JPH11175518A - Character string inputting device and method therefor and program recording medium for storing character string input program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow a user to almost exactly confirm input miss, and to easily correct an inputted character string. SOLUTION: A control part 3 allows a user to confirm input miss by displaying a synonym as a candidate word when the synonym of a character string similar to a character string inputted from an inputting part 2 has a cooccurrence relation with a word which is already used in a document which is being prepared. Therefore, even when the inputted character string is a character string which exists as a word, the user is allowed to confirm the input miss when the possibility of the input miss is high. Thus, it is possible to allow the user to almost exactly confirm the input miss, and to prevent the efficiency of document preparation from being deteriorated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention provides a method for allowing a user to confirm whether a character string entered by a user is not a character string of a word intended by the user and is likely to be an input error, The present invention relates to a character string input device, a character string input method, and a program recording medium that records a character string input program for correcting a character string input when an input error occurs.

[0002]

2. Description of the Related Art Conventionally, it has been determined whether a character string input by a user is a proper character string as a word,
If the character string is not appropriate as a word, a word (similar word) of a character string similar to the input character string is extracted as a candidate word, and the candidate word is displayed to allow the user to recognize an input error ( There is a character string input device. In this type of character string input device, when the user selects a word intended by the user from the displayed candidate words, the input character string is replaced with the character string of the word selected here, and Mistakes can be easily corrected.

[0003] For example, in an English word processor to which this type of character string input device is applied, an English-Japanese dictionary (which is dictionary data converted into electronic data) is searched to determine whether a word of a character string input by a user exists. If there is no English word of the character string, it is determined that there is an input error (spelling error), and a word of a character string similar to the input character string is extracted from the English-Japanese dictionary. By displaying the words as words, the user was made to recognize input errors. Further, when the user selects a word intended by the user from the displayed candidate words, the input character string is replaced with the character string of the selected candidate word, so that input errors can be easily corrected. Was. There is also a dictionary in which the English-Japanese dictionary is replaced with a dictionary that does not store the meaning of Japanese in correspondence with each word.

Further, when displaying candidate words, there is also a display that displays the meaning of each candidate word at the same time, so that the user can easily determine the intended word from the candidate words. .

[0005]

However, the conventional character string input device determines whether or not an input error has been made by the user based on whether or not a word having the same character string as the input character string exists. doing. For example, in an English word processor, if a word having the same character string as the character string input by the user exists in the English-Japanese dictionary, the input character string is determined to be a correct word. Has been determined.) Therefore, even if the character string input by the user is not the word (character string) intended by the user (even if the input is incorrect), if the word of the input character string exists in the English-Japanese dictionary, There is a problem that the determined character string is determined to be a correct word (not an input error). For example, if the word intended by the user is "light"
, Even if the user erroneously inputs "right", the input character string is determined to be a correct word. In particular, in the case of Japanese, "l" and "r" are often mistakenly input because the pronunciation is similar. In such a case, the user cannot recognize the input error and cannot correct it.

An object of the present invention is not only to determine whether or not a word having the same character string as the input character string exists, but also to determine whether or not a word is already input in a document being created. By determining whether or not a character string is a word intended by the user, a character string that can more reliably detect a user's input error, make the user recognize this, and correct the input character string Input device, character string input method,
Another object of the present invention is to provide a program recording medium on which a character string input program is recorded.

[0007] Another object of the present invention is to provide a method in which an input character string is a preset infrequently used word.
A character string input device, a character string input method, and a program recording medium on which a character string input program is recorded so that the user recognizes the input error and corrects the input character string by making the user confirm whether or not there is an input error. To provide.

[0008]

According to a first aspect of the present invention, there is provided an input unit for inputting a character string, and a similar word, a synonym, antonym, etc., which are set in advance for each word as a similar character string. A storage unit for directly or indirectly storing the co-occurrence word set in advance as a related idea, and the presence or absence of a co-occurrence state, and when a similar word of the input word is present in the storage unit, the similar word And a control unit for setting the co-occurrence state of the co-occurred word of the determined word when the input word is determined, on the display unit, when the similar word is in the co-occurrence state.

First, the co-occurrence referred to in the present invention is a relation between words that are related to meanings such as synonyms and antonyms and are highly likely to be used simultaneously in the same document. Note that whether or not the meaning is related between words is determined by the dictionary creator based on his / her own image, and the co-occurrence word is set for each word by the dictionary creator. For example, depending on the dictionary creator, "weight (weight)" and "light (light)" or "light"
(Light) ”and“ color (color) ”are set in advance as co-occurrence words that are related (in a co-occurrence relationship) with a semantic idea. Note that the dictionary creator changed “light” to “weigh”.
t "and" color "are set as co-occurring words in a co-occurrence relationship, but" weight "and" color "
or "is not in a co-occurrence relationship (" weight "
And "color" are not related to a semantic idea), and these words may not be set as co-occurring words. In addition, the co-occurrence state refers to a state in which a co-occurrence word related to the above-described semantic concept is more likely to be input compared to other words.

[0010] For the similar words, the dictionary creator sets in advance, for each word, a word whose character string is similar in its own image.

In the character string input device having this configuration, the presence / absence of a co-occurrence state is stored for each word, and when the input word is determined, the co-occurrence word of the determined word (the possibility of being input as compared to other words is higher). The co-occurrence state is set for the word (elevated). In addition, when there is a similar word of the input word, if the similar word is in a co-occurrence state (if it is more likely to be input than other words), the similar word is displayed as an input candidate. .

Here, even if a word of a character string input by the user exists, if a similar word of the input word is in a co-occurrence state in which there is a high possibility that the word is input compared to other words, There is a high possibility that the word intended by the user is this similar word (it is highly likely that the user has made an input error).
On the other hand, if the similar word of the input word is not in a co-occurrence state in which there is a high possibility that the word is input compared to other words, it is unlikely that the word intended by the user is a similar word (user input It is likely that it is not a mistake.) When there is a high possibility that the user has made an input error, the input candidate is displayed to allow the user to confirm whether or not the input error has occurred, so that the user can almost certainly recognize the input error.

[0013] Incidentally, if there is a similar word of the input word, it is conceivable to display the similar word to make the user confirm whether there is an input error. However, in this method, there is a possibility that the above-mentioned user input error is not made. Even when the price is high, input candidates are displayed and the user is asked to confirm whether there is any input error, and the user's document creation work is frequently interrupted, thereby causing a problem of lowering document creation efficiency.
However, in the present invention, the input problem is displayed only when there is a high possibility that the user has made an input error, and the user is made to check whether there is an input error. Therefore, the above problem does not occur.

In the invention described in claim 2, if the input word itself is already in a co-occurrence state, the input word is determined without displaying the similar word on the display unit.

In this configuration, when the word of the input character string is a word that is more likely to be input than other words, there is almost no possibility of a user input error. The word is determined.

In the invention described in claim 3, if a similar word similar to the co-occurred word of the determined word is in the co-occurrence state, the co-occurrence state of the similar word is reset, so that there is a similar relationship. A plurality of words do not co-occur. By the way, when a plurality of words having a similar relationship are in a co-occurrence state at the same time, it cannot be properly determined whether or not the input word is highly likely to be an input error. For example, first, the word “weigh” co-occurring with “light”
t ”is input, and then the word“ word ”co-occurring with“ right ”is input. When“ right ”is input at the end,“ right ”co-occurring with“ word ”is input in most cases. It is considered that
ht "is a co-occurrence state, causing the user to check for an input error, thereby causing a problem of interrupting the user's document creation work.

On the other hand, if the input word is in a co-occurrence state, the above problem can be solved by adopting a structure for determining the word, but the word "co-occurrence" in "light" first is solved. weight ”is input, and then“ rig ”
When the word "word" co-occurring with "ht" is input, and "light" is input at the end, "light" is a co-occurrence of "right" even though "right" is an input error of "right" as described above. Since the state is in the state, the input is determined, and there is a problem that the user cannot recognize the input error.

In this configuration, a plurality of words among the similar words are prohibited from co-occurring at the same time by the above configuration, thereby preventing the above-mentioned problem from occurring.

According to a fifth aspect of the present invention, there is provided an input unit for inputting a character string, a similar word set in advance as a similar character string for each word, and a word set in a predetermined low frequency of use. A storage unit for storing information indicating whether or not the input word is a word with a low use frequency, determining whether or not the input word is a word with a low use frequency;
A control unit for displaying a similar word of the input word on a display unit as an input candidate.

In this configuration, when a word set in advance as a word with low frequency of use (a word that is rarely used) is input, a similar word is displayed to determine whether the user has made an input error. Ask for confirmation. Here, when a word that is rarely used by a general user (a word that is rarely used) is input, the word intended by the user may be a word having a character string similar to the input character string. High in nature. In other words, it can be said that there is a high possibility that the user made an input error. In such a case, the user is made to confirm the input error, so that the input error of the user can be detected almost certainly, and the user can recognize the input error.

The invention described in claim 4 is a structure obtained by adding the structure of the invention described in claim 5 to the structure of the invention described in claim 1, 2 or 3.

According to a sixth aspect of the present invention, there is provided a means for inputting information on whether or not the frequency of use is low.

Normally, the frequency of use of a word differs depending on the user. For example, even if a word is rarely used by a general user, if the word is frequently used by a specific user, an input error is confirmed each time the word is input to the specific user. This lowers the document creation efficiency of the user.

In this configuration, the user can set for each word whether or not the word is an infrequently used word. Therefore, the above problem does not occur, and the user's document creation efficiency can be reduced. Absent.

[0025]

FIG. 1 is a block diagram showing a configuration of a character string input device according to an embodiment of the present invention. The character string input device 1 of this embodiment includes an input unit 2 for inputting a character string (English word), a control unit 3 for executing a process to be described later on the input character string, and an English word (hereinafter simply referred to as “English word”). A storage unit 4 storing an English-Japanese dictionary 4a in which character strings and their meanings are associated with each other and a similar word table 4b in which information indicating similar words similar to each other in the character strings are registered. A display unit 5 for displaying an input character string and the like. Reference numeral 6 denotes a program recording medium, such as a CD-ROM or a floppy disk, on which a program for causing a later-described process to be executed by the control unit 3 is recorded. The program is supplied from the program recording medium 6 to the control unit 3 in advance. In addition, RO which recorded this program in control part 3
M may be provided, and a program may be supplied from the ROM to the control unit 3.

FIG. 2 is a diagram showing the structure of the English-Japanese dictionary stored in the storage unit. A word record 11 is registered for each word in the English-Japanese dictionary 4a. The word record 11 has a dictionary i indicating the position of the record in the English-Japanese dictionary 4a.
ndex, the character string (spelling) of the word, the meaning of the word, the meaning of the meaning, the position of the word record 11 of the co-occurring word that is likely to be used simultaneously in the same document and is likely to be used simultaneously. A co-occurrence word dictionary index and a synonym table index indicating the position of a synonym record of the word registered in a synonym table 4b to be described later are registered. Note that words without co-occurring words are described in the co-occurring word dictionary ind
ex is registered with -1. For example, the example shown in FIG.
This indicates that “weight” and “light” are in a co-occurrence relationship, and that “light” and “color” are in a co-occurrence relationship. In this example, "light"
Is co-occurring with both "weight" and "color", but "weight" and "color" are not. As described above, even when the word A has a co-occurrence relationship with the words B and C, the word B and the word C may not have a co-occurrence relationship. This is because even if the word A is related to the words B and C in a semantic sense, the words B and C
Is not always relevant. It should be noted that whether semantic ideas are relevant
As described above, the dictionary is determined based on its own image, and co-occurrence words are set in advance for each word. Also, "lic
"e" is an example in which no co-occurrence word exists. For words having no similar words, -1 is registered in the similar word table index. FIG. 2 shows that the similar word record of “weight” is not registered in the similar word table 4b (see FIG. 3) (indicating that there is no similar word).

FIG. 3 is a diagram showing the structure of the similar word table. In the similar word table 4b, a similar word record 15 of a word having a similar word is registered (a similar word record 15 for a word having no similar word is not registered). The synonym record 15 includes a table index indicating the position of the record in the synonym table 4b, a character string of the word, a dictionary index indicating the position of the word record 11 in the English-Japanese dictionary 4a of the word, and the word in the document being created. A co-occurrence flag indicating whether or not the word is likely to be used, a display flag indicating whether the word is infrequently used, and
A synonym index indicating the position of a synonym record 15 of a word having a similar character string is registered. Not all the similar words are registered in each post-similar record 15, but the similar words are related in a so-called chain structure as shown by an arrow in FIG. Link information. That is, as shown by the arrows in the figure, “fight”, “light”,
If the four words “night” and “right” are similar words (there is no other similar word), “fight”
The table index indicating the synonym record 15 of “light” is registered in the synonym record 15 of “t”, and “li” is registered.
The table index indicating the synonym record 15 of “night” is registered in the synonym record 15 of “ght”,
"Right" is added to the synonym record 15 of "night"
A table index indicating the synonym record 15 of “right” is registered, and “fight” is registered in the synonym record 15 of “right”.
By registering a table index indicating the similar word record 15 of "t", these similar words are associated in a chain structure.

In this embodiment, the English-Japanese dictionary 4a and the similar word table 4b are divided into two files, and the meaning, similar words, co-occurrence words, etc. of each word are indirectly stored. Into a single file, the meaning, similar words, co-occurrence words, etc. of each word may be directly stored. Here, when the English-Japanese dictionary 4a and the synonym table 4b are stored in the hard disk, the seek time for reading the data of the English-Japanese dictionary 4a and the synonym table 4b is reduced, and the access can be performed at high speed. Is divided into Although the similar words are associated in a chain structure, all of the similar word table indexes may be registered in the similar word index of each word.

Hereinafter, the operation of the character string input device according to this embodiment will be described in detail. 4 to 6 are flowcharts showing the processing of the character string input device according to this embodiment. When the user starts creating a new document, the control unit 3 turns off the co-occurrence flag 4b in all the similar word records 15 (sets it to 0). Further, in this embodiment, the word whose display flag is on (is 1) is a word that is used less frequently. The display flag can be arbitrarily set or reset by the user, and the user can set for each word whether or not the word is a less frequently used word according to the use environment. The user operates the input unit 2 to input a character string (English characters) and the like to create a document.

When there is an input by a key operation or the like on the input unit 2 (n1), the control unit 3 determines whether the operated key is a character key, a space key, or another key (n2, n3). . If the input key is a character key, an input buffer (not shown) provided in the control unit 3
Is added (n4). The character string in the input buffer is displayed on the display unit 5, and the character string input by the user can be confirmed by this display. Also, in English, there is a rule that a space is provided between words, so if the operated key is the space key, it is determined that one word is input, and that the input word is n6 and later described later. Execute the process. When another key other than the character key and the space key, such as a line feed key, is operated, a process (line feed process, etc.) corresponding to the operated key is performed (n5).

When proceeding to n6, the control section 3 searches whether or not a synonym record 15 of a word (matching word) having the same character string as the input character string is registered in the synonym table 4b. If there is no corresponding synonym record 15 in this search, the process proceeds from n7 to n8 to search for a word (matching word) having the same character string as the character string input to the English-Japanese dictionary 4a. If there is a word to be executed, the input character string (word) is determined (n9 → n15), and the process proceeds to n16.

For example, when the user inputs a character string “weight”, a similar word record 15 of a word having the same character string as this character string is not registered in the similar word table 4a (in this embodiment, , "Weigh
It is assumed that a word having a character string similar to “t” does not exist. ), The process proceeds from n7 to n8. Also, since a word consisting of the same character string as this character string is registered in the English-Japanese dictionary 4a, the character string "we
"right".

On the other hand, if it is determined at n7 that there is a matching word, the process jumps to n20 to determine whether the co-occurrence flag of this word is on. If the co-occurrence flag is on, the word is more likely to be used in the document being created as compared to other words, as described later, so that there is almost no possibility of a user input error. For this reason n20
If the co-occurrence flag is turned on, the program proceeds to n15 to determine the input character string. On the other hand, if the co-occurrence flag is not turned on at n20, the process proceeds to n21 shown in FIG. Note that n
The processing after 21 will be described later.

If there is no matching word even in n9 (if there is no word of the input character string), it is an input error (spelling error) of the user, so that the English-Japanese dictionary 4a is searched and similar to the input character string. The word of the character string to be extracted is extracted as a candidate word (n10). The display unit 5 displays these candidate words (n11), and the control unit 3 waits until one candidate word is selected from the candidate words displayed by the user (n12). When the user selects one candidate word from the displayed candidate words, the input character string is replaced with the character string of the selected candidate word (n13), and the replaced character string (word) is determined ( n14), n1
Proceed to 6.

For example, when the user erroneously inputs the word "weight" as "weight" (however, the character string "weight" is a word not registered in the English-Japanese dictionary 4a (a word that does not exist)). , The control unit 3 sets a character string similar to this character string to “weigh”.
Words such as "t" and "right" are extracted from the English-Japanese dictionary 4a as candidate words, and these candidate words are displayed on the display unit 5 (see FIG. 7). When the user selects the word "weight" intended by the user, the input character string "weight" is replaced with the character string "weight" of the candidate word selected here, and the process proceeds to n15. Determine the character string. The selection of a word may be performed by moving a cursor (not shown) on a candidate word to be selected and clicking a mouse button (not shown), or by a specific key operation. It may be performed.

When the control unit 3 determines the word at n15,
Words co-occurring with the word determined in n16 (co-occurrence words)
Is determined, and if there is no co-occurring word, the process returns to n1 and the above processing is repeated. In addition,
Whether or not there is a word having a co-occurrence relationship is determined by the English-Japanese dictionary 4a.
Can be determined based on whether or not the co-occurrence word dictionary index of the word record 11 of the determined word in is 1. If a word co-occurring with the determined word exists, n
Then, a word whose synonym record 15 is registered in the synonym table 4b is extracted from the co-occurrence words having a co-occurrence relation with the determined word (n17). In this processing, the co-occurrence word dictionary index of the word record 11 of the word determined by searching the English-Japanese dictionary 4a is extracted, and the similar word table index of the word record 11 indicated by the co-occurrence word dictionary extracted here is extracted. This is the processing to be performed. Then, the similar word table 4a is searched, and the co-occurrence flags of all the similar word records 15 indicated by the similar word table index extracted at n17 are turned on (n18). That is, the co-occurrence flag of a word having a co-occurrence relationship with the word determined in n15 and having a similar word is turned on. It can be said that a word that has a co-occurrence relationship with the determined word is a word that is more likely to be input than other words. Also, for each word for which the co-occurrence flag is turned on in n18, if there is a similar word whose character string is similar and the co-occurrence flag is on, the co-occurrence flag of this word is turned off (n19). Returning to n1, the above processing is executed.

For example, if the determined word is "weight"
, At n17, the co-occurrence word dictionary index “13” registered in the word record 11 of “weight”
981 "," 21083 ", etc., and searches the word record 11 indicated by the co-occurrence word dictionary index, and finds a similar word table inde of the word record 11 searched here.
x “3689” or the like is extracted. In n18, “light” shown in the similar word table index extracted in n17
t "and the like, and the co-occurrence flag in these similar word records 15 is turned on (set to 1). Further, if there is a word whose co-occurrence flag is on among words similar to the word "light" or the like for which the co-occurrence flag is on in n18, the co-occurrence flag of this word is turned off. For example, if the co-occurrence flag of “night” is on, this co-occurrence flag is turned off. Thereby, when the user inputs “light”, the user may mistakenly input “light”
When "t" is input, an input candidate is displayed as described later to allow the user to confirm the input and recognize an input error, and when "right" is input accurately, Since this word is determined without prompting the user to confirm the input, the document creation efficiency of the user is not reduced.

Next, returning to n20, the synonym table 4
The case where it is determined that there is a word that matches the character string input to b will be described (see FIG. 5). The control unit 3 focuses on the synonym record 15 indicated by the synonym index in the synonym record 15 of the word (original word) that matches the input character string (n21), and determines whether the co-occurrence flag is on. Is determined (n22). If the co-occurrence flag is not turned on, attention is paid to the synonym record 15 indicated by the synonym index of the synonym record 15 of interest (n2
3) It is determined whether or not the similar word record 15 of interest is a similar word record 15 of the original word (n24).
If n24 is not the original word, the process returns to n22 and the above processing is repeated. By the processing of n22 to n24, it is determined whether or not there is a word whose co-occurrence flag is on among the similar words of the character string similar to the input character string.

Here, if there is a similar word for which the co-occurrence flag is on, the similar word for which the co-occurrence flag is on is used in the document being created as compared with other words (input). )
Since the possibility is high, it can be said that the word intended by the user is likely to be a similar word for which the co-occurrence flag is on. In other words, there is a high possibility that the user made an input error. When it is determined that the co-occurrence flag is on at n22, n
Proceeding to step 25, the display mode of the candidate word is set to type 1,
The processing after n29 is executed.

For example, as shown in FIG.
When the co-occurrence flag is turned on and a character string "right" similar to "light" is input, the display mode of the candidate word on the display unit 5 is set to type 1 in n24. It should be noted that, here, “weight”, which is co-occurring with “light”, is input to “light”.
ht ”is more likely to be input than other words. Therefore, the word intended by the user is “l
It can be said that there is a high possibility that "right" has been input due to an input error.

On the other hand, in the processing of n22 to n24, if there is no similar word for which the co-occurrence flag is on, n24 to n
Proceeding to 26, it is determined whether or not the word of the input character string is a word of low use frequency (whether or not the display flag is on). If it is determined in n26 that the word is not an infrequently used word (if the display flag is off), the possibility of a user input error is low, and the process proceeds to n15 to determine the word of the input character string. Execute the process.

If it is determined in step n26 that the input character string is a word with a low frequency of use, the display mode of the candidate word is set to type 2 in step n27, and the flow advances to step n28. here,
When the frequency of use of the word of the input character string is low and there is a word of a character string similar to the input character string, the word intended by the user is a word of low frequency of use (input word It is highly likely that the word is not a character string word) but a character string similar to the input character string. For example,
If the user inputs “license” and the frequency of use of “license” is low, the word intended by the user is changed to “license”.
It is highly possible that the name is “rice” similar to “ce”.
That is, there is a high possibility that an input error has been made by the user.

In the character string input device according to the present embodiment, when it is determined that there is a high possibility that an input error has been made by the user, the display form of the candidate word is set to type 1 or type 2 and the processing after n28 is performed. Execute In n28, the English-Japanese dictionary 4a is searched to extract similar words whose meaning is similar to the meaning of the word of the input character string and the meaning of the similar word. It is determined whether or not the display form of the candidate word is type 1 (n
29) If it is type 1, as shown in FIG. 8 (A), the word of the input character string and the similar word extracted in n28 are displayed as candidate words (n30), and it is confirmed to the user whether there is any input error. Let it. Note that it is assumed that a word of the input character string is not a user input error, and thus the candidate word includes a word of the input character string. FIG. 8A shows a case where the character string is similar to “r” when the co-occurrence flag of “light” is on.
This is an example when “light” is input. When the user selects one candidate word from the displayed candidate words (n31), the control unit 3 replaces the input character string with the candidate word (n32), proceeds to n15, and replaces the character string here. Is determined, and the processing after n16 is executed. Note that the candidate word can be selected by operating a mouse to position a cursor (not shown) on the intended word and clicking a mouse button.

As described above, instead of having the user confirm whether there is a mistake in inputting only on the condition that a word having a character string similar to the input character string exists, a similar word of the input word is replaced with another word. When the word is more likely to be used in the document being created than the word (when a word that has a co-occurrence with a similar word is already used in the document being created), check if the user has made an input mistake Therefore, it is possible to prevent the user from frequently performing this check, not to lower the document creation efficiency, and to allow the user to almost certainly recognize an input error. Note that instead of the display of FIG. 8A, only similar words for which the co-occurrence flag is on may be displayed as candidate words as shown in FIG. 8B (here, “fight”). , The co-occurrence flag of “night” is off.) In this way, the display is not complicated, and the user can more easily select a word.

If it is determined in step n29 that the display mode is not type 1, it is determined in step n34 whether or not the display mode is type 2 (see FIG. 6). In the case of type 2, the input word and its meaning are displayed on the display unit 5 as shown in FIG. 9 (n35), and the user is made to confirm whether there is an input error. Here, since the word input by the user is a word used infrequently as described above, it is determined that there is a high possibility of an input error by the user, and the user is checked whether there is an input error. FIG.
In the type 2 display mode shown in FIG. 5, the meaning of the word of the input character string is displayed to allow the user to confirm whether or not there is an input error. In FIG. 9, reference numeral 21 denotes a display button, and reference numeral 22 denotes a similar word display button.

When the type 2 display screen shown in FIG. 9 is displayed on the display unit 5, the control unit 3 waits for input of any of candidate selection, display off, and similar word display (n36 to n38).
Note that the display can be turned off by moving the cursor (not shown) on the display off button 21 and clicking the mouse button, and moving the cursor on the similar word display button 22 and clicking the mouse button. By doing so, a similar word display can be input. Further, the display may be turned off or a similar word may be input by a specific key operation. When the user inputs the similar word display, it is determined whether or not the current display mode is type 2 at n39. If the current display mode is type 2, the display mode is changed to type 3 (n40). The display mode is changed to type 2 (if it is type 3) (n41), and the process returns to n34. Note that n34
If it is determined that the input word is not Type 2, not only the input word and its meaning but also the character string of the similar word and its meaning as shown in FIG. 10 are displayed at n42, and the user's input is waited at n36 to n38. When the user inputs a similar word display in the state shown in FIG. 10, the screen of the display unit 5 returns to the screen shown in FIG.

When the user selects a candidate word, the input character string is replaced with the selected character string (n43), and n
Proceeding to step 15, the replaced word is determined, and the processing after n16 is executed. When the user inputs the display off, the display flag in the similar word table 15 of the original word is turned off (n45), the process proceeds to n15, the word of the input character string is determined, and the process after n16 is performed. Execute

As described above, when the user inputs the display off, the display flag of the word of the input character string is turned off,
This word is set as a word that is not frequently used. For this reason, even if the word is generally used infrequently, if the word is frequently used by the user himself / herself, the word can be set as a word not infrequently used, and the screen shown in FIG. 9 is displayed. Will not be done. Therefore, it is not necessary for such a user to confirm whether or not there is an input error every time this word is input, and the document creation efficiency of the user is not reduced.

In this embodiment, the character string input device according to the present invention is applied to an English word processor. However, the present invention can be applied to other languages such as German. In addition, when a word is input, it is determined whether or not this word is an input error. However, words are sequentially extracted from the beginning of the created document, and the above-described processing after n6 is performed for each extracted word. And a character string input device that allows the user to confirm whether there is an input error when there is a high possibility of an input error.

[0050]

As described above, according to the present invention, not only is it determined whether or not a word having the same character string as the input character string exists, but also based on the relationship with the already input word. Decreases the efficiency of the user's document creation by determining whether the entered character string is a word intended by the user and, when it is highly probable that the user has made an input error, prompting the user to check for the input error. Without making the user input, it is possible to make the user recognize the input error almost certainly.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a character string input device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of an English-Japanese dictionary in the character string input device of this embodiment.

FIG. 3 is a diagram showing a configuration of a similar word table in the character string input device of the embodiment.

FIG. 4 is a flowchart showing processing of the character string input device of the embodiment.

FIG. 5 is a flowchart showing processing of the character string input device of the embodiment.

FIG. 6 is a flowchart showing processing of the character string input device of the embodiment.

FIG. 7 is a diagram illustrating an example of a display screen on a display unit.

FIG. 8 is a diagram illustrating an example of a display screen on a display unit.

FIG. 9 is a diagram illustrating an example of a display screen on a display unit.

FIG. 10 is a diagram illustrating an example of a display screen on a display unit.

[Explanation of symbols]

 1-character string input device 2-input unit 3-control unit 4-dictionary file 4a-English-Japanese dictionary 4b-similar word table 5-display unit 6-program storage medium 11-word record 15-similar word record 21-display button 22 -Similar word display button

Claims (14)

[Claims] An input unit for inputting a character string, and a co-occurrence set in advance for each word as a related word having a similar idea such as a similar word, a synonym or an antonym that is similar to the character string. A storage unit for directly or indirectly storing words and the presence or absence of a co-occurrence state; and when a similar word of the input word is in the storage unit, the similar word is an input candidate if the similar word is in a co-occurrence state. And a control unit for setting a co-occurrence state of the co-occurred word of the determined word when the input word is determined. 2. The character according to claim 1, wherein, if the input word itself is already in a co-occurrence state, the control unit determines the input word without displaying the similar word on a display unit. Column input device. 3. The character string input according to claim 1, wherein the control unit resets the co-occurrence state of the similar word if a similar word similar to the co-occurrence word of the determined word is in the co-occurrence state. apparatus. 4. The storage unit further stores information indicating whether or not each of the words is a low-frequency word set in advance, and the control unit determines that the input word is a low-frequency word. 4. The storage unit according to claim 1, wherein the storage unit determines whether or not there is a word, and if the word is infrequently used, a similar word of the input word is displayed on the display unit as an input candidate. A character string input device according to Crab. 5. An input unit for inputting a character string, and storing, for each word, a similar word set in advance as a similar character string and information indicating whether the word is a predetermined infrequently used word or not. A storage unit that determines whether or not the input word is an infrequently used word by referring to the storage unit. If the input word is an infrequently used word, a similar word of the input word is input. A character string input device comprising: a control unit configured to display a candidate on a display unit. 6. The character string input device according to claim 4, further comprising means for inputting information on whether the use frequency is low. 7. For each word, a similar word set in advance as a similar character string, a co-occurrence word set in advance as a meaning having a semantic idea such as a synonym or antonym, and presence / absence of a co-occurrence state are determined. It is stored directly or indirectly, and by referring to the stored contents, it is determined whether or not there is a similar word of the input word. If there is a similar word, if the similar word is in a co-occurrence state, the similar word A character string input method that displays words as input candidates and sets the co-occurrence state for the co-occurred words of the determined word when the input word is determined. 8. The character string input method according to claim 1, wherein if the input word itself is already in a co-occurrence state, the input word is determined without displaying the similar word on a display unit. 9. The character string input device according to claim 7, wherein if a similar word similar to the determined word co-occurrence word is in the co-occurrence state, the co-occurrence state of the similar word is reset. 10. In addition to the storage contents, information indicating whether or not each word is a word of low use frequency set in advance is stored, and whether or not the input word is a word of low use frequency is determined. 10. The character string input method according to claim 7, wherein if the word is a word that is not used frequently, a similar word similar to the input word is displayed as an input candidate. 11. For each word, a similar word set in advance as a similar character string and information indicating whether or not each word is a preset infrequently used word are stored. A reference is made to determine whether or not the input word is an infrequently used word. If the input word is an infrequently used word, a character string indicating a similar word similar to the input word as an input candidate input method. 12. The character string input method according to claim 10, wherein information on whether the use frequency is low is input. 13. For each word, a similar word set in advance as a character string is similar, a co-occurrence word set in advance as a semantic idea such as a synonym or antonym is related, and the presence / absence of a co-occurrence state. A first step of determining whether or not there is a similar word of the input word by referring to the memory unit, the first step being applied to an apparatus having a storage unit for directly or indirectly storing the similar word; A second step of determining whether the word is in a co-occurrence state; a third step of displaying the similar word as an input candidate if the word is in a co-occurrence state; And a fourth step of setting a co-occurrence state with respect to the utterance. 14. A storage unit for directly or indirectly storing, for each word, a similar word preset as a similar character string and information indicating whether or not the word is a preset word having a low frequency of use. A first step of determining whether or not an input word is an infrequently used word by referring to the storage unit; and, if the input word is an infrequently used word, Displaying a similar word similar to the input word as an input candidate.