JP2680540B2 - Document layout method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001] BACKGROUND OF THE INVENTION The present invention provides a document structure for document data.
According to the hierarchical logical structure obtained by analysis, the document
To effectively support the output layout format of the data
The present invention relates to a document processing device that can be used. [0002] 2. Description of the Related Art Document processors such as word processors are used.
Is a system of code information such as character codes and punctuation marks.
Document data is input as a column. And that code
Register document data represented by a series of information in a document file
Or output to a printer or display. However, the character string indicated by the code information is
The document is very difficult to read if you just output it.
No. So, in general, a group of sentence breaks
Inserts a line break code at the position and the next point after the line break.
Insert a space code on the head and add a paragraph,
The document format is adjusted. Furthermore, the entire document is divided into a plurality of chapters and sections, for example.
Divide into the range of, and add a heading for each group,
To make the document easier to read, tabs, indents, etc.
The control code of is also inserted. [0005] [Problems to be Solved by the Invention] However, the original document data
-There is no direct relation to the data, such as the line feed code mentioned above
Creating a document while inserting the control code of
This hinders the thinking of the document and causes a decrease in document creation efficiency.
ing. Further, the document created in this way is edited.
When re-doing, delete the control code mentioned above or
It is necessary to insert it in another place. this
At this time, the document structure is controlled by the control code inserted in the document data.
Due to changes in construction, for example, several pages of documents
When editing, unify the document structure as a whole
Had problems such as requiring a lot of labor. For example
To unify the document formats, see the document formats several pages before.
Procedures such as lighting are required. Therefore, you can easily
It is impossible to proceed with the editing process and the processing efficiency is improved.
I couldn't hope for it. The present invention has been made in consideration of such circumstances.
The purpose of the document is the hierarchical structure of the document.
By actively using the logical structure for document processing
To eliminate the annoyance of document processing described above
Providing a document processing device that enables call processing
is there. [0008] According to the present invention, an input sentence
The sentence that becomes the heading from the calligraphy data and the sentence that follows this heading sentence
And a headline sentence determined from the document datasame
MasterBy sequentially obtaining the relationship between the headline sentences,
At least one of the heading sentence and the sentence following the heading sentence
The requested headlineOne anotherLayout according to the relationship of
Output the above document data as a formatmeansTo expand to
doing. [0009] [0010] [Action]As a result, according to the present invention, the input document data
The sentence that becomes the heading from the data and the sentence that follows this heading sentence
Determine. Then, for this determined headline sentence,
Compare the headline sentences and find the relationship between them.
Confuse. The headline sentence and the number of sentences following the headline sentence are small.
At least one of them, depending on the desired heading sentence companionship
The document data is expanded after the layout format is changed.
I'm trying. This allows the document data output layout
Do not refer to the relationship between headline sentences
However, the layout can be determined automatically, and the layout according to these results
Can be displayed on the display or printed out depending on the format
You. Therefore, at the time of inputting document data, or its edition
Consider the output layout format one by one when collecting data.
Not, that is, considering the insertion position of the line feed code,
Process the document without setting dents or tabs
It becomes possible. Moreover, in this way
However, the document data has a predetermined layout.
Expanded to format and displayed or printed
Will be empowered. [0012] Embodiments of the present invention will now be described with reference to the drawings.
explain. FIG. 1 is a schematic configuration diagram of the embodiment apparatus. FIG.
In FIG. 1, reference numeral 1 is a document management unit that constitutes the main body of the apparatus. Keyboard
-A series of code information via the input unit 2 composed of code etc.
The document data input as is, for example, a sentence as shown in FIG.
It has a document structure and is under the control of the document management unit 1 described above.
Is stored in the original text storage unit 3 and is used for document processing. Soshi
The document data processed by the document management unit 1 is displayed.
Under the control of the control unit 4, it is displayed on the display unit 5.
ing. FIG. 3 shows a document structure solution for input document data.
It shows the analysis processing procedure, and along with this flow
The function of the device will be described. Is the document management unit 1 the input unit 2?
Enter the document data (step a),
The delimiter code in the document data is stored in the storage unit 3.
-For example, a line feed code is detected and this delimiter code is
A group of sentences that are further separated is extracted in order. same
Sometimes the length of one sentence is measured. And extracted 1
Manages and controls the execution of the following processes in units of statements
ing. The headline extraction unit 6 is extracted by the document management unit 1.
Check whether there is a possibility that the one sentence
Information on the length of one sentence measured as described above, and the entry word dictionary
The determination is made by referring to 6a (step b). This headword
The dictionary 6a is a word or symbol that frequently appears as a headline.
For example, as shown in FIG.
It is registered in advance. Specifically, it appears as a headline
Frequent words such as "Introduction" and "Synopsis"
Are collectively registered in the category of "heading reserved word", and
Numbers and symbols that appear frequently as headings
Are registered together for each category. The headline extraction unit 6 determines that the length of the extracted sentence is
Determine whether it is within a predetermined number of characters (for example, 40 characters)
If the number of characters is within the specified number, the possibility of heading
It is determined that there is. And about this sentence, that sentence
(Words, numbers and symbols shown in the code information sequence)
It is searched whether it is registered in the word dictionary 6a (step
B), the corresponding word / phrase is registered in the headword dictionary 6a.
If this is the case, this is the heading candidate (step
C). Therefore, the headline judging section 7 determines the headline rule.
5, 6 and 7 (a) stored in the rule dictionary 7a
Based on the heading rule as shown in (b), the heading extraction
The sentence extracted as a heading candidate in part 6 is the heading rule
Is determined (step c), and
If the heading candidate matches the heading rule,
It is determined that it is a headline sentence (step e). In addition, before
The sentence in which the entry word is not detected by the entry extraction unit 6,
And even if the sentence is judged as a heading candidate,
Sentences that do not meet the issuing rules are judged not to be headline sentences
(Step f). In other words, it is the text of a document such as a paragraph.
Is determined. The document structure determination unit 8 finds the index as described above.
For each sentence determined to be a sentence or other sentence
Stored in the document structure rule dictionary 8a, for example, in FIG.
Document structure rules as shown in FIGS. 9 and 10A and 10B
The sentence is a chapter heading or section heading
The logical structure of the document such as
(Step g). This is determined by the document structure determination unit 8.
The logical structure of each of the above sentences (step h)
Logical structure memory by associating physical structure information with each sentence
It is stored in the section 10 (step i). Note that the logical structure analysis based on the document structure rule is lost.
Regarding the lost sentence, it is said that there is an error in its logical structure.
Error such as urging correction of input document data.
Processing is performed (step j). The document expanding section 9 is for the above document structure analysis processing.
Hierarchical logic for input document data obtained by
According to the structure, each sentence that is input according to its logical structure
It is expanded to the output part in the specified layout format. That is,
The layout rule dictionary 9a of the document expansion unit 9 stores its logic.
Determine the output layout format determined according to the type of structure
The layout rule to be specified is stored. Document development unit 9
Is the logic obtained for each sentence as described above.
The document data of each sentence is displayed on the display unit 5 according to the structure.
The above-mentioned layout rule dictionary 9 is used to show the layout rules for indicating.
a, and the document data of each sentence according to the layout rule.
The data are expanded in the display control unit 4. The display control unit 4 is developed in this way.
Of the document data described above according to the development structure of the document data.
The display on the display unit 5 is controlled. Hiding
According to the device configured as
The hierarchical logical structure of the document data is analyzed. When the document data is input from the input unit 2,
When the document data is sequentially stored in the original text storage unit 3,
Then, the document management unit 1 performs a division process. This break
The reason is that the input code information is line feed code and space code.
Or is it a delimiter such as "..." ";" ":"
The input code information is judged by these delimiter codes.
This is done by cutting out a series of reports for each sentence. This
In this case, the code information separated by the above delimiter code
The length of the sentence (character
Number) is measured. Now, the document data shown in FIG.
To explain the case where it is entered, the line feed code
The first line sentence "Document structure understanding system" separated by
And given the sentence "Taro Okawa" on the second line,
For each of these sentences, the corresponding phrase is found in the entry word dictionary 6a.
Since it is not registered, the headline extraction section 6
It is determined that it is not a headline. The document structure determination unit 8
For example, the sentence in the first line is a noun phrase that appears at the beginning of the document.
Since it conforms to certain rules, its attribute is the title.
It is judged as Also, the sentence on the second line is unique
A noun, especially a proper noun indicating a person's name, that appears after the title
It is judged as the author's name according to the rules such as
You. Then, the sentence "1. Introduction ”
Is given, "1" ". 』
The word "Introduction" is changed from the entry word dictionary 6a.
Each is found. As a result, this sentence is heading candidate A.
And the categories that make up the heading candidate at the same time.
Gori says “(number part) (postfix part) (heading reserved word)”
Required. Then, the headline determination section 7 determines that this headline candidate
The sentence structure judged as A conforms to the heading rules
Whether or not to search is checked by referring to the index rule dictionary 7a. this
In the heading determination process, first, the categories that form heading candidate A are
The sequence of gorigo is analyzed, and the analysis structure is as shown in FIG.
It is determined whether or not the conditions for delivery are satisfied. this
In the case of the above category “(number part) (postfix part)
About) is shown in FIGS. 5, 6 and 7 (a) (b).
According to the rule, the heading pattern is analyzed as shown in FIG.
It is confirmed that the
Is determined. It should be noted that the above categorization is performed by this determination processing.
The arrangement of the lines is shown in FIGS. 5, 6 and 7 (a) (b).
If it is determined that none of the
It is determined that the heading candidate is not a heading. When the above-mentioned heading B is obtained,
In addition, the document structure determination unit 8 has the document structure shown in FIGS.
And which of the rules illustrated in FIGS. 10 (a) and 10 (b)
It is determined whether to do it. In this case, ever analyzed
As described above, the logical structure of the sentence is “title” and “author name”.
And it is shown in Fig. 8 because the chapter headings have not appeared.
Matched with heading rules. The above sentence by this collation
"1. The headline candidate "Introduction" is the rule article shown in Fig. 8.
Items (1) (1,1) (1,1,1) (1,1,1,1)
Is found to match the heading “1. "Introduction"
It is uniquely determined that the chapter heading C is formed. So
Then, the information of the logical structure is stored in the logical structure storage unit 10.
Is done. Then, the sentences on the 4th to 5th lines are input.
If the number of characters is
Since it exceeds the number of characters, it is judged as a sentence other than the headline
Is done. And in this case, the sentence is shown in FIG.
A paragraph is constructed because it conforms to the document structure rules shown below.
It is determined that the sentence is Thereafter, similarly, the division code is used for division.
It is determined whether the cut sentence is a headline sentence, and
The document structure is sequentially obtained by collating with the construction rule dictionary 8a.
It is. For example, the sentences forming the headline are input again.
Then, it is detected as the heading candidate A in the same manner as the previous example.
FIG. 5, FIG. 6 andFigureHeading rule shown in 7 (a) (b)
It is determined that the rule is met. And in this heading B
Regarding the contents of the logical structure storage unit 10,
Since it is shown that the protrusion is detected, first, in FIG.
Referring to the document structure rules shown below, the pattern of heading B above
Find out which conditions apply. And the headline candidate
Pattern is condition (1,1) (2,1) (3,1)
If it falls under (4,1), find the chapter found earlier.
It is possible that the heading is the same level as C
Is determined. After that, whether it complies with the rules shown in FIG. 8 above
The condition (1) of the document structure rule shown in FIG. 8 is checked.
(1,1) (1,1,2) (1,1,2,2) (1,
1, 2, 2, 1), the headline is already
The chapter heading of the same level as the previous chapter heading judged by
To determine.That is, here is the finding that was sought before
Always refer to the Shino level and differ from the headlines that appeared earlier
If it is a heading of a type, it is determined as a new level heading.
Defined and is the same type of headline that came out earlier
You'll come to a decision with the same level of headlines. The headline with a circled number at the beginning gives
If a similar pattern heading was issued before that,
Matching with the rules (conditions) shown in Fig. 9 because they have not been detected
Matching is unsuccessful. As a result of this,
It is determined that the
It is. After that, in the collation with the document structure rule shown in FIG.
Therefore, for example, condition (1) (1,1) (1,1,2)
(1,1,2,2) (1,1,2,2,2) (1,1,
2, 2, 2, 1) is detected and the headline
Judged as a bulleted heading. When the sentence is determined to be a paragraph
Tells you what level of heading the paragraph received.
I may not understand. In such a case, for example,
Refer to the rules shown in 10 (b), and connect paragraphs and headings.
You should determine the continuity relationship and set the level.
No. By such document structure analysis processing,
The document structure of each sentence that makes up the input document data of
The information of the hierarchical logical structure that is obtained is input document data.
For example, as shown in FIG.
It is stored in the manufacturing storage unit 10. In the logical structure data of the document shown in FIG.
To explain briefly, the information of the document is [].
Corresponding to the document data shown in the box, {}
It is surrounded by, and the numerical value at the beginning is the hierarchical level of the sentence,
The following information describes the sentence attribute and data starting with the symbol z.
Shows the analysis result of document data. This document data
The data analysis results have the following meanings, for example. Here, z1 is a symbol part, and z2 is an alphanumeric bracket.
At the beginning, z3 is a reserved word part, z4 is an alphanumeric part 1, z5 is an alphanumeric character.
The letter part 2, z6 is the end of alphanumeric brackets, z7 is the trailing part 1, ...,
z21 is a reserved word for heading, and z22 is the starting position of the main heading body.
Position, z23 is the end position of the main body of the main heading, and z24 is the rear heading.
Symbols, ~, z28 is the start position of the content part, and z29 is the end position of the content part.
It is a position. After a symbol having such a meaning, the sentence
Information about what the writing data is or what letter character
The analysis result is expressed by adding. Well, the document exhibition
The opening 9 is for the document data obtained as described above.
According to the information of the hierarchical logical structure, the document data is as follows.
It is being developed in the output section. FIG. 13 shows the hardware structure of the document expanding section 9.
It shows. The logical structure data reading unit 11
As shown in FIG. 12 stored in the logical structure storage unit 10.
It reads the document data and its logical structure data.
The document data of is cut out for each character by the one-character cutout unit 12.
It is provided to the document data allocation unit 13. On this occasion,
The sentence attribute extracting unit 14 displays the "titl" shown in FIG.
The sentence attribute data such as "e" and "auther" is extracted and extracted.
The layout information is given to the output information detection unit 15 again.
The information analysis unit 16 is described with the above-mentioned z symbol attached.
Read out the logical structure information. In the layout rule dictionary 9a, for example,
For example, as shown in FIGS. 14 (a) and 14 (b), the document data is output.
According to the information about the frame (document frame)
From the information on how to output the sentence in layout
The following layout rules are stored. Specifically above
The frame information is displayed as shown in FIG.
Frame management number on the output device such as
Output position information and frame size information
The display image from which the document data is output.
The frame or the composition of the frame on the printing paper is, for example,
It is defined as shown in FIG. these
Frame information is set to output document data.
It is defined for each of a plurality of frames. The layout rule reading unit 17
Reading the frame rule from the layout rule dictionary 9a
It is stored in the frame format information buffer 18 and
Information is given to the arithmetic unit 19 and the maximum column of the frame
I want the number and maximum number of lines. This maximum column
The maximum number of columns and line number data
Expansion output control of character data set to 20 and described later
Used for Also stored in the layout rule dictionary 9a.
The layout rule relating to the sentence attribute is, for example, as shown in FIG.
It is defined as shown in (b). To this layout rule
Therefore, the document data for each type of sentence attribute is
How to layout on a frame defined as
Whether to do it is regulated. Here, the target character string is, for example, "lf".
Force a line break at the end of, the target character in "Ce"
Centering the columns is shown respectively. Also
One line after outputting the target document data with "ds; 1"
Is indicated as a line break, and the target document is indicated by "us; 1"
It is indicated that a line break for one line is performed before outputting the data.
You. Furthermore, "ls; 1" is 1 character from the left side of the frame.
It is shown that a margin (blank) is provided. In addition, paragraph
In the case of, the first character of the first line is blank. Such a layout rule is used for the layout.
Under the control of the text information detection unit 15, according to the type of the sentence attribute.
Is read from the layout rule dictionary 9a,
It is given to the layout information analysis unit 16. The layout information analysis unit 16
Document data allocation based on layout rules required by
The text that controls the text part 13 and is cut out character by character as described above.
It develops the calligraphy data in order to the specified frame. That is, the layout information analysis unit 16 is
A frame that develops document data according to the information in the frame
And the size of the frame at the same time.
You. Then, according to the specified frame, the sentence attribute is applied.
Then, the document data is developed according to the layout rule.
I control At this time, the document data is extracted by the one-character cutout unit 12.
Each time the data is cut out character by character, the column is synchronized with this.
・ Count value of line counter 21 is incremented
You. The count value by the column / line counter 21 is
The maximum column / line value buffer 20 in the comparison unit 22
It is being compared successively with the value set in, and the comparison result
Is determined by the determination unit 23. In this determination unit 23
The layout of the characters cut out by the one-character cutting section 12
Is the position within the frame specified at that time?
It is determined whether or not. Then, the sentence which is layout-controlled and expanded
The assigned position of the write data exceeds the specified frame
When it is determined that the frame is switched, the frame switching unit 24 is activated.
Then, the document data by the document data assigning unit 13
So that the allocation frame of the next frame is updated and controlled to the next frame.
It has become. In this way, the document expanding unit 9
The layout rule set in the out-rule dictionary 9a
Therefore, set the frame where the document data should be output.
ing. Then, the sentence attribute is added to the set frame.
According to the layout rules set according to
-We are deploying data. For example, the logical structure obtained as shown in FIG.
The first sentence is the title in the created document data.
Layout rules are indicated by "lf", "Ce", "ds; 1"
Therefore, the layout information analysis unit 16 first
From the maximum column / line value buffer 20
The maximum number of columns and lines of the frame, for example (M
AX line-the number of characters in the target sentence) / 2, etc.
It seeks the character position to be displayed by tiling. This
Of the column / line counter 21 according to the data of
The numerical value is changed in a preset manner, and the
The expansion of the document data of the target sentence, that is, the assignment of the character position
Done. After that, output allocation of character data of the target sentence
When is finished, it is forced according to the information "ds; 1"
A line of text is displayed below the character string indicating the title that has been output on a new line.
A blank line is set. Output (development) control of such document data
However, for each sentence cut out as described above,
Sequentially according to the obtained logical structure, the document data
The data will be expanded in order in a predetermined frame. As a result,
For example, as illustrated in FIG. 16, as illustrated in FIG.
Document data that requires a hierarchical logical structure
Output is expanded according to the specified layout rule.
And Thus, according to the device thus constructed,
For example, the hierarchical logical structure of the document is obtained from the input document data.
And a layer of document data that follows the hierarchical logical structure
The document data is set according to the
A layout that is set according to the sentence attributes of each
It will be expanded in the format. Specifically, the document data is written in the defined frame.
Data is assigned in sequence, and the hierarchical logical structure of the sentence is
Follows the document structure information of each sentence, that is, according to its sentence attributes.
For example, control such as forced line feed and centering is performed.
Document data is developed. The document processing at the time of document creation and document editing
ReasonHeadline sentences and their headings that are identified from document data
For the sentence following the sentence, it is calculated by comparing the headline sentences with each other.
Output document data in the layout format according to the relationship
Since it is designed to be able to be deployed in stages,Layout format
Efficiently focusing only on the document content without consideration
Can be executed. Moreover, in this way
Even if it makes sense, the hierarchical document structure analyzed as described above,
And the layer specified by its document structure (sentence attribute)
Output layout form of the document data according to the rule
The expression is automatically controlled, so the overall statement is uniform.
It is possible to create the document in written form. Therefore, in order to control the output format of the document, the document creation
Easy to process documents without disturbing thoughts such as success
It is practically useful in that it can be performed easily and effectively.
Great effect can be achieved. The present invention is limited to the above-mentioned embodiments.
Not something. For example, data format showing the logical structure of the document
And rules for obtaining the logical structure, and further document data
The layout rule that determines the output format of
It is good to set it. Also font (character font)
Extraction of headings in consideration of transformation of documents and judgment process of document structure
Alternatively, the output control may be performed. this
In case of high-intensity display and reverse display of output characters,
Of course, it is also possible to transmit such control information.
It is enough to set it as an out rule. In the above-mentioned example, the document data is hierarchical.
After determining the logical structure, control the output layout format.
I did, but even if I did the layout expansion process for each sentence
good. In this case, do not store the logical structure once.
Even if it is used directly for layout expansion processing,
good. Furthermore, multiple layouts for one sentence attribute
Prepare rules and select layout rules from them.
It is also possible to apply the rules. This makes the same sentence
Layout formats for different uses depending on gender, such as academic papers
Can be converted to formulas and in-house report formats, etc.
It is possible to easily edit the layout. Also
Although Japanese was used as an example here, the same applies to other languages.
It is needless to say that the above can be applied. Others
Akira implements various modifications without departing from the scope of the invention.
be able to. [0055] As described above, according to the present invention,Sentence
Following the headline sentence and the headline sentence that is determined from the calligraphy data
The relationship obtained by comparing the headline sentences with each other
Expand the document data to the output means in the layout format according to
Because it is possible, consider the layout format for document processing.
Can be run efficiently withoutDocument publication
Control of force form does not interfere with the thought process.
In addition, it is possible to process the document effectively.
It has a great effect in practice.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic configuration diagram of one embodiment of the present invention. FIG. 2 is a diagram showing an example of input document data. FIG. 3 is an exemplary flowchart showing a flow of a document structure analysis process in the apparatus according to the embodiment. FIG. 4 is a diagram showing an example of a headword dictionary. FIG. 5 is a diagram showing a configuration example of a heading extraction rule dictionary. FIG. 6 is a diagram showing a configuration example of a heading extraction rule dictionary. FIG. 7 is a diagram showing a configuration example of a heading extraction rule dictionary. FIG. 8 is a diagram showing a configuration example of a document structure rule dictionary. FIG. 9 is a diagram showing a configuration example of a document structure rule dictionary. FIG. 10 is a diagram showing a configuration example of a document structure rule dictionary. FIG. 11 is a diagram showing an analysis structure of a headline sentence. FIG. 12 is a diagram showing an example of document structure information stored in a logical structure storage unit. FIG. 13 is a diagram illustrating a configuration example of a document development unit. FIG. 14 is a diagram showing a configuration example of a layout rule dictionary. FIG. 15 is a diagram showing an example of a frame (document frame) indicated by a layout rule. FIG. 16 is a diagram showing an example of a document format output after layout development processing. [Description of Signs] 1 ... Document management unit, 2 ... Input unit, 3 ... Original text storage unit, 4 ... Display control unit, 5 ... Display unit, 6 ... Heading extraction unit, 6a ... Headword dictionary, 7 ... Heading determination unit , 7a ... heading rule dictionary,
8: Document structure determination unit, 8a: Document structure rule unit, 9: Document development unit, 9a: Layout rule dictionary, 10: Logical structure storage unit.

   ────────────────────────────────────────────────── ─── Continuation of front page    (72) Inventor Toshio Okamoto               1 Komukai Toshiba-cho, Saiwai-ku, Kawasaki-shi, Kanagawa               Toshiba Research Institute, Ltd.                (56) References JP-A-60-24622 (JP, A)                 JP 60-17522 (JP, A)                 bit, Vol. 15, No. 5 Kyoritsu               Publishing Co. P. 44-50                 Coupling Survey               s, Vol. 14, No. 3 (Sep.               1982) P. 417 ~ 472

Claims (1)

(57) [Claims] A sentence that becomes a headline from the input document data and a sentence that follows this headline sentence are discriminated, the headline sentences discriminated from the document data are compared with each other, and the relation between the headline sentences is sequentially obtained. document layout wherein the deploying the and output means said document data to at least one of the statements following the該見out statement as a layout format corresponding to the relationship caption between obtained above.