JPS63184864A - Documentation device - Google Patents

Abstract

PURPOSE:To easily check a lack of coordination on representations in a document by preserving different representation information together with document data if plural representations exist for one word. CONSTITUTION:The character code sent from an input device 1 is stored in a key input buffer 44 and is used as the key to retrieve a code conversion dictionary 31. Retrieved contents are stored in a dictionary look-up buffer 41; and if plural candidates are obtained as the result of dictionary look-up, one of them is selected and determined by the indication from the input device 1, and different representation data is stored in a check item buffer 43 if the determined candidate is a word having different representations. This data is preserved in a document file 32 together with the document at the time of preserving the document. At the time of checking the representation, the document and different representation data are called to a document buffer 42 and the buffer 43 respectively from the document file 32, and different representation data is used to check a lack of coordination on representation in the document.

Description

DETAILED DESCRIPTION OF THE INVENTION [Objective of the Invention] C11t Business Application) This invention relates to a text creation device for processing a document including words written in different ways.

(Prior Art) Opportunities to create a sentence using a sentence-forming device are increasing year by year, but when converting a Japanese kana notation into a kanji character, there are multiple candidates. Homophone O1! Although the user can uniquely determine the term, there are some j4L words that have multiple spellings even if they are synonyms. For example, there are multiple ways to play the okurigana.

It is strange that d9 in this case has a uniform way of marking the number. The Kana/Kanji converter, which has a learning function, temporarily memorizes these information. However, it is not easy when considering the need to maintain control over a single meeting with different users or between 41 sentences.

However, in the conventional technology, a specific character string in a document
There are some that have the ability to perform IL, but there is no such thing that can be done with a single instruction for multiple items.

There is also no way to check the uniformity of notation between multiple documents.

(Interlude that the invention attempts to solve) As mentioned in the previous section, multiple notations are allowed for Japanese kanji, kana, and second sentences, and it is desirable to unify these six notations, but one sentence It's a pain to have to worry about using kanji notation while creating.

The present invention performs a unified check on Japanese kanji notation during writing or proofreading.

If there is no uniformity in notation, this information is kg! It is an object of the present invention to invent a text creation device that has a function of informing the user and correcting it on the spot if necessary.

[Structure of the invention]

(Tane point? The tth way to solve the problem) For things that have multiple notations during kana-kanji conversion,
Register in this single Ifik test item dictionary. This registration may be done in advance, or the completed sentence 1laF1 may be specified and registered even during the creation of sentence-d. For example, the kanji notations for ``Moshikomi'' include ``Application'', ``Application'', and ``Application'', and these three notations are stored together with the selected notation code information Shichimonbara code. After that, the test item dictionary and the notations used in the document and t
Check @ and search for inconsistency vI on one expression. Corrections made after the search may be made by alerting the text creator, or may be made all at once.

(Function) By using the nine methods mentioned in the previous section, you can easily check for unreadable notation in a document and correct it on the spot.
Not only can the process be completed in a shorter time than the conventional procedure of adding inspection items and then searching and replacing characters, but it also allows for uniformity in notation across multiple documents.

(Implementation row) The second block configuration for implementing the present invention is shown in Figure 1.

Sentence creation *R is done manually by converting the characters entered by the user by typing on the keyboard R key into R code. A display device that takes the character code and displays it on the character r display.Create it with this device and store it in the code conversion dictionary #, such as Kana-kanji conversion, which is necessary when saving nine documents or creating one document.
Storage device A-31 Temporarily necessary information when creating documents
It consists of a storage device B-4 for storing data, and a control device 5 for controlling these devices. There is 1 in the control device 5.
Control program 5 showing specific control procedures for the apparatus of ~5
1 is stored, and the control device 5 stores a control program 51.
Control is performed by interpreting the controller 'l1l-1 step at a time.

Device A-3 contains a code conversion dictionary containing essential information when converting from hiragana code to kanji code.
#31, a document file 32 containing the character codes of nine documents created by this device, and a dictionary 33 of nine test items containing words with different spellings among homophones are stored. t
The storage device 11B-4 includes a dictionary lookup buffer 411!1. which temporarily stores the contents of the code conversion dictionary f31. ,
A document buffer 42 that temporarily stores the contents of the document or document file being created, a list of abbreviations with different spellings, an inspection item buffer 43 that temporarily stores the contents of the document or document file being created, and a temporary character code sound sent from the input device. There is a key manual buffer 44 that stores keys.

Next, the control program 51 in the control device 5 will be explained using the second figure sound. When the control program is started, the devices 1 to 5 are initialized to make one text creation device ready for use.

(81) Next, display the work menu 1 on the display device 2,
Prompt the user to select a work item. There are three work items: text creation, correction, and spelling check, which can be selected manually from input device 1 (S
2). When 1-piece (S3) is selected, memory fililB
It is within -4. Clear the contents of the four buffers (8
4), move on to editing the text (85). -1 corrected (S6)
When is selected, document file 3 in storage device A-3
Select the target sentence #kura from 2 (87), edit the sentence (S
Proceed to 5) 0 When editing is completed, give an appropriate document name r in the 1 document file 32 and save it as a document (88) L.

Return to 1 job selection (S2). If the notation check (S9) is selected during work selection (S2), a notation check is performed to see if the document edited immediately before is present. At this time, if necessary, f'T! from one document file 32! Completed sentence 4i
This can also be checked by pulling it into the document buffer 42 (810).

If a different spelling of Rin is found by checking, this message will be displayed! 112 and can be modified (811) if necessary. @Figure 3 is a display screen diagram after checking for different spellings. Here, 301 and 302 indicate the correct spelling of the word r.

Next, we will explain in detail how to check whether words used in a created document are spelled differently.

First of all, the structure 1 of the code conversion dictionary $31 is shown in FIG. In this column, the keyword ``moshikomi'' corresponds to the m-flat word ``application'' and ``application.''

The keyword "Owari" is matched with the converted word "Owari", "End", "End", "Tsuribari", "Owari" r, and the result is tla. The converted words ``application'' and ``application'' are spelled differently, with different street names.

The same is true of the relationship between the kings of ``the end'', ``the end'', and ``the end''. ``Tsuribari'' and ``80%'' are treated as conversion words that have different meanings from the groups ``end'', ``end'', and ``end''.

That is, Hiragana middle-word “Motsushikomi” l Wari J
(Mr, M2) has a conversion word pointer that points to S-gana-kanji conversion information for #ζ (PI, P3).

These conversion word pointers PI and P3 further point to homophone pointers (P2, P4 to P6), and the number of homophone pointers is equal to the number of homophones of the hiragana keyword that is the parent of the pointer. (“
(1 mark for ``Entry'', 3 marks for ``End'') The pointer ends with a pointer end mark (gl, E3) indicating the end of the pointer. Then, substitute 2 for the kanji notation code (7 for Kl~) at the position pointed to by each open-heteronym pointer. Furthermore, for homophones, use a different notation terminator ('I'1-T7)
Separated by 23? ), if there is no s notation, it ends with a different notation end mark (g2, g4-g6).

After making the preparations in this way, we create a list of words with different spellings and test item buffers 43 and ζ while creating sentences in #Ja collection (S5). This flowchart is shown in Figure 7. When a character code is sent from the input device tl (821), if it is not a kana-kanji conversion code, the key code is stored in the key buffer 44 (521), but if it is a kana-kanji conversion code, it is stored in the key human buffer 44. The code conversion dictionary 31 is searched using the code as a key. The search contents are stored in the dictionary lookup buffer 41 (823
) If there are multiple candidates as a result of dictionary lookup, display them on the display device.
824) Select from the candidates.

input! 1Jtl (825), and if the determined candidate is a word with a different spelling, a list of different spellings is created (826).

The different notation list is created from the conversion dictionary data copied to the dictionary lookup buffer 41.

The keyword “
The data structure of "Moushikomi" and "2Wasu" is as shown in Figure 5. Here, if the converted word "application" is selected for the keyword "moshikomi", there is another candidate "application" that can be referenced with the homophone pointer P2, so "
It turns out that "Application" and "Application" have different notations, and the O information is stored in the test item buffer 43 in a data structure as shown in FIG. This consists of a list of alternate orthographic words 401 and a list of authentic orthographic words 402 . Also, for Application J (403), add a selection mark (404) to indicate that this notation has been selected.

For the homophonic notation [application J (405), insert the variant ``application j (406)''. On the other hand, data for 91 notations can be selected and stored in one document.

When the conversion word "end" is selected for the keyword "3wari", member notation resistance data is created (407
~412) When "r Owari" or "20%" is selected, is it different written word data? There is no need to collect 1i. These data are saved together with the document in one document file 32 when the document is saved (@2, FIG. 88).

Next, referring to the method of checking the orthography using the different orthographic word data r shown in Figure @4, the 0 notation check (Fig. 2 89) is performed from the document file 32 to the sentence gramophone document buffer 4
is stored together with one document! For example, the data in the test item buffer 43 is realized by calling the different notation data into the test item buffer 43 and first performing a comparison check between the IA notation data in the test item buffer 43 and the test item dictionary 33. is as shown in Figure @4, and the data in the test item dictionary - $33 is as shown in Figure g'7. Among the data in the test item buffer 43, 1i
The test item dictionary 33 is searched using the notation t keyword with the 4-choice mark (404). That is, if you search for the test item dictionary $331!- with the keyword "Application J (403)", you will obtain the items "Application" → "Application" (703, 704). In the inspection item dictionary 33, keyword Ti is included.
Since no selection mark is attached to , there is a mismatch between the contents of the different notation buffer 43 and the inspection item dictionary 33.

(@Figure 2 s Io) Through the above steps, the discrepancy in notation is recognized and ef! Then, the normal notation items (701 to 704) are copied from the test item dictionary 33 to the test item buffer 43, and the different notation items (701 to 704) are copied from the test item dictionary 33 to the test item buffer 43.
03) f! :dP- to search for the sentence w in the document buffer 42, and if it is relevant, it may be corrected at that time, or the relevant portion may be corrected all at once. (81
1) Mismatch information of t1 notation r When displaying on the display device 2, the correct notation item and the different notation item are displayed differently (color, brightness?
(e.g. frog) will make it easier to find the relevant location.

In this way, by devising the structure of code conversion dictionary v31 data and storing variant spelling information together with the document data, when searching for words with variant spellings, you can search for all words with variant spellings. There is no need to do this, and the processing time is greatly reduced. In other words, information on different spellings of words used in the text is recorded together with the document (2), and the checking process when these words are reused is greatly shortened.

Is it a fist story strike with @ notation in every document? When checking the notation, please check the inspection item name 7JF33L-.
Instead of referring to it, if you define the specific sentence 4F and 11 and use it when checking the comparison of Imoki Seihokura included in this,
It is possible to modify search preferences more quickly than with words t that are written differently between two documents. In this embodiment, we have described checking the okurikana notation, but there is no need to be particular about this.For example, if you group vulgar words with multiple katakana notations and register them in the code conversion dictionary, you can check the katakana notation. It can also be used for

The structure of the nine conversion chopsticks shown in FIG. 5 is an example.

Instead of connecting the conversion items with pointers, a method of arranging the data by showing one item at the beginning may be considered.

Furthermore, when displaying ephemeral kana-kanji conversion candidates (@6, 824), the conversion dictionary (Fig. 5) data in the dictionary lookup buffer 41 is used for commercial purposes, and only the correct spelling items of the kana-kanji candidates are rfed.
14PF! By displaying it as I.

The conversion process can be made more efficient. In other words, for the keyword ``moshikomi'', the R kansan ``application'' and oshikomi'' are different spellings.
Therefore, when displaying candidates (824), only ``Application'' is displayed, and candidates are not displayed for ``Application'', but a message indicating that there are alternative spelling candidates is displayed. However, this method is more effective for words that have many different spellings.

Retirement age - for the word ``2wari'', the f-conjunction ``end''
It can be seen that "end" and "end" have different notations according to the homophone pointer P4, so when displaying candidates (824), the homophone pointer Ps, P5. P6kffi displays only the three items ``End'', ``Hanging'', and ``Owari'', and displays a message indicating that there is an item with a different notation for the ``End'' item. Then, if necessary, you can add the ``end J (DI4 display item) to the conversion candidates.Which item among the different notation items should be corrected is determined when creating the code conversion dictionary 31. It is also a good idea to decide on this and change this 'CR when a different notation item is decided.

In short, various modifications can be made without departing from the gist of the present invention.

〔Effect of the invention〕

The text explained above is Kuicheng V&Yu T1! ! Words that have multiple spellings in Japanese depending on the two words they are used for! Even if the document contains text, it is possible to easily unify the notation within the document, as well as to unify the notation between multiple sentences.

In other words, is it possible to quickly search for differences in notation between two documents written by Yoseiji? Thus, it becomes possible to modify the variant notation @ along with the reference document fζ.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a text creation device according to an embodiment of the present invention, and FIG. 2 is a block diagram of a text creation device according to an embodiment of the present invention. III. Flowchart of the WJ program. Figure 3 shows the displayed columns after checking for different spellings. Figure 4 shows the data string r in the test item buffer. Figure 5 shows the data structure of the conversion dictionary. Figure 6 is a flowchart for creating a 4-notation list, and Figure 7 is a flowchart of the data in the inspection item dictionary.
This is a diagram showing l and k. l...Input device, 2...Display device, 3...Characterization device, A4...Storage device 8.5...Control device, 31
...Code conversion dictionary, 32...Document file, 33
...Frugal item dictionary-1iF, 41...Dictionary lookup buffer, 42...Bunsei bakufu, 43...Inspection item buffer, 44...Φ-human power buffer, 51...System-program . Agent Patent Attorney Nori Ken Yudo Takehana Kikuo No. 1 Fig. 2 Fig. 5 Fig. 7

Claims (2)

[Claims] (1) An input section for inputting character information, a storage section storing a conversion dictionary for performing kana-kanji conversion and a test item dictionary for searching for words with different spellings, and character information from the input section. a control unit that performs kana-kanji conversion using the conversion dictionary, and an output unit that outputs the kana-kanji conversion results from this control unit, and the control unit refers to the test item dictionary for the kana-kanji conversion results. A text creation device characterized by ejecting a different notation word and outputting a message to that effect to the output unit. (2) The input unit inputs a selection instruction signal for a word written in a different way, and the control unit identifies and stores the word in a different way for which selection is instructed in the test item dictionary. Sentence writing device described in section.