JP3353947B2 - Word selection method in sentence creation device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a word selecting method in a text preparation device. In recent years, the spread of Japanese document creation devices (Japanese word processors) has been remarkable. The basic function of the Japanese document creation device is to convert the input kana into kanji, but it is expected that the scene will be used as a hobby in the coming aging society. It is desired to be more versatile.

[0002]

2. Description of the Related Art In a conventional Japanese document creating apparatus, in addition to a function of converting an input word (kana character) as it is into a kanji, a plurality of words corresponding to the input word (kana character or converted kanji) are added. (Japanese Patent Application Laid-Open No. Sho 60-129875) having a function of displaying and selecting a seasonal word has been proposed.
issue).

According to this conventional word selection method, words such as "candy", "cloud" and "kion" are registered in advance together with the months in which similar words are used,
When a word is input, seasonal matches that are registered in response to the word are displayed on the screen in order, so that the operator can select an appropriate one from among them.

[0004]

However, in the above-described conventional word selection method, since only a specific word is selected from words (season words) registered in advance, a specific word is not selected. The spread is small and words with different nuances related to the original word cannot be selected from a wide vocabulary range.

[0005] For example, when creating a haiku or a waka, there is a case where a user wants to find a word that comes up with a certain word but has a slightly different nuance. In such a case, the conventional word selection method simply searches for a synonym for the input word,
This does not mean that the meaning of a word is grasped and a concept word associated with the word is searched for, or a word related to them is searched for from a plurality of words. For this reason, even when a word that has been conceived or a searched word is not considered to be the optimal word, the user often has to use the word reluctantly.

SUMMARY OF THE INVENTION In view of the above problems, the present invention enables a word related to an input original word to be selected from a wide range, and a word selection in a sentence creating apparatus capable of selecting a word closer to an operator's idea. The aim is to provide a method.

[0007]

According to a first aspect of the present invention, there is provided a method for solving the above-mentioned problems, as shown in FIG.
This is a word selection method in the text creation device 1 for creating a text by inputting a word WD1, and displays one or a plurality of words WD2 having the same concept as the input word WD1 and displays the same. Word WD with concept
2 displays one or more words WD3 having a concept close to the specified word when one of the two is specified;
When one of the words WD3 having the adjacent concept is designated, the one having the concept associated with the designated word is designated.
One or more words WD4 are displayed, and a required word is selected from any of the displayed words WD2, WD3, and WD4.

According to the method of the present invention, one or a plurality of words WDA3 and WDB3 having a concept associated with each of the plurality of input words WDA1 and WDB1 are searched, and each of the searched words WDA3 and WDB3 is searched. One or more words WDA4, WDB having a concept close to WDB3
4 and the searched words WDA3, WDB3, WD
Words WDD3, WDD4 duplicated in A4, WDB4
Is displayed, and a required word is selected from the displayed words WDD3 and WDD4.

According to a third aspect of the present invention, there is provided a method comprising the same concept dictionary 23, the proximity concept dictionary 24, and the associative concept dictionary 25, each of which includes a plurality of words. Is associated with a word having the same concept for each word and a word in the close concept dictionary 24 having a concept close to each word, and includes a word included in the close concept dictionary 24 Is associated with a word in the same concept dictionary 23 having a concept close to each word and a word in the associative concept dictionary 25 having a concept associated with each word. The words included in the concept dictionary 25 are associated with words in the proximity concept dictionary 24 having a concept associated with each word, and a plurality of input words WDA are included.
1, the associative concept dictionary 25 corresponding to WDB1 respectively.
Words WDA3 and WDB3 associated with the word
Is searched from the proximity concept dictionary 24, the words WDA4, WDB4 associated with the words WDA3, WDB3 searched from the proximity concept dictionary 24 are searched from the same concept dictionary 23, and among the searched words, Displays the duplicate words WDD3 and WDD4, and displays the displayed word WDD.
3, a word selection method characterized by selecting a required word from WDD4.

According to a fourth aspect of the present invention, each word included in the same concept dictionary 23, the close concept dictionary 24, and the associative concept dictionary 25 is any one of a Japanese word, a Chinese word, and a foreign word. This is a word selection method characterized by being classified into types and performing a search according to the type.

According to a fifth aspect of the invention, there is provided a first dictionary 23, a second dictionary 24, each comprising a plurality of words.
And the first dictionary 23, the second dictionary 25,
The words included in the dictionary 24 and the third dictionary 25 are associated with other words having a predetermined related concept for each word, and a plurality of input words WDA1 and WDB1 are input.
The words WDA3 and WDB3 associated with the words of the third dictionary 25 respectively corresponding to
4 and the word WDA associated with each of the words WDA3 and WDB3 searched from the second dictionary 24.
4, searching the first dictionary 23 for the WDB4, displaying the overlapping words WDD3 and WDD4 among the searched words, and selecting a necessary word from the displayed words WDD3 and WDD4. This is the method of selecting words.

[0012]

With the input word as a keyword, a word having a predetermined related concept is searched for and displayed. When a specific word is specified from the displayed words, a word having a predetermined related concept is searched for using the specified word as a keyword, and is displayed. In this way, words whose concept is broadened while being related to the input word are displayed, and a suitable word is selected from them.

When a plurality of words are input, a word having a predetermined related concept for each word and a word having a predetermined related concept for each word are searched. Duplicate words are displayed among the searched words, and a word that is appropriate is selected from them.

[0014]

FIG. 1 is a block diagram of a Japanese document creating apparatus 1 according to the present invention, and FIG. 2 is a diagram showing a configuration of each word data of a dictionary provided in the Japanese document creating apparatus 1.

The Japanese document creating apparatus 1 is composed of a processing unit 11 including an MPU (microprocessor unit) and a memory, an input unit 12 including a keyboard, a display unit 13 having a screen, and the like.

The processing section 11 is provided with an input stack 21, an output stack 22, an identical concept dictionary 23, a close concept dictionary 24, and an associative concept dictionary 25 on its memory.

The input stack 21 stores one or a plurality of words (keywords) input from the input unit 12. The output stack 22 stores words searched for in relation to the words stored in the input stack 21, and the contents are displayed on the display unit 13.

In the same concept dictionary 23 and the near concept dictionary 24, and in the near concept dictionary 24 and the associative concept dictionary 25, mutually related words are associated with each other by pointers. In the same concept dictionary 23, words having the same concept are associated with each other by a pointer.

More specifically, words included in the same concept dictionary 23 include words in the same concept dictionary 23 that have the same concept for each word, and words in the same concept dictionary 23 that have a concept that is close to each word. It is associated with a word in the concept dictionary 24.

That is, as shown in FIG. 2A, word data 2 for each word contained in the same concept dictionary 23
3W is composed of the word WD, the type KD, the number of pointers NP and pointers P for words in the same concept dictionary 23, and the number of pointers NP and pointers (forward pointers) P for words in the close concept dictionary 24.

Here, the type KD indicates one of a Japanese word, a Chinese word, and a foreign word, and each word WD is classified into one of these types KD. In addition, the words included in the proximity concept dictionary 24 are associated with words in the same concept dictionary 23 having a concept close to each word and words in the association concept dictionary 25 having a concept associated with each word. Have been.

That is, as shown in FIG. 2B, the word data 2 for each word contained in the proximity concept dictionary 24
4W is the word WD, the type KD, the number of pointers NP for words in the same concept dictionary 23, and the pointer (reverse pointer)
P, the number of pointers NP to words in the associative concept dictionary 25
And a pointer (forward pointer) P.

Further, the words included in the associative concept dictionary 25 are the adjacent concept dictionaries 2 having the concept associated with each word.
4 is associated with the word. That is, FIG.
As shown in (c), the word data 25W for each word included in the associative concept dictionary 25 includes a word WD, a type K
D, number of pointers NP to words in proximity concept dictionary 24
And a pointer (reverse pointer) P.

These pointers P are preset based on the meaning or concept of each word. Further, the words in the same concept dictionary 23 or the associative concept dictionary 25 are associated with the words stored in the input stack 21.

The modes for selecting a related word from the input words include an enlargement mode and a convergence mode.
In each mode, the search range is set to the type KD.
Either “type search” for each type or “mixed search” for all types KD can be specified.
When "type search" is designated, the type of the word is automatically identified by the type KD attached to the input word.

First, the enlargement mode will be described. FIG.
Is a diagram showing an outline of a word selection method in the enlargement mode, and FIG. 5 is a flowchart showing a flow of processing or operation in the enlargement mode.

First, the word WD1 is input as a keyword, and a search type is input (# 11). Input word W
A word corresponding to D1 is searched from the same concept dictionary 23,
A word (word group) WD2 in the same concept dictionary 23 specified by the pointer P of the searched word (word WD1) is searched (# 12) and displayed on the screen of the display unit 13 (# 14). . If there is an appropriate word in the displayed word WD2 (Yes in # 15), by selecting it (# 16), it is taken into the text being created.

If there is no appropriate word in the displayed word WD2 (No in # 15), a word that is considered to be closest to the idea of the operator is designated, and the search is further performed using that word as a keyword. (# 17). As a result, the word WD3 in the proximity concept dictionary 24 specified by the pointer P of the word specified as the keyword is searched (# 1).
8), which is displayed on the screen (# 20).

If there is an appropriate word among the displayed words WD3, it is selected (# 21, # 22). If there is no appropriate word in the displayed words WD3, one of the words is designated as a keyword (# 23), and the word WD4 in the associative concept dictionary 25 designated by the pointer P of the word is searched. (# 24), and it is displayed on the screen (# 25). An optimum word is selected from the displayed words WD4 (# 26).

Thus, the concept of the input word WD1 is sequentially expanded to the word WD2, the word WD3, and the word WD4, and the image is expanded to expand the range of word selection.
In addition, the idea can be expanded by obtaining a new suggestion, and as a result, the optimal word can be selected.

Next, the convergence mode will be described. FIG.
Is a diagram showing an outline of a word selection method in the convergence mode, and FIG. 6 is a flowchart showing a flow of processing or operation in the convergence mode.

First, a plurality of words WDA1 and WDB1 are input as keywords, and a search type is input (# 3).
1). Words respectively corresponding to the input words WDA1 and WDB1 are searched from the associative concept dictionary 25, and the words WDA3 and WDB3 in the proximity concept dictionary 24 designated by the pointer P of the searched words WDA2 and WDB2 are searched. The word WDA in the same concept dictionary 23 designated by the pointer P of these words WDA3 and WDB3
4, WDB 4 is searched (# 32, 33).

When there is a certain number or more of the same words (sometimes referred to as "duplicate words") in the searched words WD, the duplicate words are displayed on the screen (# 34). In extracting a duplicate word, the words WDA3 and WDB3 searched from the proximity concept dictionary 24
And words WDA4, W retrieved from the same concept dictionary 23.
When there are two or more words that are identical to each other in DB4, they are regarded as duplicate words. Further, the words WDA3 and WDA4 and the words WDB3 and WD
B4 or all words WDA3, WDA4, W
The determination may be made based on whether or not DB3 and WDB4 overlap each other. In addition, three or more may be used instead of two or more.

The operator can select an optimum word from the words WD displayed on the screen (# 35) and take it into the text being created. As described above, in the convergence mode, a plurality of words W related to the concept of the word desired by the operator are set.
By inputting DA and WDB as keywords, words most relevant to them are narrowed down and displayed, and an optimum word can be selected from them. Therefore, the potential vocabulary of the operator is expanded and maximized by the word selection method in the convergence mode.

Next, specific examples in each mode will be described. Here, it is assumed that the operator searches for an appropriate word when creating a haiku or a waka. First, the case of the enlargement mode will be described.

The operator feels that the coming of spring is thrilled by the sun that is getting stronger and stronger, and at this time he wants to find a suitable word (word) that expresses the sunshine in a cheerful mood. Therefore, "sunlight" is input as the keyword word WD1, and "mixed search" is input as the search type.

The next word WD2 is retrieved from the same concept dictionary 23 and displayed on the screen. a. Sunshine b. Sunlight c. Nikko d. Sunlight e. Since there is no optimal word among these, the operator selects "sunlight" as the word that best matches the image, and proceeds to the next search using this as a keyword. Proximity concept dictionary 2
4 is searched for the next word WD3 and displayed on the screen.

A. Kosatsu weather b. Harumitsu c. Sun
d. Yang e. Sunshine Here, the operator decides that "Shunmitsu" is the most suitable one for his own state of mind and uses this word. Further, when "spring light" is selected to obtain a keyword for enlarging the image, the next word WD4 is searched from the association concept dictionary 25 and displayed on the screen.

A. Warm b. Nodular c. Beautiful
d. Spring breeze e. Spring color f. Spring Dawn g. Spring suburbs h. Sakura i.
Admission j. Ugui k. Hibernation l. Spring thunder m. Spring manure n. Spring warm o. Equinox The operator decided to use the word “uraraka” from this, and created the following haiku.

"Increase by one or more in spring light" As described above, according to the Japanese document creating apparatus 1, it is possible to support the selection of terms from a wide range in consideration of the degree of approach of the meaning of words. .

Next, the case of the convergence mode will be described.
The operator feels the vitality of the creatures in the forest when he goes up to Mt. Natsu, and wants to find a suitable word for this. Therefore, “creature” and “life force” are input as the keyword words WDA1 and WDB1, and “mixed search” is input as the search type.

For "creature", the next words WDA3 and WDA4 are searched from the proximity concept dictionary 24 and the same concept dictionary 23. Proximity concept word group (WDA3) a. Plants b. Green c. Life d. Life e. Lifetime f.
Beating g. Animal h. Breathing i. Breath j. Embryonic movements: Same concept word group (WDA4) a. Branch a. Leaves c. Root d. Generation e. Reincarnation f. Pulsation g. Beast. Sigh. Breathing Wriggling: The next words WDB3 and WDB4 are searched from the proximity concept dictionary 24 and the same concept dictionary 23 for "life force".

Proximity Concept Words (WDB3) a. Genetic b. Gene c. Egg d. Longevity e. prosperity
f. Beating g. Recovery h. Playback i. Unchanged j. Blood ... Same concept word group (WDB4) a. Tradition a. Transmission c. Nuclear d. Elderly e. Proliferation
F. Beating key. Constant Pulsation Red blood cells Tradition
...... The Japanese document creation device 1 searches for these words WDA
From 3, WDA4, WDB3, and WDB4, the word “pulsation” of the group of adjacent concept words and the word “pulsation” of the same concept word group are overlapped by the word WD.
D3 and WDD4 are extracted and displayed on the screen.

The operator has selected "pulsation" as the word that best fits his image. According to the above embodiment,
It is possible to select not only synonyms or synonyms as in the past, but also from a group of words with one flow, expand the vocabulary of the operator, select words with rich ideas, and more suitable You can select a word. In the convergence mode, you can enter multiple words that you can think of and select words that are closely related to them.
You can easily select and use words you do not think of.

In the above embodiment, the same concept dictionary 2
3. It has been described that the proximity concept dictionary 24 and the associative concept dictionary 25 are separately and independently configured. However, in a single large dictionary, the pointers P to the same concept word, the proximity concept word, and the associative concept word are respectively set. These dictionaries may be substantially formed by registering words. In addition, the configuration of the word data 23W, 24W, and 25W, the content or order of the flowchart, the configuration of the whole or each part of the Japanese document creation device 1, and the like can be variously changed according to the gist of the present invention. INDUSTRIAL APPLICABILITY The present invention is applicable to a sentence creation device other than Japanese, and is also applicable to a sentence creation device having various configurations and functions.

[0046]

According to the present invention, a word related to the input original word can be selected from a wide range, and a word closer to the idea of the operator can be selected.

According to the second to fifth aspects of the present invention, by inputting a plurality of words that come to mind, words that are closely related to them can be selected, and words that cannot be thought of easily can be easily selected. .

[Brief description of the drawings]

FIG. 1 is a block diagram of a Japanese document creation device according to the present invention.

FIG. 2 is a diagram showing a configuration of each word data of a dictionary provided in the Japanese document creation device.

FIG. 3 is a diagram illustrating an outline of a word selection method in an enlargement mode.

FIG. 4 is a diagram showing an outline of a word selection method in a convergence mode.

FIG. 5 is a flowchart illustrating a flow of processing or operation in an enlargement mode.

FIG. 6 is a flowchart illustrating a flow of a process or an operation in a convergence mode.

[Explanation of symbols]

 1 Japanese document creation device (sentence creation device) 23 Same concept dictionary (first dictionary) 24 Proximity concept dictionary (second dictionary) 25 Associative concept dictionary (third dictionary) WD, WD1-4 Words WDA1-4, WDB1 4 words WDD3-4 Duplicate words (duplicate words)

Claims (5)

(57) [Claims] 1. A word selecting method for a sentence creating apparatus (1) for creating a sentence by inputting a word (WD1), wherein the word selecting method has the same concept as the input word (WD1). One or more words (WD2) are displayed, and when one of the words (WD2) having the same concept is designated, one or more words (WD3) having a concept close to the designated word are displayed. When one of the words (WD3) having the adjacent concept is designated, one or more words (WD4) having the concept associated with the designated word are displayed. Word (WD2) (WD3) (WD
4) A necessary word is selected from among the above, a word selecting method in a sentence creating apparatus. 2. A method for selecting a word in a sentence creating apparatus (1) for creating a sentence by inputting a word (WDA1, WDB1), the associating from a plurality of input words (WDA1, WDB1). One or more words (WDA3,
WDB3), and one or more words (WDA4, WD) having a concept close to each searched word (WDA3, WDB3).
B4), and the searched words (WDA3, WDB3, WDA4, WD)
B4), displaying a duplicate word (WDD3, WDD4), and selecting a required word from the displayed words (WDD3, WDD4). 3. A word selection method in a text preparation device (1) for preparing a text by inputting a word (WDA1, WDB1), the same concept dictionary (2) each including a plurality of words.
3), the proximity concept dictionary (24), and the associative concept dictionary (2)
5), wherein the words included in the same concept dictionary (23) are words having the same concept for each word, and the close concept dictionary (24) having a concept close to each word. The words included in the close concept dictionary (24) are associated with words in the same concept dictionary (23).
Are associated with words in the associative concept dictionary (25) having a concept associated with each word, and words included in the associative concept dictionary (25) are associated with each word. Associated with a word in the proximity concept dictionary (24) having a concept to be input, and a word in the associative concept dictionary (25) corresponding to each of the plurality of input words (WDA1, WDB1). An associated word (WDA3, WDB3) is searched from the proximity concept dictionary (24), and each word (WD) searched from the proximity concept dictionary (24)
A3, WDB3) (WDA)
4, WDB4) from the same concept dictionary (23), and duplicate words (WDD3, WDD) among the searched words.
4) displaying, and selecting a required word from the displayed words (WDD3, WDD4). 4. Each word included in the same concept dictionary (23), the close concept dictionary (24), and the associative concept dictionary (25) is classified into one of a Japanese language, a Chinese language, and a foreign language. 4. The method according to claim 3, wherein the search is performed in accordance with the type of the document. 5. A word selection method in a text preparation device (1) for preparing a text by inputting a word (WDA1, WDB1), wherein the first dictionary (23) includes a plurality of words. ),
It has a second dictionary (24) and a third dictionary (25), and the words included in each of the first dictionary (23), the second dictionary (24), and the third dictionary (25) are The word (WDA1, WDB1) is associated with another word having a predetermined related concept, and the word (WDA1, WDB1) is associated with the word of the third dictionary (25). WDA3, WDB3) from the second dictionary (24), and each word (WDA) searched from the second dictionary (24).
3, WDB3) (WDA)
4, WDB4) from the first dictionary (23), and duplicate words (WDD3, WDD) among the searched words.
4) displaying, and selecting a required word from the displayed words (WDD3, WDD4).