JPH0311459A - Document preparing device - Google Patents

Abstract

PURPOSE:To prevent the inappropriate operation of a clause re-segmenting learning function performed by applying the clause re-segmenting learning function only when reading that coincides with each other exists in an input KANA (Japanese syllabary) character string. CONSTITUTION:It is judged whether or not the reading coinciding with that of data being registered on a clause re-segmenting learning dictionary 8 exists in a KANA character string from an input part 1 with a reference part 6. When the reading coinciding with that of the data being registered on the clause re-segmenting learning dictionary 8 exists in the input KANA character string, it is judged whether or not a block used in automatic blocking for KANA/ KANJI (Chinese character) conversion is included in a range where the reading coincides with each other with a control part 5. When the block used in the automatic blocking for the KANA/KANJI conversion is included in the range where the reading coincides with other, the clause re-segmenting learning function using the dictionary 8 is applied.

Description

[Detailed description of the invention] [Purpose of the invention] (Industrial application field) The present invention relates to a document creation device.

(Prior Art) Document creation devices, such as word processors, are known to have a learning function for changing phrases.

For example, if you want to create a document that says ``Here, take off your kimono,'' first, ``Here you will take off your kimono.'' (Figure 7 (a)
) is input, and the kana-kanji conversion section performs kana-kanji conversion on this kana character string. As a result, it may be erroneously converted into, for example, "Take off your footwear here" (FIG. 7(b)). This is because the kana character string "Koko wa Kimono wo" was incorrectly recognized and processed as consisting of the clauses "Koko de" and "Hakimono wo", and it was necessary to change it to the desired clause. There is. Therefore, for the kana character string ``Koko wa kimono wo'', use arrow A etc. to specify the point between ``Koko wa'' and ``Kimono wo'' as the bunsetsu re-cutting point (Figure 7(C)). Re-convert the kanji to create the desired sentence ``Here I take off my kimono'' (
Figure 7(d)) was obtained.

In order to make the learning function work for such phrase re-cutting, as shown in Figure 8, all the pronunciations of the phrase (first phrase) immediately before the phrase re-cut position must be changed to ``kokowa'' and The reading of the clause immediately after the correction position (second clause) is “kimono wo”
Edit the phrase and register it in the learning dictionary.

From now on, use kana character strings that include ``here kimono'', such as ``
The learning function is applied when "Here I will buy a kimono" (Figure 7 (e)) is input, and "Here I will buy a kimono" is input.
I got the conversion result.

(Problem to be solved by the invention) However, such a learning function for changing phrases,
If the input kana character string matches the contents of the phrase reshuffling learning dictionary, the phrase is unconditionally reshuffled at the registered position, so characters that are homophones with the character string registered in the learning dictionary are used. Even when a column is input, the learning function works inappropriately and the desired result cannot be obtained.

The present invention was made to solve such problems,
The purpose is to provide a document creation device in which the learning function for re-cutting phrases does not work inappropriately.

[Configuration of the Invention (Means for Solving the Problem) In order to achieve the above object, the present invention is capable of performing kana-kanji conversion on an input kana character string by using automatic block cutting. and means for determining whether or not an input kana character string has a pronunciation that matches data registered in a phrase re-cutting learning dictionary. , If there is a pronunciation in the input kana character string that matches the data registered in the bunsetsu recutting learning dictionary, automatic block cutting for kana-kanji conversion will be performed within the range of this - matching pronunciation. A means for determining whether or not a break to be used is included, and a method for re-cutting bunsetsu if a break to be used for automatic block cutting for kana-kanji conversion is included within the range of matching readings. The present invention is characterized by providing a means for applying the learning function.

(Function) In the present invention, it is determined whether or not there is a pronunciation in the input kana character string that matches the data registered in the bunsetsu trimming learning dictionary, and if there is a pronunciation that matches, It is determined whether or not the range of matching readings includes a break used for automatic block cutting for Kana-Kanji conversion, and only if this break is included, the phrase re-segment learning function is applied. applies.

(Embodiment) Hereinafter, a word processor according to an embodiment of the document creation device of the present invention will be described with reference to the drawings.

The input unit 1 has, for example, a keyboard, can input a kana character string indicating a document to be created, and sends the input data signal to the control unit 2. The control unit 2 outputs a predetermined signal based on the sent data signal to the output unit 3 and the kana-kanji conversion unit 4.
It also performs basic operational control of the output section 3, kana-kanji conversion section 4, and other components.

The output section 3 has, for example, a CRT display, and the control section 2
Displays the input kana character string and the results of kana-kanji conversion based on a predetermined signal sent from. The kana-kanji conversion unit 4 performs kana-kanji conversion on a predetermined kana character string by automatic block cutting using a maximum of 4 clauses and shortest of 3 clauses, for example.
A data signal indicating the conversion result is sent to the control section 2.

When the kana-kanji converter 4 is operating, the recutting control unit 5 controls the reference unit 6 if there is no request for phrase recutting, and controls the registration unit 7 according to the recutting position if there is a request for phrase recutting. The reference unit 6 refers to the phrase reshuffling learning dictionary 8 and determines whether or not the input kana character string has a pronunciation that matches the data registered in the dictionary. The registration unit 7 registers the reading of the phrase immediately before the phrase re-cutting position (first phrase) in the "first phrase reading" column of the phrase-cutting learning dictionary 8, and also registers the reading of the phrase immediately before the phrase re-cutting position (the first phrase), and The pronunciation of (second clause) is registered in the "pronunciation of second clause" column of the clause reshuffling learning dictionary 8.

The operation of this word processor will be explained below.

For example, a case will be explained in which a document that says ``I will take off my kimono'' will be created.

First, when a kana character string ``Here I wipe my kimono'' is input into the input unit 1, a data signal indicating this kana character string is sent to the control unit 2. The control section 2 sends a predetermined signal based on this data signal to the output section 3 and the kana-kanji conversion section 4.

The output unit 3 displays this kana character string ``Here, wipe your kimono'' on the CRT display (Fig. 3(a)).
. The kana-kanji converter 4 is able to write this ``wipe your kimono here.''
This automatically blocks the kana character string and performs kana-kanji conversion. During this kana-kanji conversion, the reswitching control unit 5 performs the following processing. Note that in this case, since a new document is being created, there is no request for segment re-cutting, and the re-cutting control section 5 controls the reference section 6. The reference unit 6 refers to the phrase re-cutting learning dictionary 8 and determines whether or not there is a pronunciation that matches the registered data in the input character string "Here, wipe the kimono" (step 201).

Clause re-cutting learning dictionary 8 says, “Here I wipe my kimono.”
Since data related to the reading has not yet been registered, it is determined that there is no data to be referred to, and normal kana-kanji conversion is performed by the kana-kanji conversion unit 4 (step 202). In this case, the conversion result is, for example, "take off your footwear here." The output unit 3 displays the sentence ``Now take off your footwear'' according to the above conversion result (FIG. 3(b)).

However, since the phrase group of the displayed sentence is different from that of the desired sentence "Here I take off my kimono," it is necessary to correct (reconvert).

Therefore, by operating the input unit 1, the user specifies the range to be re-converted, ``footwear here''.

The control unit 2 reads back all the characters in the specified range into hiragana as originally input, and causes the output unit 3 to display ``Now take off your kimono'' (FIG. 3(C)).

Next, by operating input section 1, select "Here"
Specify the position between and ``kimono wo'' as the phrase re-cutting position. If, for example, a phrase change conversion key is pressed on the keyboard of the input unit 1 to request phrase change, the following process is performed.

The kana-kanji conversion unit 4 performs kana-kanji conversion on the first phrase "here". In this case, there is no part that is converted to kanji, and it remains as ``here''.

On the other hand, the re-selection control unit 5 controls the registration unit 7 to re-select the reading of the first phrase “here” in the learning dictionary 8.
Register it in the "Reading of the clause" column.

The kana-kanji conversion unit 4 further performs kana-kanji conversion on the second phrase "kimono wo" to obtain a conversion result of "kimono wo." On the other hand, the registration unit 7 registers the reading of the second phrase "Kimono wo" in the column of the "Reading of the second phrase" of the phrase reshuffling learning dictionary 7. In addition, as shown in FIG. 4, the phrase re-cutting learning dictionary 8 at this point has "Here" and "Here" in the "First phrase reading" column and "Second phrase reading" column, respectively. Kimono.”
are registered in correspondence with each other.

Based on the above conversion result, a data signal indicating the sentence "Here, take off your kimono" is sent from the kana-kanji conversion section 4 to the control section 2. Based on this data signal, the control section 2 causes the output section 3 to display the sentence "Here, take off your kimono" (FIG. 3(d)).

Thereafter, when a kana character string including the kana character string ``Here we wear kimono'' is input, for example, ``Here we wear kimono'', the following processing is performed.

The output unit 3 displays the kana character string ``I am wearing kimono here'' (FIG. 3(e)). On the other hand, at this point, there is no request for recutting, and the recutting control unit 5 allows the reference unit 6 to refer to the phrase recutting learning dictionary 8, and in the input kana character string ``Here I am wearing a kimono''. It is determined whether there is a pronunciation that matches the registered data (step 201). In this case, the kana character string "Koko wa kimono wo" in the kana character string "Koko wa kimono wo" is changed to "Koko wa kimono wo" in the first clause already registered in the phrase reshuffling learning dictionary 8. Since this matches the character string consisting of the reading of the second clause, ``Kimono'', it is determined that there is data to be referenced.

Next, it is determined whether or not the range of the kana character string “here kimono wo” that matches the data in the bunsetsu re-cutting learning dictionary 8 includes the delimiters used for automatic block cutting for kana-kanji conversion. A determination is made (step 203).

In this case, since that break is not included, the normal kana-kanji conversion is performed in the kana-kanji conversion unit 4 without applying the phrase re-segment learning function (step 202). Get the conversion result.

Such a conversion result can be easily converted into a desired homonym document, for example, ``Here, take off your kimono'' by pressing a ``next candidate key'' (not shown) provided in the input unit 1. can.

Next, as another example, a case will be described in which a document titled "70th United High School Baseball Championship" is created.

First, when a kana character string such as ``Dai 70 Kaizenkoku Koukou Yaki Yusenshi Yuken Taikai'' is input into the input unit 1, a data signal indicating this kana character string is sent to the control unit 2. The control section 2 sends a predetermined signal based on this data signal to the output section 3 and the kana-kanji conversion section 4.

The output unit 3 displays on the CRT display the kana character string ``Dai 70 Kaizenkoku Kouyaki Yusenshi Yuken Taigai''. The kana-kanji conversion unit 4 performs kana-kanji conversion on the kana character string "Dai 70 Kaizenkoku Koukouyaki Yusenshi Yuken Taigai" by automatic block cutting using, for example, the 4-segment longest and 3-segment shortest method.

During this kana-kanji conversion, the reswitching control unit 5 performs the following processing. In this case, since a new document is being created, there is no request for phrase recutting, and the recutting control section 5 controls the reference section 6. The reference unit 6 refers to the phrase reshuffling learning dictionary 8 and finds a pronunciation that matches the already registered data in the input character string "Dai 70 kaisenkoku kou yaki yuusenshi yukentai gai". It is determined whether there is one (step 201). In this case, since data related to the pronunciation "dai 70 kaizenkoku kou kou yaki yuusenshi yuken taikai" has not yet been registered in the phrase reshuffling learning dictionary 8, it is determined that there is no data to refer to, and normally The kana-kanji conversion unit 4 performs kana-kanji conversion (step 202).

As a result, the automatic block cutting results in an erroneous conversion result of, for example, ``70th National High School and Armistice Sovereignty Tournament''. The output unit 3 displays the above conversion result, that is, the sentence [70th National High School and Armistice Sovereignty Tournament] (see Figure 5(a)
)).

However, since the phrase group of the displayed sentence is different from that of the desired sentence "70th United High School Baseball Championships," it is necessary to correct it (retransform it again).

Therefore, by operating the input unit 1, the range to be re-converted, ``high school and armistice sovereignty,'' is specified, for example, by underlining it (not shown). The control unit 2 reads back all the characters in this specified range into hiragana as originally input, and outputs "Kou Kouyaki Yusen Shiyuken" on the output unit 3.
is displayed.

Next, by operating input section 1,
Specify the position between and ``yakyu'' as the phrase recutting position. If, for example, a phrase change conversion key is pressed on the keyboard of the input unit 1 to request phrase change, the following process is performed.

The kana-kanji conversion unit 4 performs kana-kanji conversion on the first phrase "kou-kou" and obtains the result "high school."

On the other hand, the re-selection control unit 5 controls the registration unit 7 to re-select the pronunciation of the first phrase “Kou-kou” in the learning dictionary 8.
Register it in the "Reading of the clause" column.

Further, the kana-kanji conversion unit 4 performs kana-kanji conversion on the second clause "Yakyu" to obtain the result "baseball".

On the other hand, the registration unit 7 registers the reading of the second phrase "Yakyu" in the column of the "Reading of the second phrase" of the phrase reshuffling learning dictionary 8. In addition, the phrase re-cutting learning dictionary 8 at this point includes:
As shown in Figure 8, the column ``Reading of the 1st clause'' and the ``Reading of the 2nd clause''
``Kou-kou'' and ``Yakyu'' are registered in the ``pronunciation of clauses'' column in correspondence with each other.

The kana-kanji conversion section 4 further sends a data signal indicating the text "70th Kana High School Baseball Championship" to the control section 2 based on the above conversion result. Based on this data signal, the control unit 2 causes the output unit 3 to display the message "70th National High School Baseball Championship Tournament" (see Figure 5(b)).
)).

From now on, we will use kana character strings that include the kana character string ``Kou Kou Ya Kyusenshu''.
・” is input, the following processing is performed.

Output part 3 is "...Kousouya Kyushuenshu..."
The kana character string (FIG. 5(C)) is displayed.

On the other hand, at this point, there is no request for reshuffling, and the reshuffling control unit 5 allows the reference unit 6 to refer to the phrase reshuffling learning dictionary 8.
The input "...Kousouya Kyusenshu..."
It is determined whether or not there is a pronunciation that matches the registered data in the kana character string (step 201).

In this case, the kana character string ``Kōkō yakyu'' in the kana character string ``...Kōkō ya kyūsenshu...J'' is the reading of the first clause already registered in the phrase reshuffling learning dictionary 8. Since it matches the character string consisting of "Kokou" and the reading of the second clause "Yakyu", it is determined that there is data to be referenced.

Next, it is determined whether or not the delimiter in automatic block cutting during kana-kanji conversion is within the range of the kana character string "Kou Kouyakyu" that matches the registered data of the phrase re-cutting learning dictionary 8 ( Step 203). in this case,"
"Kou Kouya" in the kana character string "Kou Kou Yakyuu"
Separator B (
5(d)), a learning function using the phrase reshuffling learning dictionary 8 is applied. That is, step 20
According to the data obtained in 1, the reading of the first clause and the second clause are
In step 204), the kana-kanji conversion unit 4 performs kana-kanji conversion at this re-cutting position (step 205), and converts the desired document, for example, "...high school baseball player... . . ” is displayed on the output unit 3 (FIG. 5(e)).

Thus, according to the word processor of this embodiment, even if the input kana character string matches the contents of the phrase re-cutting learning dictionary 8, the phrase re-cutting learning function is not applied unconditionally, but automatically. Since we took into account the presence or absence of block breaks, the learning function would work inappropriately for character strings that are homophones of the character strings registered in the bunsetsu re-cutting learning dictionary 8. do not have.

Note that the automatic block cutting of the document creation device of the present invention can be applied to various methods such as, for example, a 4-phrase maximum and 3-phrase shortest method, and a 2-phrase longest matching method.

[Effects of the Invention] As described in detail above, according to the document creation device of the present invention, even when the input kana character string matches the contents of the phrase re-cutting learning dictionary, the phrase re-cutting learning function is not activated. is not applied unconditionally, but instead takes into account the presence or absence of automatic block separation, so the learning function may be inappropriate for character strings that are homophones of character strings registered in the learning dictionary. I never end up working.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a word processor according to an embodiment of the present invention, FIG. 2 is a flowchart showing the operation of the word processor of this embodiment, and FIGS. 3, 5, and 7 are similar to those shown in FIG. A diagram showing the status displayed on the output section shown in
4, FIG. 6, and FIG. 8 are diagrams showing examples of registered contents of the phrase reshuffling learning dictionary shown in FIG. 1. 1... Input section, 2... Control section, 3... Output section, 4
... Kana-kanji conversion unit, 5... Re-switching control unit, 6.
...Reference section, 7...Registration section, 8...Phrase re-cutting learning dictionary 2 Figure 1 (a) Here: 1 Yoshomono Otonugu. (g) No. Z-1... 5 Tadashi Fu//, Self-weight turnip 0.17tηsu1;
Figure 4 (a) Innoshima school at 70 years of age and Armistice Sovereignty Conference Figure 5 Figure 6 Figure 2 (a) 'S, Ga. 1 Yosamo Lhgu. (b) Sharp and rounded objects and the cutting edge. (f) Gosate・1ma1suvttL>.

Claims (1)

[Claims] (1) A document creation device that is capable of converting input kana character strings into kana-kanji using automatic block cutting and has a learning function for re-cutting phrases, converts input kana character strings into A method for determining whether or not there is a pronunciation that matches the data registered in the bunsetsu reshuffling learning dictionary; If there is a reading that matches, a means for determining whether or not the range of this matching reading includes a delimiter used for automatic block cutting for Kana-Kanji conversion, and a range of the matching reading. 1. A document creation device comprising: means for applying a learning function for re-cutting phrases when the document includes a delimiter used for automatic block-cutting for kana-kanji conversion.