JPH05151199A - Document preparation device - Google Patents

Abstract

PURPOSE:To improve operability by displaying clearly the strength of cogeneration between homonyms displayed as conversion candidates by difference in attribution so as to be easily confirmed by a user. CONSTITUTION:A conversion candidate storage buffer 82 stores an input character string i.e., a reading character string stored in an input character buffer 81 by referring to a ROM 9 according to the instruction of a KANA(Japanse syllabary)/KANJI(Chinese character) conversion processing part 4 and retrieving the homophone of a corresponding KANJI. Furthermore, a conversion result storage buffer 83 displays the strength of the cogeneration of the words of the homophone that is the conversion candidate stored in the conversion candidate storage buffer 82 by retrieving a cogeneration dictionary 92 in the ROM 9 by a cogenerative conversion processing part 5, and stores the connection information of the words with cogenerative relation. An attribute information setting part 10 retrieves an attribute pattern storage part 93 in the ROE 9 corresponding to the strength of the cogenerative relation, and displays it on a display 7 by changing the attribution based on the difference of strength of the cogenerative relation.

Description

Detailed Description of the Invention

 [Object of the Invention]

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document preparation apparatus having a co-occurrence function for performing efficient kana-kanji conversion by considering the semantic connection of input words.

[0002]

2. Description of the Related Art Conventionally, in a document creation device such as a Japanese word processor, several conversion candidates are usually created and displayed by the process of converting a kana character string input from a keyboard into a sentence containing kana-kanji characters, and the user can A desired word is specified from the conversion candidates by pressing a specific key (for example, operation of the select / execute key) on the keyboard.

By the way, in the document creating apparatus having the above conversion function, among the words in the sentence converted into Kana-Kanji,
There is one that considers a relationship (referred to as a co-occurrence relationship) that is easily semantically combined with the surrounding words. For example, "Natsu is hot"
When the Kana-Kanji conversion process is performed on the input character string, it is expected that there are conversion candidates such as "summer is thick", "summer is hot", and "summer is hot". In this case, the words "thick" and "hot" in the latter word are compared with the words "summer" in the first word.
Is a word that is not normally used. On the other hand, "it is hot in summer" is a relationship that is semantically easy to combine, that is, "it has a co-occurrence relationship".

[0004] A document creation device having a co-occurrence function that preferentially outputs conversion candidates having such a co-occurrence relationship is being developed year by year. , Kana readings, and co-occurrence information indicating co-occurrence words corresponding to the kana readings, are stored in the dictionary in advance, and during the kana-kanji conversion process, kana kanji characters that are usually meaningful conversion candidates are used. It is possible to efficiently create mixed sentences.

Further, in the document creating apparatus having the co-occurrence function as described above, when the input reading has various homonyms as conversion candidates, underlining, overlining, and inversion are used as a method of notifying the user of the existence. , High brightness, etc. were used.

However, regardless of the strength of the co-occurrence relation, only one type of the same attribution is used, and it is difficult for the user to understand the strength of the co-occurrence relation.

With the co-occurrence transformation, it is no exaggeration to say that the first appearing candidate is the correct answer for about 70% or more homophones appearing by the transforming process. Although the correct answer candidate has already been displayed, the user has to check all the homonyms after all, which is troublesome in operation.

[0008]

SUMMARY OF THE INVENTION In the present invention, in a kana-kanji conversion device having a co-occurrence function as described above, a user can easily confirm the strength of the co-occurrence relationship of words displayed as conversion candidates. By clearly showing the attribution by the difference in the attribution, it is possible to reduce the troublesome processing such as checking and confirming whether or not the homonym is the desired one. [Constitution of Invention]

[0009]

A kana-kanji conversion processing means for executing kana-kanji conversion by referring to dictionary information stored in a dictionary part of a reading character string input by an input device to generate and display conversion candidates. In a document creation device having
Co-occurrence dictionary based on the co-occurrence information among the conversion candidates generated by the kana-kanji conversion processing means, which stores the word information that is easily combined with the preceding and following words semantically and that is easy to combine. Co-occurrence processing means for outputting the conversion candidate of the relationship,
The present invention is characterized by having a co-occurrence display means for expressing the strength of the co-occurrence relationship by a difference in attribution display.

[0010]

In the document creating apparatus according to the present invention having the above-mentioned structure, first, the kana-kanji conversion processing means searches the kana-kanji conversion dictionary for an input phonetic character string for a homophone. Furthermore, the co-occurrence processing means selects the comrades that have a high semantic connection between the preceding and following words by searching the co-occurrence dictionary, and the co-occurrence display means that shows the difference in strength and weakness in the co-occurrence relationship by the difference in attribution display This allows users to see the relationship between the homonyms of words that are homologous to each other, so that the user can confirm the relationship, and alleviate the troublesome process of checking and confirming whether or not the homonym is the desired one. You can

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the document creation apparatus of this embodiment.

In FIG. 1, a CPU 1 controls the processing of each processing unit and controls the data. Kana-Kanji conversion processing unit 4, co-occurrence conversion processing unit 5, attribute information setting unit 1
0, control of input control unit 3, output control unit 6, RAM 8, R
Control processing such as data search of the OM 9 is performed.

On the keyboard 2 of the apparatus, character keys 21 for inputting character data such as hiragana, katakana, roman letters, numbers, various symbols, etc., and a conversion for giving an instruction for kana-kanji conversion of the inputted reading character string. A key 22 and other keys 23 for instructing cursor movement and various functions are provided.

The input control unit 3 controls in response to a character input instruction and a kana-kanji conversion instruction by pressing the keyboard 2, and stores a reading character in the input character buffer 81 of the RAM 8 corresponding to the character key 21. Control.

The display 7 shows the reading characters stored in the input character buffer 81 of the RAM 8, the conversion candidates stored in the conversion candidate storage buffer 82, and the conversion results stored in the conversion result storage buffer 83. The same sound word is displayed, and the output control unit 6 controls the character output and the symbol output on the display 7.

The Kana-Kanji conversion processing unit 4 is a ROM, which will be described later, for processing for searching for a Kanji corresponding to the input phonetic character string.
This is executed by referring to the Kana-Kanji conversion dictionary 91 of 9 to generate combination candidates (conversion candidates) consisting of Kanji and Hiragana, which are stored and held in the conversion candidate storage buffer 82 of the RAM 8 described later.

The co-occurrence conversion processing unit 5 re-converts the conversion candidates by the semantic connection between the Kanji and Kana mixed words that are the conversion candidates stored in the conversion candidate storage buffer 82 of the RAM 8 by the Kana-Kanji conversion processing unit 4. The process of constructing is performed, and the words that have a co-occurrence relationship and the strength of the co-occurrence relationship of the word comrades are added as information and stored in the conversion result storage buffer 83 of the RAM 8.

The attribute information setting unit 10 uses the RA
A method for retrieving the strength information of the co-occurrence relationship between words stored in the conversion result storage buffer 83 of M8, calling the corresponding attribute pattern from the attribute pattern storage unit 93 of the ROM 9, and storing and holding it in the conversion result storage buffer 83. Is.

The ROM 9 comprises a kana-kanji conversion dictionary 91, a co-occurrence dictionary 92, and an attribute pattern storage section 93. The kana-kanji conversion dictionary 91 has a part of speech and connection information corresponding to the entry word of the kana character string. , Already keyboard 2
The character string in the input character buffer 81 of the RAM 8 in which the character string input from is stored is necessary when converting into a kanji-kana mixed sentence according to the instruction of the kana-kanji conversion processing unit 4, and the reading character string It is a dictionary that stores the kanji corresponding to. The co-occurrence dictionary 92 is composed of word-to-word connection relationships that can be combined depending on the strength of the semantic connection based on the word and word part-of-speech information, and connection strength information, and according to the instructions of the co-occurrence conversion processing unit 5. Is called. The attribute pattern storage unit 93 uses the co-occurrence dictionary 92.
It is a memory in which several types of attribute patterns are stored, corresponding to the strength of the co-occurrence relation added between words. The strength information of the co-occurrence relationship and the attribute information are stored and held in the conversion result storage buffer of the RAM 8 according to the instruction of the attribute information setting unit 10.

The RAM 8 includes an input character buffer 81, a conversion candidate storage buffer 82, and a conversion result storage buffer 82.
The input character buffer 81 is an area for storing a read character as an input character string in response to a character key pressed by the keyboard 2 controlled by the input control unit 3. The conversion candidate storage buffer 82 corresponds to the input character string stored in the input character buffer 81, that is, the reading character string, by referring to the Kana-Kanji conversion dictionary 91 of the ROM 91 according to the instruction of the Kana-Kanji conversion processing unit 4. This is a memory area for searching and storing homophones of Chinese characters. The conversion result storage buffer 83 uses the co-occurrence conversion processing unit 5 to determine the strength of the co-occurrence relationship between the words of the same phoneme that is the conversion candidate stored in the conversion candidate storage buffer 82 by the co-occurrence dictionary 9 of the ROM 9.
2 is a memory area for searching for 2 and storing connection information between words having a co-occurrence relationship. In addition, the attribute information setting unit 10 searches the attribute pattern storage unit 93 of the ROM 9 according to the strength of the co-occurrence relationship, and the attribute pattern information is also stored. The operation of this embodiment will be described with reference to the flowcharts of FIGS. 2 to 4 and the schematic diagrams of FIGS.

First, a reading character string to be converted into a kanji / kana mixed sentence is input. It is determined whether or not the key pressed on the keyboard 2 is the character key 21 (step S1: the step is hereinafter represented by S). The pressed key is the character key 21.
If it is, the reading character input by the input control unit 3 in response to the pressing of the key is stored in the input character buffer 81 of the RAM 8 by the CPU 1 (S2). If the pressed key is not the character key 21, the CPU 1 judges whether it is the cursor movement instruction or the other key 23 for instructing various functions (S3), and performs the processing according to the key input information (S3). S4) or, again, the key input from the keyboard 2 is awaited.

Next, the input reading character string is converted into a sentence containing kanji and kana. After inputting a phonetic character to be converted from the keyboard 2, the conversion key 22 is pressed (S5)
Then, the CPU 1 activates the kana-kanji conversion processing unit 5 (S
6). When the kana-kanji conversion processing unit 5 is activated, the RAM 8
The reading character stored in the input character buffer 81 of is read out (S7). A ROM in which the kanji corresponding to the read character string later stores some homophones of the kanji corresponding to the headword using the reading character as the headword.
By searching the Kana-Kanji conversion dictionary 91 (FIG. 5) of 9 (S7), it is extracted as a conversion candidate and stored in the conversion candidate storage buffer 82 of the RAM 8 (S10). If the kanji candidate corresponding to the reading character does not exist in the kana-kanji conversion dictionary 91, the input reading character string is displayed on the display device 7 as it is according to the instruction of the input control unit 6 (S9). For example, if the key of the reading character "Natsuwa Atsui" is pressed from the keyboard 2,
The phonetic characters once stored in the input character buffer 81 are
Morphological analysis (morphological analysis is a syntax analysis method that analyzes the input sentence structure such as a solid written sentence, decomposes it into words, and adds the part-of-speech information of the words), "Natsu", It is decomposed into words such as "ha" and "atsu", and at the same time, part-of-speech information is also added. After that, each word is searched in the Kana-Kanji conversion dictionary 91, and "Natsu" is "Natsu"
The same phonetic word "kosaku", "hot" is "hot""thick"
Is stored in the conversion candidate storage unit 82.
Since "ha" is a postpositional particle, it cannot be converted to kanji.

Next, the co-occurrence information is added to the words that have been extracted as conversion candidates. First, the CPU 1 activates the co-occurrence conversion processing unit 5 and extracts one word that is a verb among the words stored in the conversion candidate storage buffer 82 (S11). Next, a forward noun is searched for (S12), a tree structure is created, and points are assigned based on the information in the co-occurrence dictionary 92 of the ROM 9 (S13, S).
14). This scoring is the strength of the co-occurrence relationship and is determined by the meaning and usage of the words. For example,
When a phonetic string "I am in the line" is entered, the Kana-Kanji conversion process

[0024]

[Table 1] Morphological analysis is performed as described above, and the conversion candidates of the words excluding the particle are stored in the conversion candidate storage buffer 82. Of these words, one word of the verbs “meet”, “fit”, and “encounter” is extracted, the co-occurrence dictionary 92 in the ROM 9 is searched, and the co-occurrence dictionary 92 is retrieved.
The co-occurrence relationship strength information stored as semantic information knowledge is added. As shown in FIG. 6, the co-occurrence dictionary 92 holds a semantic connection as a score in the relationship between the verb and the particle connecting the nouns, and this is the strength information of the co-occurrence relationship. In the above example, 1 I have an accident 2 I have an accident 3 I have a self 4 I fit myself 5 I have a self 6 I have an accident It is high.

As described above, the co-occurrence dictionary 9 of the ROM 9
The co-occurrence relationship strength information in 2 is stored in the conversion result storage buffer 83 of the RAM 8 together with the homophones that are conversion candidates for each word (S15).

Finally, the attribute pattern is displayed differently in the conversion result converted using the co-occurrence dictionary so that the strength of the co-occurrence relationship can be seen at a glance. ROM9
As shown in FIG. 7, the attribute pattern storage unit 93 of FIG. 7 stores attribute patterns such as meshes corresponding to scores according to the strength of the co-occurrence relation in the co-occurrence dictionary 92, that is, the strength of the joint relation between the words. When the conversion result is stored and displayed on the display unit, the attribute pattern corresponding to the score is also displayed. As a result, the user can recognize at a glance which word should be selected, it is possible to eliminate an important mistake in selecting a homophone, and it is possible to reduce the burden of the act of selecting a homophone. For example, when the phonetic reading "I have trained my heart so I can warm up the whole body." Is entered,

[0027]

[Table 2] A homophone conversion candidate such as is obtained, and the strength of the co-occurrence relation is the strongest as information in the co-occurrence dictionary 92.

[0028]

[Table 3] Is displayed. However, since there is no co-occurrence information such as "Heya Zentai ~, Ryozentai ~", the priority of "Zentai can be warmed up" is determined based on the frequency of use. So, the example sentence above is

[0029]

[Table 4] "Machine" and "Repair" have a high probability of being the first candidate, and "Warmed" is a word that requires attention or check for homophone selection. You can see at a glance.

[0030]

According to the present invention based on the above-described embodiment, in the kana-kanji conversion device having the co-occurrence function, the strength of the co-occurrence relationship between homonyms that are displayed as conversion candidates is
By making it clear by the difference in attribution so that the user can easily confirm it, it is possible to reduce troublesome operations such as checking whether the same phoneme is the desired one and confirming each one, and creating documents with high operability. The device can be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a document creation support device according to an embodiment of the present invention.

FIG. 2 is a flowchart for explaining the operation of the embodiment of the present invention.

FIG. 3 is a flowchart for explaining the operation of the embodiment of the present invention.

FIG. 4 is a flowchart for explaining the operation of the embodiment of the present invention.

FIG. 5 is a schematic diagram showing a structure of a kana-kanji conversion dictionary in a ROM according to an embodiment of the present invention.

FIG. 6 is a schematic diagram showing the structure of a co-occurrence dictionary in a ROM according to an embodiment of the present invention.

FIG. 7 is a schematic diagram showing a structure of an attribute pattern storage section in a ROM according to an embodiment of the present invention.

[Explanation of symbols]

 1 CPU 2 keyboard 21 character key 22 conversion key 23 other keys 3 input control unit 4 kana / kanji conversion processing unit 5 co-occurrence conversion processing unit 6 output control unit 7 display unit 8 RAM 81 input character buffer 82 conversion candidate storage buffer 83 conversion Result storage buffer 9 ROM 91 Kana-Kanji conversion dictionary 92 Co-occurrence dictionary 93 Attribute pattern storage unit

Claims (1)

[Claims] 1. A kana-kanji conversion processing means for executing kana-kanji conversion by referring to dictionary information stored in a dictionary part of a reading character string input by an input device to generate and display conversion candidates. In the document creation device, a co-occurrence dictionary that stores information about words that are easily and semantically combined with preceding and following words, and the co-occurrence among the conversion candidates generated by the kana-kanji conversion processing unit. A document creation apparatus comprising: a co-occurrence processing unit that outputs a conversion candidate of a co-occurrence relation based on information; and a co-occurrence display unit that indicates the strength of the co-occurrence relation by a difference in attribution display.