JPH06203014A - Kana to kanji converter - Google Patents

Abstract

PURPOSE:To constitute a converter so that when other character, etc., than a KANA (Japanese syllabary) are inputted, its character, etc., are set as delimiters for a grammatical analysis, the grammatical analysis is executed, and thereafter, KANA to KANJI (Chinese character) conversion is executed. CONSTITUTION:When a character is inputted (701), the kind of the character is discriminated (702). In the case the inputted character is a KANA, 4a write processing is executed from a phrase portion by a prescribed grammatical analysis together with a KANA character-string inputted alreaby (708 to 711), and it is outputted as a KANA and KANJI mixed phrase candidate (715). In the case the inputted character is other character, etc., than a KANA, the connection with the KANA character-string inputted already is not executed, and its character, etc., become delimiters for a grammatical analysis (704). In this case, the inputted character, etc., form a phrase by one character (705). Thus, the delimiter for the grammatical analysis is defined, and the conversion efficiency can be improved.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a kana-kanji conversion device for converting a kana character string input for inputting a Japanese sentence into a sentence mixed with kana-kanji.

[0002]

2. Description of the Related Art In a conventional kana-kanji conversion device, when performing kana-kanji conversion, the user presses the conversion key or equivalent conversion instruction after inputting all kana character strings. In addition, in order to convert an input long kana character string into a kana-kanji mixed sentence, an analysis process such as the two-segment longest matching method is applied to the entire input kana character string.

[0003]

However, according to such a conventional technique, in the kana-kanji conversion device, after inputting all the kana character strings to be converted into kana-kanji, the conversion key is pressed or a conversion instruction equivalent thereto is input. Therefore, there is a problem that the conversion operation is performed for the first time, and therefore the conversion instruction operation must be performed one by one. Moreover, since the processing such as grammar analysis for Kana-Kanji conversion is applied to the entire Kana character string input so far, in order to improve the conversion efficiency, the conversion instruction should be placed at a position where it is easy to grammar analyze. I need to do it. Therefore, the user has to consider the timing of instructing the kana-kanji change, and the flow of thought is interrupted each time. On the other hand, it was difficult to obtain a correct conversion result if the conversion instruction was given at an appropriate place.

The present invention has been made in order to solve the above-mentioned drawbacks, and has an object to realize a kana-kanji conversion which is performed automatically and sequentially without any conversion operation and which is excellent in conversion efficiency.

[0005]

A kana-kanji conversion device according to the present invention comprises an input means capable of inputting at least two or more of a plurality of types of character elements such as hiragana, katakana, letters, numbers and symbols. A kana-kanji conversion device for converting a character element string input from the input means into a kana-kanji mixed sentence, in which the character elements sequentially input from the input means are arranged and stored as an unconverted character element string. When a character element of a different type from the character element of the type input immediately before is input by the unconverted character element string storage means and the input means, the position of the character element of the different type is used as a delimiter in grammar analysis. , Analysis processing means for performing analysis processing such as grammar analysis and phrase segmentation processing for the unconverted character element string, and conversion means for performing kana-kanji conversion for the clauses obtained by the analysis. And summarized in that with.

[0006]

In the kana-kanji conversion device of the present invention configured as described above, the character element input from the input means is stored as an unconverted character element string by the unconverted character element string storage means. On the other hand, when a character element of a type different from the type of the already input character element is input, the analysis processing means,
Using the positions of character elements of different types as grammar analysis delimiters, analysis processing such as grammar analysis and phrase segmentation processing is performed on unconverted character element strings. The conversion means performs kana-kanji conversion for the bunsetsu obtained by this analysis.

[0007]

The preferred embodiments of the present invention will be described below in order to further clarify the constitution and operation of the present invention described above.

FIG. 1 is a block diagram showing an example of a well-known hardware structure of an apparatus for creating and editing a document as an embodiment. As shown in the figure, this device is provided with the following respective parts centered on the CPU 21 and interconnected by a bus 31. Each part mutually connected by the CPU 21 and the bus 31 will be briefly described.

ROM 22: Mask memory for storing kana-kanji conversion program, RAM 23: Readable / writable memory constituting main memory, Keyboard interface 25: Interface for controlling key input from keyboard 24, CRTC 27: Displayable in color A CRT controller for controlling signal output to the CRT 26, a printer interface 29: an interface for controlling output of data to the printer 28. In addition to these units, the bus 31 is provided with an expansion slot 30 for future expansion. Are connected.

With the hardware thus configured,
Text is input, Kana-Kanji conversion, editing, display, printing, etc. That is, the character string input from the keyboard 24 is subjected to predetermined processing by the CPU 21, and the RAM 2
3 is stored in a predetermined area, and a CRT is sent via the CRTC 27.
It is displayed on the screen of 26.

Next, the function executed by the hardware thus constructed will be described with reference to FIG. Figure 2
It is a block diagram which shows the function of the kana-kanji conversion device of this invention by a block. Reference numeral 1 in the figure denotes a central processing unit (CPU 21) that controls the entire system based on the data input from the input unit 2. The CPU controls all of the apparatus.

The kana characters input by the input unit 2 are sent to the kana-kanji conversion device 7 where kana-kanji conversion is performed, and while the bunsetsu candidates are displayed on the analysis result display unit 3, the bunsetsu candidate selecting unit 4 selects the bunsetsu unit. The bunsetsu candidate is selected and displayed, and the bunsetsu candidate deciding unit 5 decides a desired bunsetsu candidate or is automatically decided by the bunsetsu candidate internal confirming unit 19, and the kana-kanji character string is stored as a sentence in the document storing unit 6. To be done.

In the kana-kanji conversion device 7, the kana kanji, alphanumeric characters, symbols, control characters, etc. are received by the input character acquisition part 8, and the input characters are stored / corrected by the input character storage / correction part 9 based on these. , The stored character is 1
The characters are sent to the grammar analysis unit 15 on a character-by-character basis for grammar analysis. Further, when the input character is corrected, the fact is also sent to the grammar analysis unit 15 and the grammar is reanalyzed if necessary.

The word dictionary search instructing section 10 searches the independent word dictionary 12 by the independent word dictionary searching section 11 as needed based on the analysis result of the grammar analyzing section 15, and sends the information to the grammar analyzing section 15. The grammar analysis unit 15 searches the adjunct word dictionary 17 by the adjunct word dictionary search unit 16 based on the content of the independence word, performs the join test of the independence word and the adjunct word, and the phrase segmentation and writing process, and stores the analysis result in the phrase candidate storage unit 14. Is stored, and the analysis result output unit 20 outputs it in a phrase segmented state.

Further, based on the analysis result of the grammar analysis unit 15, the phrase candidate internal confirmation unit 19 requests the phrase candidate selection unit 4 to output the previous phrase / next phrase candidate or the input kana character string from the input unit 2. Correction or phrase candidate confirmation part 5
Unless there is a forced confirmation, the clause candidates will be internally confirmed if necessary. The internally fixed phrase candidate is deleted from the phrase candidate storage unit 14, and the corresponding kana character string is deleted from the input character storage correction unit 9. The word learning unit 22 learns the independent word usage frequency of the internally determined phrase information in the independent word dictionary.

For the phrase candidates displayed in the analysis result display unit 3 in the phrase segmented state, the phrase candidate selection unit 4 calls the previous candidate and the next candidate to obtain a desired phrase candidate. The analysis result output unit 20 outputs necessary phrase candidates from the phrase candidate storage unit 14 by the output instruction unit 13.
The read analysis result is output to the display unit 3. At this time, if there is no desired phrase in the phrase candidate storage unit 14 and a phrase candidate is requested, the phrase reanalysis unit 18 causes the independent word dictionary 1
2 and the adjunct word dictionary 17 are searched, and a grammatical analysis is performed by a method different from that of the grammar analysis unit 15 to store a new phrase candidate in the phrase candidate storage unit 14 and an analysis result output unit 20.
Is output by.

When the phrase candidate confirming unit 5 forcibly confirms the phrase candidate group displayed on the analysis result display unit 3, the confirmed phrase information receiving unit 21
The information of the decided phrase is received, and the word learning unit 22 learns the independent word usage frequency and the like in the independent word dictionary 12.

FIG. 3 is a flow chart for explaining the operation of the kana-kanji conversion device of the present invention. As described above, this kana-kanji conversion device receives a character element according to the operation of the keyboard 24 at 701, and verifies at 702 what kind of character this is. Here, if the input character is of a type instructing conversion, conversion is performed on all the kana character strings that have been input by 703, and the processing ends with all undetermined phrases. The kana-kanji conversion device according to the present embodiment has a function of automatically entering a conversion process when a character type changes, as will be described later. However, when inputting only the same type of character elements, for example, hiragana characters, the user There is also a case in which the user wants to perform the conversion process with the intention of, so that the conversion of all the input characters can be started by the input character (for example, the space key) instructing the conversion.

If the input character is an alphabetic character, numeral, symbol, or katakana, if the character before it is hiragana, then 70
Convert all undetermined hiragana character strings by 4, and at 705,
A phrase is formed by one character other than the entered hiragana. If the already entered character is something other than hiragana, if the character entered this time is of the same character type as the immediately preceding character, then that character is added to the preceding character's phrase to form a phrase (not shown). Otherwise, 705 forms a new phrase with one character,
The clause created in 705 is an undetermined clause,
The analysis result is output by 715 together with the conversion result in 4. After that, the process returns to 701 and the processing is repeated.

If the input character is an edit control character for correcting the input character, the input character is corrected by 706 and the target input character is grammatically analyzed again by 707.

If the input character is a hiragana character, an adjunct word concatenation process is performed on the input character by 708.
If it is necessary to search for an independent word by 9, the independent word dictionary is searched by 710, and further grammar analysis is performed by 711.

The results analyzed by 707 and 711 are
If there is a clause that can be confirmed by 712, if there is a clause that can be confirmed, the clause is confirmed by 713, the confirmed clause is output by 714, and the remaining analysis result is output by 715, but if not, the analysis result is output by 715. The process returns to step 701 and the processing is repeated.

4 to 8 are diagrams for explaining the operation of the kana-kanji conversion device of the present invention similarly. here,
An example will be described in which a kana character string and punctuations such as "Okusan ha kobu otou san." Are input in succession.

In FIG. 4 (a), the first character is input, and the independent word dictionary reads the position beginning with the first character as the head of the heading (the symbol "r" in the figure indicates the independent word dictionary). Indicates the start position of the heading for searching. (Omitted below), 2nd
Also for the fifth character, the bunsetsu candidate as shown in the figure is obtained for each input digit in a state in which it is collated with the independent word on the basis of "r" and the adjunct is concatenated. Here, since the candidates are always created in phrase units, the symbol "e" indicates the end of the phrase. Further, in this state, a candidate in which the difference between the last digit of the heading and the last digit of the input of the candidate is 3 characters or more is rounded down as no next concatenation (the symbol "*" in the figure is omitted).

The state of FIG. 4 (b) is a state in which the sixth character has been input, and here the first phrase candidate group cannot cut out objects up to the sixth character due to the concatenation of adjunct words. . Therefore, the next digit after the last digit of the phrase having the smallest digit with respect to the valid phrase is set as the heading first digit for reading the next independent word dictionary, and the independent word is read (hereinafter referred to as independent word analysis position calculation reading). Here, “out” and “out” are cut out, but due to the principle of the minimum number of bunsetsu, there is a first bunsetsu having the same bunsetsu end position, so it is cut off (symbol “c” in the figure is omitted). .
Also, "car" has no possibility of connecting with the next independent word,
It is rounded down (abbreviated below "n" in the figure).

The state of FIG. 5 (a) is a state in which the seventh character has been input, and the fifth character is obtained by reading the independent word analysis position calculation.
Independent words are searched for from the letters. Here, of the first bunsetsu group, those that are connected to the second bunsetsu group are output as Kanji character strings as temporary bunsetsu candidates, and those that are not are output as kana character strings (symbol “k” in the figure). "Hereafter omitted). After that, the provisional phrase candidates are not used for position calculation for reading independent words.

The state of FIG. 5 (b) is the state in which the independent word analysis position calculation read is performed, and the state of FIG. 6 (a) is the state in which the independent word analysis position calculation read is performed, which is the third here. Since the bunsetsu has been cut out, the priority of bunsetsu cost calculation etc. is calculated for the combination that connects 3 bunsetsu,
The first clause group is provisionally confirmed. The provisionally determined phrase is designated as a provisionally determined phrase (symbols in the figure below "○" are omitted) and displayed as a provisionally determined phrase thereafter separately from the others, and the previous candidate / next candidate process for this is enabled. By this processing, the one with the lower priority is discarded (the symbol "▽" in the figure is omitted).

The state of FIG. 6B is a state in which the independent word analysis position calculation read-out is performed after the provisional decision. The state of FIG. 7 is a state in which the bunsetsu is provisionally determined for the second bunsetsu group, the second bunsetsu group is provisionally determined, and the first bunsetsu group is completely determined. The current display candidate of the bunsetsu candidate is automatically output as a fixed bunsetsu (abbreviated below the symbol “⊚” in the figure), and is thereafter excluded from the kana-kanji conversion processing. In addition, by this processing, the end display candidate of the bunsetsu group is truncated (the symbol "★" in the figure is omitted).

The state of FIG. 8 shows the case where the type of the input character string has changed, and here the case where the punctuation mark "." Is input after the input of the hiragana. Similar processing is performed when the conversion key is pressed or when an alphanumeric character or the like is input following the hiragana character string. In this case, the position where different types of characters are input is regarded as the segment delimiter position, and the above-described grammar analysis and the like are performed. Even in this case, the phrase candidates are not definitely output, and the phrase candidate selection state (previous phrase / next phrase selectable state,
The symbol "☆" in the figure) is output as a definite clause when the definite processing is performed.

In the kana-kanji conversion device of the present embodiment described above, when characters different from hiragana, such as letters, numbers, symbols (including punctuation marks), katakana, etc., are input, these characters are used as delimiters for segmentation and segmentation processing. Is done,
Grammar analysis can be performed efficiently. Moreover, since the operation for converting the kanji is not required, the work efficiency can be improved without interrupting the thinking when inputting the Japanese sentence. In addition, since the clauses are sequentially determined after the input of a predetermined kana character, the objects of grammar analysis and kana-kanji conversion can be reduced, and the processing can be performed quickly. Further, since the conversion operation can be performed by the operator's intention, it is possible to correspond to the Kanji conversion of a difficult kana character.

The embodiment of the present invention has been described above.
The present invention is not limited to these examples, and it goes without saying that the present invention can be implemented in various modes without departing from the scope of the present invention.

[0032]

In the kana-kanji conversion device of the present invention, when a character element of a different type from the already input character element string is input, the different type of character element is used as a delimiter for grammatical analysis and grammar analysis or clause. The process required for Kana-to-Kanji conversion, such as the segmentation process, is performed, improving the efficiency of grammar analysis.
The conversion process can be performed accurately and quickly. Moreover, since no special conversion operation is required, it is possible to improve work efficiency without interrupting the thinking of the operator when inputting a Japanese sentence.

[Brief description of drawings]

FIG. 1 is a block diagram showing an outline of a hardware configuration of a kana-kanji conversion device according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a functional configuration of a kana-kanji conversion device according to an embodiment.

FIG. 3 is a flowchart illustrating an operation of the kana-kanji conversion device according to the embodiment.

FIG. 4 is an explanatory diagram illustrating the processing steps of the kana-kanji conversion device according to the embodiment.

FIG. 5 is an explanatory diagram illustrating a process of the kana-kanji conversion device according to the embodiment.

FIG. 6 is an explanatory diagram illustrating the process of the kana-kanji conversion device according to the embodiment.

FIG. 7 is also an explanatory diagram illustrating a process of the kana-kanji conversion device according to the embodiment.

FIG. 8 is an explanatory diagram illustrating a process of the kana-kanji conversion device according to the embodiment.

[Explanation of Codes] 21 ... CPU 22 ... ROM 23 ... RAM 24 ... Keyboard 25 ... Keyboard Interface 26 ... CRT 27 ... CRTC 28 ... Pronter 29 ... Printer Interface 30 ... Expansion Slot

Claims (1)

[Claims] 1. An input means capable of inputting at least two or more kinds of a plurality of kinds of character elements such as hiragana, katakana, letters, numbers and symbols, and a character element string inputted from the input means is kana kanji. A kana-kanji conversion device for converting a mixed sentence, wherein the character elements sequentially input from the input means are arranged and stored as unconverted character element string storage means, and the input means , When a character element of a different type from the character element of the last input is input, the position of the character element of a different type is used as a delimiter in the grammatical analysis, and the grammar analysis for the unconverted character element string and the phrase segmentation are performed. A kana-kanji conversion device comprising analysis processing means for performing analysis processing such as processing, and conversion means for performing kana-kanji conversion for the phrase obtained by the analysis.