JP2941482B2 - Document creation device and kana-kanji conversion method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document preparation apparatus having a kana-kanji conversion function, and more particularly to a document preparation apparatus and a kana-kanji conversion method for realizing short-term learning focusing on the concept of a word.

[0002]

2. Description of the Related Art Conventionally, when a user selects a word,
When priority information is added to the word and the same pronunciation is input and converted the next time, the word to which the priority information is added is searched from a plurality of words (same words) corresponding to the pronunciation. Then, there is a learning function of trying to output preferentially. This is to add priority information to words (kanji headings corresponding to readings), and is called short-term learning or provisional learning. Note that long-term learning manages word frequency information, whereas short-term learning does not perform such frequency management and gives priority to words. Here, short-term learning will be described.

[0003] For example, input "Atsugi" and enter "
In the case of conversion in the order of "Atsugi", if "Atsugi" is selected, priority is given to "Atsugi" and then "Atsugi"
Is input first, "Atsugi" will be output first. However, when "Akita" is input, "Akita", which is the same place name word as "Atsugi", is not given priority, and may be converted in the order of "Tired" and "Akita".

[0004]

As described above, in the conventional method, since the words themselves are trained, words belonging to the same type cannot be simultaneously learned. For example, in the above example, even if the word "Atsugi" is selected, the learning at that time is not reflected on other place name words such as "Akita". For this reason, even if it is known that words that can be classified into “place names”, “person names”, “post titles”, etc. are frequently used in input of an address book or a novel, priority can be given only to words alone. However, it was very inconvenient in document preparation.

SUMMARY OF THE INVENTION The present invention has been made in view of the above points, and provides a document preparation apparatus and a kana-kanji conversion device capable of learning words belonging to the same type at the same time for a short period of time and using them with the highest priority.
The purpose is to provide a replacement method .

[0006]

A document creation apparatus according to the present invention adds concept information to each word registered in a conversion dictionary, and stores the priority of each concept information in a learning table. Thereby, when each word corresponding to the input reading information is searched, the searched word is output as a conversion result based on the priority of the concept information , and
When one word is selected , refer to the mode table
To learn the concept information of the selected word
Is determined, and only when the learning is possible, the priority of each of the concept information stored in the learning table is updated so that the concept information of the word has the highest priority.

[0007]

According to the above arrangement, words belonging to the same type determined by the concept information can be simultaneously learned. Therefore, even if the word has never appeared, if the word having the same type (concept) is selected first, the word can be used preferentially.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A description will now be given, with reference to the accompanying drawings, of an embodiment of the present invention.

FIG. 1 is a block diagram showing the configuration of a document creating apparatus according to one embodiment of the present invention. In FIG. 1, a keyboard 11 is a device for inputting various kinds of information necessary for document creation, and has function keys K1 to K7 and N1 as shown in FIG. K1-K
7, K1: conversion instruction key, K2: selection key,
K3: Cancel key, K4: Next candidate instruction key, K5: No conversion instruction key, K6: Katakana conversion instruction key, K7: Single Kanji conversion key. N1 is a mode setting key for instructing concept learning of the present invention.

The input control unit 12 determines a key input by the keyboard 11, and prompts the units 13, 14, and 16 to perform a process corresponding to the key. The mode setting processing unit 13 executes a mode setting process as shown in FIG. 4 and sets an instruction of “learn / not learn” for each concept information of each word in the learning mode table M1. Learning mode table M1
Stores a learning flag for each concept information as shown in FIG. This learning flag indicates “learn” when “1”, and indicates “do not learn” when “0”.

The kana-kanji conversion unit 14 converts an input hiragana character string into a kanji-kana mixed character string with reference to a kana-kanji conversion dictionary M2. Leave it to unit 15. Dictionary M2
3, as shown in FIG. 3, stores not only the corresponding kanji (converted word information), grammar information, and additional information, but also conceptual information indicating which type each word belongs to. The concept short-term learning reference unit 15 determines the output rank of each word obtained by the kana-kanji conversion unit 14 with reference to the concept learning table M3. The concept learning table M3 stores the priority of each concept information as shown in FIG.

The editing work control unit 16 controls the kana-kanji conversion unit 1
4 prompts the screen display unit 18 to display data such as the result of the conversion and the input character string on the display device 19. Here, the concept information of the selected word is sent to the concept short-term learning registration unit 17. give. Concept short term learning registration part 1
7 updates the information in the concept learning table M3 so that the concept information of the selected word has the highest priority. At that time, the concept short-term learning registration unit 17 sets the learning mode table M1.
, Only the concept information that is described as “learning” is updated. The screen display unit 18 displays an instructed character string on the screen, and performs operations such as reverse display and bold underline display. The display device 19 is a device for displaying characters and the like. Next, the operation of the embodiment will be described. here,
The three processes of (1) mode setting process, (2) reference process, and (3) registration process will be described separately. (1) Mode setting processing

As shown in the flowchart of FIG. 4, when the user operates the mode setting key N1 of the keyboard 11 (step A1), the mode setting processing unit 13 displays a mode setting screen as shown in FIG. 5 (step A2). ). On this mode setting screen, a selection item of “learn / do not learn” is provided for each concept information such as “city name”, “yamakawa name”, “town name” and the like registered in advance in the kana-kanji conversion dictionary M2. Is displayed. The user designates “learning / non-learning” for each concept information using the cursor key or the like while viewing this screen in advance. It is assumed that "learn / not learn" can be designated collectively.

If "learn" is specified (step A3), the mode setting processing unit 13 sets the learning flag of the concept information stored in the learning mode table M1 (step A4). When “do not learn” is specified (step A3), the mode setting processing unit 1
No. 3 resets the learning flag (step A5). The mode setting processing unit 13 performs this processing for all the concept information (step A6). FIG. 6 shows an example of the learning mode table M1 set as described above. In FIG. 6, the learning flag “1” means “learn”,
“0” means “do not learn”. (2) Reference processing

As shown in the flowchart of FIG. 7, for example, when reading information such as "Atsugi no Shigai" is input (step B1), the kana-kanji conversion unit 14 refers to the kana-kanji conversion dictionary M2 and executes the same. “Atsugi” and “Atsugi” are read as words corresponding to the reading “Atsugi” from the dictionary M2, and “J” and “1” are read as conceptual information of each word.
4 is temporarily stored in a buffer (not shown) (step B2). Similarly, the kana-kanji conversion unit 14 interprets the words “city”, “outside the city”, “corpse” and the concepts of the words “Q”, “Q”, and “S” in response to the reading “shigai”. The information is read and these are temporarily stored in a buffer.
FIG. 8 shows the contents of the buffer at this time.

Here, when a plurality of words are obtained for the same reading, that is, when there are homophones (step B3), the kana-kanji conversion unit 14 determines the output order of each word by using the concept short-term learning reference unit. Leave it to 15. The concept short-term learning reference section 15 refers to the concept learning table M3 and first reads out the concept information to be given the highest priority from the table M3 (step B4). FIG. 9 shows the contents of the concept learning table M3 at this time. In this example, the “city name” of the concept information “1” is obtained as the one to be given the highest priority.

In this case, the learning mode table M1
Is optional. In other words, the concept information that is “not learned” in the learning mode table M1 is interpreted as “does not learn” including the reference process here, or the new registration process is not performed. May be interpreted as only reference processing, or either may be used.

Next, the concept short-term learning reference unit 15 checks whether there is a word (conversion candidate) having the highest priority concept information (step B5). Here, "Atsugi" corresponds to it (step B6). Accordingly, the concept short-term learning reference unit 15 determines the word output order for “Atsugi” with “Atsugi” as the highest priority (step B7),
Kana-Kanji conversion unit 1 converts each word (conversion candidate) in the order of
4 (step B8).

On the other hand, there is no corresponding candidate for “Shi” (step B6). Therefore, the concept short-term learning reference unit 15 reads out the next priority concept information from the concept learning table M3 (step B9) and performs the same processing as described above. In this case, if the concept information "none" to be prioritized is read (step B10), the words (conversion candidates) are set in a preset order without rearranging the priorities.
Is output to the kana-kanji conversion unit 14 (step B8). For example, as for “Shigai”, there is no candidate that matches either the “priority name” of the next priority concept information “5” or the “first and last name” of the next priority concept information “7”. Information that the concept information is “none” is obtained. Therefore, the priority is not rearranged for
For example, they are output in a predetermined order such as “city”, “urban area”, and “corpse”. (3) Registration process

As shown in the flowchart of FIG. 10, when the user selects one of a plurality of words (same words) obtained by the above-mentioned reference processing through the selection key K2 of the keyboard 11 (step C1), an editing operation is performed. The control unit 16 provides the concept information of the selected word to the concept short term learning registration unit 17 (Step C2). The concept short term learning registration unit 17
First, with reference to the learning mode table M1, it is searched whether or not learning is possible for the concept information (step C3). If learning is possible (step C4), the concept learning table M3 is set so that the concept information has the highest priority. Update (Step C
5).

For example, it is assumed that the selection is made after conversion to "Takai-cho". The conceptual information of “Takai” is “7” (first and last name), the conceptual information of “manager” is “9” (title), and the conceptual information of “tono” is “H”. This example is shown in FIG. In this example,
Although it appears to be registered (updated) collectively, it may be registered each time one word is selected, such as registered when "Takai" is selected.

When “Takai” is selected, the concept information “7” of “Takai” is given to the concept short term learning registration unit 17. In the example of FIG. 6, the learning flag of the conceptual information “7” is “1”, indicating “learn”. Therefore, the concept short-term learning registration unit 17 updates the concept learning table M3 so that the concept information “7” has the highest priority. In this case, as shown in FIG. 9, the concept information "7" is given a priority "1". Since the priority "3" has already been given, the priority "3" is deleted. When updated, the concept learning table M3 becomes as shown in FIG.

Similarly, it is assumed that "Manager" is selected.
The concept information of “manager” is “9”. It can be seen from the learning mode table M1 of FIG. 6 that learning is not possible. FIG.
Is not updated as it is. It is also assumed that “dono” has been selected. The concept information of “dono” is “H”. Although not described in the learning mode table M1 of FIG. 6, it is assumed that learning is possible. Therefore, in this case, the concept learning table M3 in FIG. 12 is updated to be as shown in FIG. In this way, the priority of each concept information is updated. In this case, the priority order is assigned to a plurality of concepts, but only one piece of concept information to be prioritized may be stored.

As described above, words belonging to the same type determined by the concept information can be simultaneously learned for a short period of time. Therefore, a great effect is obtained when many words having concepts such as "city name", "town and village name", "first name and last name", and "person name" are used, such as when creating an address book. In addition, words such as novels that have the concept of "person name" and "first name and last name" are often used, but words that have other concepts are also used at random. If it is set to "learn" only, adverse effects can be avoided.

[0025]

As described above, according to the present invention, words belonging to the same type can be simultaneously learned for a short period of time. Therefore, 1
If a word having the same type (concept) is selected first even for a word that has never appeared, the word can be used preferentially, and the document creation efficiency is greatly improved.

Further, since the learning mode can be set arbitrarily for each type (concept), only the necessary types of words can be used preferentially. This further improves the document creation efficiency.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an overall configuration according to an embodiment of the present invention.

FIG. 2 is a diagram showing function keys provided on the keyboard of FIG. 1;

FIG. 3 is an exemplary view showing contents of a dictionary according to the embodiment;

FIG. 4 is a flowchart for explaining the operation of a mode setting process of the embodiment.

FIG. 5 is an exemplary view showing a concept learning mode setting screen of the embodiment.

FIG. 6 is an exemplary view showing contents of a learning mode table of the embodiment.

FIG. 7 is a flowchart for explaining the operation of a reference process according to the embodiment;

FIG. 8 is a view showing the contents of an internal buffer of the kana-kanji conversion unit of the embodiment.

FIG. 9 is an exemplary view showing contents of a concept learning table of the embodiment.

FIG. 10 is an exemplary flowchart for explaining the operation of the registration process of the embodiment.

FIG. 11 is a diagram showing words to be subjected to the registration processing and conceptual information thereof.

FIG. 12 is a diagram showing contents of a concept learning table updated in the registration processing.

FIG. 13 is a view showing contents of a concept learning table updated in the registration processing.

[Explanation of symbols]

11 keyboard, 12 input control unit, 13 mode setting processing unit, 14 kana-kanji conversion unit, 15 concept short-term learning reference unit, 16 edit editing control unit, 17 concept short-term learning registration unit, 18 screen display Part, 19 ... Display device, M
1: learning mode table, M2: dictionary, M3: concept learning table.

Claims (2)

(57) [Claims] 1. A dictionary for conversion which stores readings and respective words corresponding to the readings, and concept information indicating to which type each word belongs, and each of the concepts stored in the dictionary A learning table for storing the priorities of information, and whether or not learning is possible for each of the concept information stored in the dictionary.
A learning designating means for arbitrarily designating each of the concept information according to the designation by the learning designating means;
Mode table storing mode information indicating whether or not learning is possible
When a candidate search means for searching each word and the concept information corresponding to the input read information from the dictionary, with reference to the learning table, each word retrieved by said candidate search means to the priority of the concept information On the basis of
Word output means for outputting as a conversion result, and one of the words output by the word output means
Word selecting means for selecting one of the words , based on the mode information stored in the mode table.
And the concept of the word selected by the word selecting means
Learning determining means for determining whether or not learning on the information is possible; and a method for determining whether learning is possible by the learning determining means.
Only when the concept information of the word has the highest priority,
A document updating device for updating the priority of each of the concept information stored in the learning table. 2. The method according to claim 1, wherein the input reading information is associated with the reading information.
Kana-Kanji conversion method to convert to corresponding word and output
Te, each word corresponding to the reading and reading, and every time this each word
Stores the concept information indicating which type the word belongs to.
Conversion dictionary and the conceptual information stored in the dictionary.
A learning table for storing the priorities of reports, and determining whether or not learning is possible for each of the concept information stored in the dictionary.
Each concept information is specified according to the specified contents
Mode information indicating whether or not learning is possible is stored in the mode table for each
In addition, by inputting the reading information, the input reading information can be
Each corresponding word and its concept information are searched, and each word searched from the dictionary with reference to the learning table.
Convert words to conversion results based on the priority of the concept information
And when one of the output words is selected,
Based on the mode information stored in the mode table,
To learn the concept information of the selected word.
Judgment is made, and the concept information of the word is given the highest priority only when learning is possible
So that each of the above outlines stored in the learning table is
Kana-Kanji, characterized by updating the priority of information
Conversion method.