JP2021064019A - Sentence creation support device, sentence creation support system and program - Google Patents

Abstract

To provide a sentence creation support device, a sentence creation support system, and a program capable of improving accuracy of a prediction of a character or a character string used in the entire referenced document, compared to the one that predicts as a candidate to be input, when creating a new sentence by referring to one or more documents.SOLUTION: A sentence creation support device 10 includes: predicting means 44 for predicting a character or a character string extracted from a part of a second document that is different from a first document referred to by the user a part of the first document referred to by the user as a candidate to be input when creating a new sentence, when a user edits a first portion in the first document; and display control means 38 for controlling the character or the character string predicted by the prediction means 44 so as to be displayed when the user edits the first portion.SELECTED DRAWING: Figure 3

Description

The present invention relates to a sentence creation support device, a sentence creation support system, and a program.

Patent Document 1 describes the text of the received e-mail in a communication terminal that converts the input characters into characters based on the dictionary data, edits and sends the sent e-mail, and displays the received e-mail received from the outside. A character string acquisition means for acquiring a character string composed of one or a plurality of characters separated by a predetermined condition, a storage means for storing the character string acquired by the character string acquisition means, and the storage means. A control means for storing the character string acquired by the character string acquisition means, and an input for selectively displaying the character string stored in the storage means when editing a transmission e-mail to be replied to the received e-mail. The text of the transmitted e-mail based on the candidate display means, the character string selection means that accepts the operation of the operator who selects the character string displayed by the input candidate display means, and the operation received by the character string selection means. Disclose a communication terminal characterized by comprising an editing means for editing a character string.

Patent Document 2 describes a storage unit in which a character conversion dictionary is stored, an extraction means for extracting words and sentences from the data by analyzing data representing materials, and the above words and sentences extracted by the above extraction means. Discloses a document creation device including a registration means for registering in a conversion dictionary and a display processing means for displaying the words and sentences registered in the character conversion dictionary on a display unit as conversion candidates. ..

Japanese Unexamined Patent Publication No. 2008-52702 Japanese Unexamined Patent Publication No. 2014-56401

In the present invention, when a new sentence is created by referring to one or more documents, the character or character string used in the entire referenced document is predicted as a candidate to be input. It is an object of the present invention to provide a sentence creation support device, a sentence creation support system, and a program capable of improving the accuracy of prediction.

The present invention according to claim 1 is a part of the first document referred to by the user or a first portion referred to by the user when the user edits the first part of the first document. A predictive means that predicts a character or character string extracted from a part of a second document different from the document of the above as a candidate to be input when creating a new sentence, and a character or character predicted by the predictive means. A sentence creation support device comprising display control means for controlling a column to be displayed when the user edits the first portion.

According to the second aspect of the present invention, the display control means controls to display a part of the first document or the second document when editing the first part with reference to the first part. It is a sentence creation support device described in 1.

According to the third aspect of the present invention, the display control means obtains a character or a character string from a pre-divided part of the first document or the second document referred to by the prediction means. The sentence creation support device according to claim 1, which is predicted as a candidate to be input when creating a sentence.

According to the fourth aspect of the present invention, the predictive means predicts a character or a character string from the page of the first document or the second document referred to as a candidate to be input when creating a new sentence. This is the sentence creation support device according to claim 3.

According to the fifth aspect of the present invention, the predictive means predicts a character or a character string from the chapter of the first document or the second document referred to as a candidate to be input when creating a new sentence. The sentence creation support device according to claim 4.

The present invention according to claim 6 claims that the predicting means predicts a character or a character string of a document in an operable state as a candidate to be input when creating a new sentence. 1 There is a sentence creation support device described.

The sentence according to claim 6, wherein the prediction means predicts a character or a character string of a document that is not in an operable state as a candidate to be input when creating a new sentence. It is a creation support device.

In the present invention according to claim 8, the predicting means inputs characters or character strings extracted from the first document or a part of the second document referred to when creating a new sentence. The first prediction unit that predicts as a candidate to be input and the character or character string extracted from the first document or the entire second document that is referred to are candidates to be input when creating a new sentence. The sentence creation support device according to claim 1, which has a second prediction unit for prediction, and the first prediction unit has priority over the second prediction unit.

In the present invention according to claim 9, the predicting means inputs a character or a character string extracted from the first document or a part of the second document referred to when creating a new sentence. It has a first prediction unit that predicts as a candidate to be power, and a third prediction unit that predicts as a candidate to be input when newly created from a general dictionary, and the first prediction unit is the second prediction unit. The sentence creation support device according to claim 1, which has priority over the prediction unit.

In the present invention according to claim 10, the predicting means inputs a character or a character string extracted from the first document or a part of the second document referred to when creating a new sentence. The first prediction unit that predicts as a candidate to be predicted and the character or character string extracted from the first document or the entire second document that is referred to are candidates to be input when creating a new sentence. It has a second prediction unit for prediction and a third prediction unit for predicting as a candidate to be input when newly created from a general dictionary, and the first prediction unit includes the second prediction unit and the second prediction unit. The sentence creation support device according to claim 1, which has priority over the third prediction unit.

The present invention according to claim 11 has an extraction means for extracting characters or character strings from a document, a storage means for storing characters or character strings extracted by the extraction means, and characters or characters stored in the storage means. A character string, which is a part of the first document referred to by the user or the first document referred to by the user when the user edits the first part of the first document. Displays a predictive means that predicts a character or character string extracted from a part of a different second document as a candidate to be input when creating a new sentence, and a character or character string predicted by the predictive means. It is a sentence creation support system having a display control means for controlling the operation.

The present invention according to claim 12 is a part of the first document referred to by the user or a first portion referred to by the user when the user edits the first part of the first document. The step of predicting a character or character string extracted from a part of a second document different from the document of 1 as a candidate to be input when creating a new sentence, and the predicted character or character string of the first This is a program for causing a computer having a step of controlling the part to be displayed when the user edits it.

According to the present invention according to any one of claims 1, 11 or 12, when creating a new sentence by referring to one or more documents, the character or character string used in the entire referenced document is used. The accuracy of the prediction can be improved as compared with the one predicted as a candidate to be input.

According to the second aspect of the present invention, in addition to the effect of the present invention according to the first aspect, a part of the first document or the second document referred to can be displayed.

According to the third aspect of the present invention, in addition to the effect of the invention according to the first aspect, candidates to be input can be extracted from a predetermined portion referred to.

According to the fourth aspect of the present invention, in addition to the effect of the third aspect of the present invention, candidates to be input can be extracted from the referenced page.

According to the fifth aspect of the present invention, in addition to the effect of the present invention according to the fourth aspect, candidates to be input can be extracted from the referenced chapter.

According to the sixth aspect of the present invention, in addition to the effect of the present invention according to the first aspect, candidates to be input can be extracted from a document in an operable state.

According to the seventh aspect of the present invention, in addition to the effect of the present invention according to the sixth aspect, candidates to be input can be extracted from a document that is not in an operable state.

According to the eighth aspect of the present invention, in addition to the effect of the present invention according to the first aspect, the first document or the second document in addition to the first document or a part of the second document being referred to. Candidates to be input can be extracted from the entire second document.

According to the present invention according to claim 9, in addition to the effect of the present invention according to claim 1, in addition to the first document or a part of the second document referred to, input is also made from a general dictionary. Candidates to be extracted can be extracted.

According to the tenth aspect of the present invention, in addition to the effect of the present invention according to the first aspect, the first document or the first document or the second document in addition to the first document or a part of the second document being referred to. Candidates to be entered can be extracted from the entire second document and the general dictionary.

It is a block diagram which shows the sentence creation support system which concerns on embodiment of this invention. It is a block diagram which shows the hardware of the sentence creation support apparatus which concerns on embodiment of this invention. It is a block diagram which shows the function of the sentence creation support apparatus which concerns on embodiment of this invention. It is a flowchart which shows the operation flow of the sentence creation support apparatus which concerns on embodiment of this invention. It is a screen view which shows the display screen in the other embodiment of the sentence creation support device which concerns on embodiment of this invention. It is a figure which shows the content of the 1st document in the example to which this invention is applied. In an example to which the present invention is applied, the structure of the index is shown in a chart. It is a figure which shows the comment input in the example to which this invention is applied. It is a figure which shows the operation procedure in the example to which this invention is applied. It is a figure which shows the change of the display content with respect to the change of the display priority when only the index of an active document is used in the example to which this invention is applied. It is a figure which shows the change of the display content with respect to the change of the display priority when the index of the active document and the inactive document is used in the example to which this invention is applied.

Next, an embodiment of the present invention will be described in detail with reference to the drawings.
FIG. 1 shows the entire sentence creation support system 12 having the sentence creation support device 10 according to the embodiment of the present invention.

In the sentence creation support system 12, for example, client devices 14a and 14b, which are personal computers, are connected to each other via a network 16. The network 16 may be a local area network or the Internet.

The sentence creation support device 10 has a server 18 and a database 20 which is a large-capacity storage device. The database 20 constitutes a storage means in which a document and an index, which will be described later, are stored.

As shown in FIG. 2, the server 18 has a CPU 22, a memory 24, a storage device 26, and a network interface 28, and the CPU 22, the memory 24, the storage device 26, and the network interface 28 are connected via the bus 30. ..

The CPU 22 executes a predetermined process based on the control program stored in the memory 24. The storage device 26 is composed of, for example, a hard disk, and stores required software and data. The network interface 28 inputs / outputs data via the network 16 described above.

FIG. 3 is an explanatory diagram illustrating a schematic function of the sentence creation support device 10.
The sentence creation support device 10 has a document storage unit 32 and a document editing unit 34.

The document storage unit 32 has an index creation unit 36. The index creation unit 36 creates an index according to the description contents of the received document. The index is an index obtained by extracting a character string to be searched in advance from a document. This index has a table structure in which character strings, file locations, file modification dates, appearance frequencies, etc. are listed in a list format. The index created by the index creation unit 36 is stored in the database 20 described above together with the received document. The database 20 also stores edited documents. The received document and the edited document are stored in the database 20 as, for example, a tree structure.

The document editing unit 34 is a part that edits a document. Here, editing a document means processing the document based on a certain policy and purpose. For example, a part of the document is changed, deleted, added, or a comment is added to the document. It is an act of doing. In this embodiment, a comment is inserted in the document when the document is reviewed.

The document editing unit 34 has a display control unit 38 for displaying a document or the like. The display control unit 38 controls the documents stored in the database 20 of the document storage unit 32 so as to be displayed on the client devices 14a and 14b.

Further, the document editing unit 34 has a display page acquisition unit 40. The display page acquisition unit 40 acquires the displayed document and the target page information from the display control unit 38. The displayed document page is part of the document.

Further, the document editing unit 34 has an operation setting unit 42. The operation setting unit 42 sets the display priority and the index range. The display priority is the priority of input prediction, and as will be described later, it is prioritized in the order of the character string of the target page of the review target document, the character string of the review target document, and the character string of the general dictionary. , The priority can be changed from the setting screen. The index range is the range to which the index is applied, and includes, for example, the pages and chapters of the referenced document. The number of referenced documents is not limited to one, and may be plural. Further, the referenced document may be limited to a document in an operable state (hereinafter referred to as an active document), or a document that is not operable but is being viewed (hereinafter referred to as an inactive document). May be included. Similarly, the range of this index can be changed by the setting screen.

Further, the document editing unit 34 has an input prediction generation unit 44. The input prediction generation unit 44 indexes from the document storage unit 32 based on the page information of the displayed document acquired by the display page acquisition unit 40, the display priority information set by the operation setting unit 42, and the index range information. Is acquired, and the external dictionary unit 46 is acquired to perform input prediction. The input prediction generation unit 44 includes a first prediction unit that makes an input prediction from an index included in a part of a document such as a page or a chapter, a second prediction unit that makes an input prediction from an index included in the entire document, and an external device. It has a third prediction unit that predicts from a dictionary.

The comment receiving unit 48 receives comments from the client devices 14a and 14b. When the characters constituting the comment are input by the comment receiving unit 48, the display control unit 38 controls the display control unit 38 to display the predicted input candidates generated by the input prediction generation unit 44 on the client devices 14a and 14b. To. The edited document containing the comment is stored in the database 20.

The client devices 14a and 14b instruct the document storage unit 32 to store the document, and the document editing unit 34 to instruct the document display and operation setting, or input a comment. ..

FIG. 4 is a flowchart showing an example of the control operation of the CPU 22 described above.

First, the index creation will be described. In step S10, the document is stored in the database 20. In the next step S12, information necessary for index creation is acquired from the document stored in step S20. Information required for index creation includes character strings, file locations, file modification dates, frequency of occurrence, etc., but here, the minimum required information is information that identifies a document (for example, document ID), document. Page, character string.

In the next step S14, an index is created from the information acquired in step S12. In the next step S16, the index created in step S12 is stored in the database 20 and the process ends.

Next, document editing will be described. In step S18, the display control unit 38 acquires the document to be edited from the database 20 and the index associated with the document, and controls the client devices 14a and 14b to display the document. .. In the next step S20, the target page income unit 40 acquires the information of the page displayed on the display control unit 38. In step S22, the input prediction unit 44 acquires an index from the display control unit 38. Further, in step S24, the operation setting unit 42 acquires the set operation information, that is, the display priority and the index range.

In step S26, the input prediction unit 44 creates an input candidate from the index, the existing dictionary, the target page information, and the operation setting information. In the next step S28, the comment receiving unit 38 selects a character string from the input candidates and receives a comment sentence. In the next step S30, the comment receiving unit 38 sends the edited document containing the comment to the database 20. In step S32, the database 20 stores the edited document and ends the process.

In the above embodiment, the input prediction is made by the index extracted from a part (page or chapter) of the document to be edited (referred to as the first document), but the present invention is not limited to this.

That is, as shown in FIG. 5, the first document 50 has a comment input field 52 for the user to input a comment, the first document 50 is an active document, and the user can use the first document. Assuming that the second document 54 other than 50 is referenced, the second document 54 is an inactive document, but not only from the index extracted from the first document 50, but also from the second document 54. The extracted index may be included. Further, the first document 50 may display not only one page but also a plurality of pages, and input prediction may be performed from an index extracted from the plurality of pages. Similarly, the second document 54 may display not only one page but also a plurality of pages, and make an input prediction from an index extracted from the plurality of pages. Further, not only the displayed document is not necessarily the target of the index, but the index of the document referred to by the user until immediately before may be included in the target of the input prediction. For example, when the first document 50 being edited is displayed small, the user may operate to close the second document 54 once and display the first document 50 large, but the second document before closing. The index of the document 54 of the above may be included.

Next, an example to which the present invention is applied will be described.
As shown in FIG. 6, the first document 50 is composed of, for example, three pages.
Windows is a registered trademark of Microsoft Corporation.

The index shown in FIG. 7 is created for the first document 50. The index is composed of [Page | String], and only the examples necessary for explanation are described here.

For such a first document 50, with two pages of the first document 50 displayed, as shown in FIG. 8, a comment "Is the character code UTF8 only in Windows 2012 correct?" input. Here, the index extracted from the open page of the first document 50 in the index range of the operation setting unit 42 described above, the index extracted from the entire first document 50, and the character string from the general dictionary are mixed. It is assumed that it is set as.

FIG. 9 shows an operation when a comment is input.
First, in operation 1, it is in the initial state, and the comment input field is blank.
In the next operation 2, "W" is input. In the next operation 3, input candidates for "W" are displayed. For example, "★ Displaying" is displayed for the input candidates extracted from the open page, and the corresponding page is displayed for the input candidates extracted from other pages. Nothing is displayed in the input candidates extracted from the general dictionary. The display priority is displayed, other pages, and a general dictionary. Here, "Windows 2012" is displayed first.

In operation 4, the corresponding candidate is selected. The selected input candidates are highlighted, for example. Here, it is "Windows 2012".

In operation 5, the selected input candidate "Windows 2012" is input in the comment field.

In operation 6, "only character code" is input after "Windows 2012".

When "U" is input in the operation 7, input candidates corresponding to the "U" are displayed in the operation 8. "UTF8" extracted from the two displayed pages is displayed first, and "UTF16" extracted from other pages is displayed next.

In operation 9, the corresponding candidate is selected. Here, it is "UTF8".

In operation 10, the selected input candidate "UTF8" is input in the comment field.

In the operation 11, "is it correct?" Is input after "UTF8", and the comment is confirmed.

Next, a case of changing the display priority will be described.

In FIG. 10, the index range is limited to the first document (active document), and the input candidates (referred to as pages) extracted from the page displayed by the first document and the first document are displayed. A first example is shown in which the display priority of the input candidates (referred to as another page) extracted from the entire first document excluding the existing pages and the input candidates (referred to as general) extracted from the general dictionary is changed.

Specifically, when "W" is entered in the comment field, the character strings used in the entire first document are "Windows 2008", "windows 2012", and "Windows 2016", which are used on the target page. The character string is "Windows 2012", and the general dictionary is "Win" and "Windows".

Here, in FIG. 10, the display priority is changed to Page> Document> General, Page> General> Document, Document> Page> General, Document> General> Page, General> Page> Document, General> Document> Page. It shows the change of the display contents in the case.

In FIG. 11, the index range is defined as the first document (active document) and the second document (inactive), and in addition to the page, another page, and general, the input candidates extracted from the second document (separate). A second example is shown with the addition of (referred to as a document).

Similar to the first example, when "W" is entered in the comment field, the character strings used in the entire first document are "Windows 2008", "windows 2012", and "Windows 2016", and on the target page. The character string used is "Windows 2012", and the general dictionary is "Win" and "Windows". In this second example, the character strings used in the entire second document are "Windows 2003" and "Windows 2008".

Here, FIG. 11 shows the display priority of page> document> separate document> general, page> document> general> separate document, page> separate document> document> general, page> separate document> general> document, page> It shows the change of the displayed contents when it is changed to General> Document> Separate Document, Page> General> Separate Document> Document.

In the above embodiment, an index is created in advance and input candidates are extracted using the index. However, other methods may be used, for example, the target page or the target document is sequentially scanned when the user edits. And the input candidates may be extracted.

10 Sentence creation support device 12 Sentence creation support system 14a, 14b Client device 16 Network 18 Server 20 Database 22 CPU
24 Memory 26 Storage device 28 Network interface 30 Bus 32 Document storage unit 34 Document editing unit 36 Index creation unit 38 Display control unit 40 Display page acquisition unit 42 Operation setting unit 44 Input prediction generation unit 46 External dictionary unit 48 Comment reception unit 50th One document 52 Second document 54 Comment section

Claims (12)

When the user edits the first part of the first document, a part of the first document referred to by the user or a second document different from the first document referred to by the user. A predictive means that predicts a character or character string extracted from a part of the text as a candidate to be input when creating a new sentence.
A display control means for controlling the character or character string predicted by the prediction means so as to be displayed when the user edits the first portion.
Sentence creation support device with. The sentence creation support device according to claim 1, wherein the display control means controls to display a part of the first document or the second document when editing the first part with reference to the first part. The claim that the predicting means predicts a character or a character string from a preliminarily divided part of the first document or the second document referred to as a candidate to be input when creating a new sentence. 1 Described sentence creation support device. The sentence creation support according to claim 3, wherein the prediction means predicts a character or a character string from the page of the first document or the second document referred to as a candidate to be input when creating a new sentence. apparatus. The sentence creation support according to claim 4, wherein the prediction means predicts a character or a character string from the chapter of the first document or the second document referred to as a candidate to be input when creating a new sentence. apparatus. The sentence creation support device according to claim 1, wherein the prediction means predicts a character or a character string of a document in an operable state as a candidate to be input when creating a new sentence. The sentence creation support device according to claim 6, wherein the prediction means further predicts characters or character strings of a document that are not in an operable state as candidates to be input when creating a new sentence. The prediction means predicts a character or a character string extracted from the first document or a part of the second document that is referred to as a candidate to be input when creating a new sentence. A part, a second prediction part that predicts a character or a character string extracted from the first document or the entire second document that is referred to as a candidate to be input when creating a new sentence, and a second prediction part. The sentence creation support device according to claim 1, wherein the first prediction unit has priority over the second prediction unit. The prediction means predicts a character or a character string extracted from a part of the first document or a part of the second document that is referred to as a candidate to be input when creating a new sentence. A claim that includes a unit and a third prediction unit that predicts as a candidate to be input when newly created from a general dictionary, and the first prediction unit has priority over the second prediction unit. 1 Described sentence creation support device. The prediction means predicts a character or a character string extracted from a part of the first document or a part of the second document that is referred to as a candidate to be input when creating a new sentence. A part, a second prediction part that predicts a character or a character string extracted from the first document or the entire second document that is referred to as a candidate to be input when creating a new sentence, and a second prediction part. It has a third prediction unit that predicts as a candidate to be input when newly created from a general dictionary, and the first prediction unit has priority over the second prediction unit and the third prediction unit. The sentence creation support device according to claim 1. An extraction method that extracts characters or character strings from a document,
A storage means for storing characters or character strings extracted by the extraction means, and
A part of the first document or the user that is referenced by the user when the user edits the first part of the first document, which is a character or character string stored in the storage means. A predictive means that predicts a character or character string extracted from a part of a second document that is different from the first document referenced by, as a candidate to be input when creating a new sentence.
A display control means for controlling the display of characters or character strings predicted by the prediction means, and
Sentence creation support system with. When the user edits the first part of the first document, a part of the first document referred to by the user or a second document different from the first document referred to by the user. The step of predicting the character or character string extracted from a part of the text as a candidate to be input when creating a new sentence, and
A step of controlling the predicted character or character string to be displayed when the user edits the first part, and
A program to run on a computer that has.

Images