kana character string, and 302 is an input field. A kana character string length field 303 is used to store the length of the converted kana character string, and is an unconverted character string pointer indicating the beginning position of the unconverted part. The contents of the kana character string buffer 3 immediately after the kana character string is input are as shown in FIG. 3(b).

FIGS. 4(a), 4(b), and 4(c) show an example of the structure of the Kanji/Kana mixed character string buffer 4 and explanations for its functions.

In FIG. 4(a), 401 is a kanji/kana mixed character string field that actually stores the kanji/kana mixed character string;
2 is a kanji/kana mixed character string length field for storing the length of the kanji/kana mixed character string.

The transition table storage device 6 is a table that describes the syntactic information of Japanese candy, and specifically describes the procedure for analyzing input character strings.

FIG. 5 is a diagram showing an example in which the input character string analysis procedure of FIG. 1 is described in the form of a transition table. The transition pointer 7 points to the transition state number in the transition table storage device 6. (Initial value is 1
It is assumed that is set. ) Also, the prefixes, independent words, etc. of the search target dictionary entries in the transition table storage device 6 are information that point to the dictionary storage devices s-i to 8-n, respectively.

(For convenience of explanation dictionary storage device 8-1.8-2.8-3.8
-4 are respectively a nominal dictionary, a pragmatic dictionary, a prefix dictionary, and a suffix dictionary. ) Upon receiving the control signal from the control unit 1, the analysis device IO first refers to the record with the transition state number = 1 in the transition table storage device 6 pointed to by the transition pointer 7, and searches the search target dictionary for the nominal dictionary (the above-mentioned dictionary). In this case, the dictionary storage device 8-1 (!:
fAru), and issues a search request to the dictionary search device 9-1 using the unconverted kana character string "Taiharaken" in the kana-mata character string buffer 3 as a key. Dictionary search device 9-1 +,
! Upon receiving a signal from the control section 7, the first storage device 8-1 is searched using, for example, the longest match method. Here, the dictionary search device 9-1, which assumes that no suitable word exists in the dictionary storage device 8-1, informs the analysis device 10 that the search has failed. The screen analyzer @10 receives the Iv report and the signal from the control unit 1, and again refers to the record with the Jfl transition state number = 1 in the transition table storage device ji6 indicated by the transition pointer 7, and determines the unanalyzed data. Since the character string still exists and the search failed, the transition state number = 2, which is the transition destination that cannot be searched.
is set in transition pointer 7. Next, the analysis device 10
Upon receiving the signal from the control unit l, the record with the transition state number = 2 in the transition table storage device 6 pointed to by the transition pointer 7 is referred to, and the dictionary to be searched is set to the dictionary storage device 8 (in the above assignment, 8-2)) and the dictionary search device 9-
2, the unconverted kana string in character string buffer 3 “
Make a search request using "Daiharaken" as the key. The dictionary search device 9-2 uses the same method as the dictionary search device @9-1.
Search the dictionary storage device 8-2. It is also assumed here that there is no suitable word in the dictionary storage device 8-2. Dictionary 1 search device M
9-2 notifies the analyzer 101 that the search has failed.

The analysis device 10 receives this information and the signal from the control unit l, refers again to the record with the transition state number = 1 in the transition table storage device 6 indicated by the transition pointer 7, and determines that an unanalyzed character string still exists. Since the search has failed, the transition state number=3, which is a transition destination that cannot be searched, is set in the transition pointer 7. Next, the analysis device IO receives the signal from the control unit l, refers to the record with transition state number = 3 in the transition table storage device 6 pointed to by the transition pointer 7, and changes the search target dictionary to the prefix dictionary ( In the above allocation, the dictionary storage device 8-3 is determined as the dictionary storage device 8-3), and the unconverted kana character string "Taiharaken" in the character string buffer 3 is determined as the dictionary storage device 8-3.
The dictionary search device 9-3, which issues a search request using the key, searches the dictionary storage device 9 in the same manner as the dictionary search device 9-1.
-Search for 3. As a result, it is assumed that "large" corresponding to die J is retrieved.

The dictionary retrieval device 9-3 informs the analysis device 101 that the retrieval was successful and the corresponding “dai” was obtained.

The analysis device IO receives the search result from the dictionary search device 9-3 and the signal from the control unit 1, and inputs the kana character string buffer 3.
, the contents of the Kanji/Kana mixed character string buffer 4, and the contents of the connection pointer 9 are updated. As a result of updating,
The kana character string buffer 3 is in the state shown in FIG. 3(c), and the kanji/kana mixed character string buffer 4 is in the state shown in FIG. 4(b).

Next, the analysis device 1o examines the kana character string buffer 3,
Confirm that the unparsed character string remains, refer to the transition table storage device 6, determine the migration destination as 4 if the record with transition state number = 3 can be searched, and set the transition pointer 7 to 4. Update.

The flow from here on is almost the same as the one described above, so only the outline will be explained. The analysis device lo refers to the record with the transition state number=4 in the transition table storage device 6 and searches the dictionary search device 9-1.
The dictionary storage device 8-1 (nominal dictionary) is searched through.

(At this time, search for ``Kana heading ``X-kken.'')
As a result, it is assumed that "discovery" corresponding to "Haraken" is searched.

The search results are sent to the analysis device H1, and the kana character string buffer and the mixed kanji/kana character string buffer 4 are updated in the same manner as described above.

At this time, the character string buffer 4 containing kanji and kana is shown in Figure 4 (
The state becomes c).

Then, when there are no more unanalyzed kana character strings, a signal is received from the control unit 1 (because the procedure when there are no more unanalyzed characters in the record with transition state number = 4 ends as an analysis success). Kanji/kana mixed character string output device 5
As a result, the obtained Kanji/Kana mixed character string can be read out from the Kanji/Kana mixed character string buffer 4.

FIG. 5 is a diagram showing an example of formula (1) described in [81] written in the transition table storage device 6. By using formula (2) instead of formula (1), it is possible to If you want to change the syntax definition and analysis procedure so that it is possible to input words (successive independent words), in Figure 5, change the transition destination when the transition state number is 1 and 4, which are searchable. 5 to 4, and change the migration destination to 5 when the transition state number=4 cannot be searched.

As described above, according to the present invention, the syntax definition and analysis procedure can be easily changed. By inputting a Naorina character string and searching the dictionary storage means, you can find Kanji or η[
The method of extracting mixed sentences has already been realized by several methods, so a description thereof will be omitted.

In addition, in FIG. 2, the dictionary search device [9-1 to 9-n
Although it is described that the same number of dictionary storage devices t8-1 to t8-n are prepared, in reality, the number does not necessarily have to be the same. For example, one dictionary storage device may have multiple dictionary storage devices. It is also quite possible to take the form of searching a dictionary storage device. Further, the format and configuration of each device used in the explanation of the present invention, the product classification, syntax format, dictionary search device, etc. that appear in the explanation are merely examples, and there are no limitations on the present invention. A) It is not a given.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing the analysis procedure of an input character string in kana-kanji conversion, Fig. 2 is a block diagram showing an embodiment for realizing the present invention, Fig. 3 (a) (b) (c) Figures 4(a) and (c) are diagrams for explaining an example of the configuration of a kana character string buffer and its function implementation, and Figures 4 (a) and (c) are diagrams for explaining an example of the configuration of a character string buffer containing kanji and kana and its functions. Figure 5 is a diagram showing an example of what is written in the transition table storage device. In the figure, l is a control unit, 2 is a kana character string input device, 3 is a kana character string buffer, and 4 is a mixture of kanji and kana. buffer,
5 is a character string output device containing kanji and kana, 6 is a transition table storage device, 7 is a transition pointer, 8-1 to 8-n are dictionary storage devices, 9-1 to 9-n are dictionary search devices, and IO is an analysis device 0 1 ζ (!) ノ m whisper 2 Illustrated funeral collection 4 diagrams each showing

Claims (1)

[Scope of Claims] 1. A kana character string input means for inputting a kana character string, and during the kana-kanji conversion, at least an unused part of the kana character string input from the kana character string input device. a kana character string storage means for holding a converted part; a plurality of dictionary storage means each having at least seven pronunciations of words;
a dictionary search means for searching the dictionary storage means; and a character string updating means for outputting the notation of the word searched by the dictionary search means and updating an unconverted portion in the kana character string storage means. In a kana-kanji conversion device that makes it possible to obtain a kanji-kana mixed character string from a human-generated kana character string by It has a self-memorizing transition table storage means, and determines the transition destination in the dictionary storage means to be searched and the h-transition table storage means to which the next jζ transition should be made based on the M syntactic information possessed by the transition table storage means. A kana-kanji conversion device characterized by: 2. The transition table storage means has at least information for determining the dictionary storage means to be searched and information regarding the next transfer destination. Kana-Kanji conversion device described in Section 1. 3. The kana-kanji conversion device according to claim 1, wherein the transition destination in the transition table storage means is determined by a search result obtained by the dictionary search means ζ.