JP3074191B2 - Document processing device that inspects the table of contents - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document processing apparatus for inspecting a table of contents for performing a table of contents of a document prepared by a document processing apparatus such as a word processor.

When writing a manual or the like, what kind of information to collect at the design stage before writing,
It is necessary to examine the order of explanation. When this design stage is completed, a draft of the table of contents is almost completed, and the assignment of writing etc. is determined based on the table of contents. For this reason, it is necessary to make use of the first draft of the table of contents, and to understand how much the written manuscript differs from the first draft of the table of contents.

[0003] By reconsidering the differences, it is possible to judge which of the original table of contents and the written manuscript are more rigid, and to help improve the manuscript such as a manual. .

[0004] In addition, despite the fact that a manuscript is prepared in accordance with the table of contents, there are many cases where the term of the heading is different or the item number is shifted due to an input error or the like. In the method of automatically creating a table of contents from a manuscript according to the related art, there is a risk that the table of contents may be used as it is without correcting errors caused by erroneous entry of heading terms. If you know how much the written manuscript differs from the table of contents, you can also find differences due to typographical errors and prevent such dangers.

[0005]

2. Description of the Related Art Conventionally, in a document creating apparatus, a symbol for heading recognition is inserted in a manuscript in advance, and a portion thereof is extracted to create a table of contents, or a chapter, section, section, subsection, etc. A technique is used in which the format of a heading is determined (for example, a chapter heading starts to be written in “Chapter” ”), and the table of contents is automatically created by recognizing the format.

However, it has not been possible to automatically check whether or not a digitized manuscript has a table of contents, and whether or not an existing table of contents matches a headline in the text. Therefore, when examining whether the table of contents matches the heading of the text, it is necessary for a human to check each item visually by displaying the digitized original on a display or printing it with a printer. Was needed.

[0007]

[0006] As the review and correction of a manual or the like are repeated, the headings of the table of contents may not match the headings of the text. That is, if the headings in the table of contents do not match the headings in the text,
Although there is a table of contents, there is no headline in the text,
Conversely, it often happened that it was in the text but not in the table of contents.

In the case of examining such a problem, there has been a problem that a great deal of labor and time are required since a human had to perform the above. In addition, there is a problem that inspection leakage may occur. SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned problems and to provide a means for automatically checking and indicating a mismatch between a heading in a table of contents and a heading in a text using an electronic manuscript. I have.

[0009]

FIG. 1 is a block diagram showing the principle of the present invention. In FIG. 1, 10 is an external storage device for storing document information, 11 is an electronic manuscript created by a document creation function of a computer, 12 is a processing device including a CPU and a memory, 13 is a headline extraction processing unit, 14 is a table of contents presence / absence check processing unit, 15 is a table of contents page judgment processing unit, 1
Reference numeral 6 denotes a table of contents / text comparison / inspection processing unit, and reference numeral 17 denotes an output result indicating coincidence or non-coincidence between the heading of the table of contents and the heading of the text.

The heading extraction processing unit 13 reads a record of the digitized manuscript 11, checks whether there is a heading such as a chapter, a section, a section, a subsection, or a heading of a figure or a table. A process for extracting the record including the record is performed.

The table-of-contents presence / absence check processing unit 14 checks whether or not the same headline exists in the record group of the headline extracted by the headline extraction processing unit 13, and, based on the result, the digitized original 11 to be checked. A process for checking whether or not there is a table of contents is performed.

When the table of contents presence / absence check processing unit 14 determines that there is a table of contents based on the inspection result, the table of contents page determination processing unit 15 checks how much the table of contents occupies and separates the table of contents from the table of contents. is there.

The table of contents / text comparison / inspection processing section 16 performs a comparison test between the table of contents index and the table of text, and outputs an output result 17 indicating the position of a mismatched part when there is a mismatch.

[0014]

The heading extraction processing unit extracts a record group including a heading, the table of contents presence / absence checking processing unit automatically checks presence / absence of a table of contents, and the table of contents page determination processing unit 15 separates the heading from the text. Since the table of contents / text comparison / inspection processing unit 16 checks the difference between the table of contents and the table of contents, it is possible to automatically obtain the desired result only by specifying the first document to be checked.

In particular, since there is no need to specify the range of the table of contents in the digitized manuscript 11 or to point out one by one whether or not the heading is the target of the table of contents, the operation is extremely difficult. Easy.

[0016]

FIG. 2 is a flowchart showing a process of an embodiment of the present invention, FIG. 3 is a diagram showing an example of a headline extraction process according to an embodiment of the present invention, and FIG. FIG. 5 is an explanatory diagram of an inspection process, FIG. 5 is an explanatory diagram of a table of contents page determination process according to an embodiment of the present invention, and FIG. 6 is a diagram illustrating an example of an output result according to an embodiment of the present invention.

An embodiment of the present invention will now be described with reference to FIGS. 3 to 6 in accordance with the flowchart shown in FIG. [1] The designated digitized original 11 is input by the processing 20 shown in FIG.

[2] Next, by processing 21 shown in FIG.
As shown in FIG. 3, a heading record group 30 is created from the input electronic document 11. Here, from the digitized manuscript 11, a record having a heading such as a chapter, section, section, subsection, or a heading of a figure or table is selected. Then, a record is created by adding a symbol, such as a flag, which can determine which of the headings such as chapter, section, section, subsection, figure, table, etc. has been recognized, and the number of pages where the heading actually exists.
Arrange the created records in the order extracted from the manuscript,
The heading record group 30 is set.

As shown in FIG. 3, each record of the heading record group 30 includes, for example, a heading type flag 31 indicating the type of heading, and a comparison source record to be used later or a comparison destination record. Flag 32 for identifying a record, and a flag 33 indicating the presence or absence of the same heading indicating whether there is another record having the same heading.
Are provided.

[3] Next, it is checked whether or not the original has a table of contents by the processing 22 shown in FIG. The checking method is as follows. a. First, the head record is taken from the heading record group 30, and is set as the comparison source record.

B. Records after the comparison source record are regarded as a comparison destination record group. In the comparison target record group, a record having the same heading type flag 31 as the comparison source record and having no same heading presence / absence flag 33 is set as a comparison destination record. Then, the item number and the item name of the comparison source record and the comparison destination record are compared. At this time, blanks are not included in the comparison. If the comparison result shows that the comparison source record and the comparison destination record match, it is determined that the same heading exists, and the presence / absence flag 33 of the same heading is set for both the comparison source and the comparison destination records.
Also, a comparison source / comparison destination flag 32 for distinguishing between the comparison source record and the comparison destination record is set.

However, if the same index presence / absence flag 33 of the comparison source record has already been set, this flag (the same index presence / absence flag 33) is set in the comparison destination record. Also, a comparison source / comparison destination flag 32 for distinguishing between the comparison source record and the comparison destination record is set.

The comparison source / comparison destination flag 32 is used later to identify whether the record is a table of contents record or a body record. The above comparison is performed up to the end of the record group of the comparison destination.

C. When the processing of b is completed for one comparison source record, the next comparison source record is determined.
To do this, the next record in the current comparison destination record group that does not have the same heading flag 33 (the same heading has not yet been found) and is closest to the comparison source record is compared. Original record.

D. The above processes b to c are repeated until there is no more record to be compared. FIG. 4 shows an example of the above processing. As shown in FIG. 4A, the next record after the checked record is set as the comparison source record, and it is determined whether or not there is the same heading one by one with the subsequent comparison destination record group. I will investigate. When the check is completed, the presence / absence flag 33 of the same headline and the comparison source / comparison destination flag 32 are set as shown in FIG.

As a result of performing the above-described processing, if a flag indicating that all records have the same heading (the same heading presence / absence flag 33) is not set, it is regarded that the document is a text-only original having no table of contents.

If the flag indicating that the same heading is present is set, there is a possibility that both the table of contents and the text exist, so the determination is made by the following processing. a. The number of records in which the comparison source / comparison destination flag 32 is set as a comparison target record flag and the same heading presence / absence flag 33 is set as a flag indicating that the same heading is present is counted. Let the number be A.

B. A record whose comparison source / comparison destination flag 32 has a flag of a comparison destination record, and
Counts the number of records that have the same heading flag 33 in the same heading flag 33 and the flag that the heading type flag 31 has a section, figure, figure, table, table or subheading. I do.
Let the number be B.

C. If A = B, it is regarded as a manuscript consisting only of a text without a table of contents. If A ≠ B, it is considered that the manuscript has both the table of contents and the text. Section, figure, appendix,
For tables, appendixes, and subheadings, the same headline may appear many times, even if the manuscript consists only of the text. A section is sometimes used multiple times when there is a chapter door, and a figure or table is used multiple times when there is a figure or table that spans multiple pages. Also, the same headline may be used many times.

Here, consider a manuscript having only a text page. If the manuscript is only a text page, there is no heading for the same chapter or section. On the other hand, even in a manuscript containing only the text, the same heading may appear many times in sections, figures, attached figures, tables, attached tables, and subheadings, and these are the records to be compared. That is, in the case of a manuscript consisting only of a text page, the number A of records to be compared and the number B of sections, figures, attached figures, tables, attached tables, and subheadings in the compared records match. Therefore, in the case of A = B, it is regarded as a manuscript of only the text without a table of contents.

[4] For a manuscript that is considered to have both a table of contents and a text, the following procedure is used to check how much the record of the table of contents occupies in the record group by the process 23 shown in FIG. Of the records in which the heading type flag 31 has a flag indicating that it is a chapter heading, the head record is the inspection source record. Records that are located after the inspection source record and for which a flag indicating that the heading is a chapter heading in the heading type flag 31 are set as inspection destination records.

It is checked whether the chapter number of the inspection source record matches the chapter number of the inspection destination record.
If the chapter number of the record of the inspection source and the chapter number of the record of the inspection destination match, the record group of the heading preceding the record of the inspection destination is regarded as the table of contents, and the record group after the inspection destination record is regarded as the text. . From this result, for example, as shown in FIG.
A flag (◎) indicating a headline of the text is set for each record group. Then, the presence / absence flag 33 of the same heading of all records is cleared.

When the chapter number of the record of the inspection source and the chapter number of the record of the inspection destination do not match, in the records after the inspection destination record, a flag indicating that the chapter is a chapter heading is set in the heading type flag 31. The standing record is set as the next inspection destination record, and the above processing is performed. The above processing is repeated until the chapter number of the inspection source record matches the chapter number of the inspection destination record.

[5] Next, by processing 24 shown in the figure, it is checked whether or not the heading in the table of contents matches the heading in the text. Follow the procedure below. The record with the flag of the index of the table of contents set as the inspection source record, and the record with the heading of the text flag set and having the same heading type flag 31 as the inspection source record as the inspection destination record .
It checks whether the heading (item number and item name) of the inspection source record matches the heading (item number and item name) of the inspection destination record, one character at a time. At this time, blanks are not subject to inspection.

If a match is found,
A flag indicating that there is a match is set in the presence / absence flag 33 of the same heading in both the inspection source record and the inspection destination record. Then, the record next to the record of the inspection source is set as the inspection source record, and the record with the flag of the headline of the text and having the same heading type flag 31 as the inspection source record is set as the inspection destination record. The inspection of the above processing is repeated. If they do not match, the inspection source record is left as it is, and a record after the inspection destination record and a record having the same heading type flag 31 as the inspection source record are processed as the inspection destination record. .

Even if the record to be inspected reaches the end of the record group flagged as the headline of the text, if no record matching the record of the inspection source is found, The record next to the record is set as a new inspection source record, and a record having the same heading type flag 31 as the inspection source record is searched from the head of the record group flagged with the heading of the text, and is checked. Repeat the processing of and for the previous record. The above processing is repeated until the end of the record group with the flag of the index of the table of contents.

After executing the processing (1) to (4), a flag indicating that there is a match between a record group in which the flag of the head of the table of contents is set and a record group in which the flag of the head of the body is set is set. Flags all unmatched records as having no match.

A list is created by outputting a record group flagged as a heading of a table of contents and a heading portion of a record group flagged as a heading of a text. At this time, a symbol (for example, ★) indicating that there is no match is added to the heading portion of the record in which the flag indicating that there is no match is set. In addition to the paper output, a list can be displayed on a screen, and a mismatched portion can be clearly indicated by moving a cursor or the like.

[6] By the process 25 shown in FIG. 2, the result of the inspection by the above process is output. FIG. 6 shows an example of the output result. 6A shows the output result 17a of the table of contents, and FIG. 6B shows the output result 17b of the text. In the case of the table of contents, the item name in section 1.1 is "Overall image of xx"
In the case of the text, the item name in section 1.1 is "Overview of XX". Therefore, the symbol ★ indicating no match is
Add to each heading and display or print it.

There are the following five cases of heading inconsistency, all of which can be pointed out by inspection of this processing. FIG. 6 shows an example of (a). (a) The terms are different. Example: In the table of contents, "-Overview" is described as "Overview" in the text. (b) In the case of an error in a term or the like due to an input error or the like. Example: In the table of contents, "environment for operation" is described as "environment for operation" in the text, and so on. In the table of contents, “composition of” is described, but in the text, “offensive of” is described. (c) When the item numbers are incorrect. (d) The heading is in the table of contents but not in the text. (e) The headline is in the text and not in the table of contents. As a result, the table of contents or the text can be modified, and a document having the correct table of contents can be created.

This method can also be used to check the heading of a chapter and a header and the heading of the text, and the mismatch between the chapter and the heading of the text can easily be checked. Note that FIG. 6 shows an example of output by a line printer, but output by a display, which is another display means, can also be output by applying the present technology equally.

[0042]

As described above, according to the present invention,
By simply specifying the document to be inspected, the inconsistency between the table of contents and the text that has been overlooked so far can be determined.
It can be inspected accurately and quickly. In addition, since errors in heading terms and deviations in item numbers due to input errors and the like can be found, there is no danger of creating a table of contents while including errors.

[Brief description of the drawings]

FIG. 1 is a principle configuration diagram of the present invention.

FIG. 2 is a processing flowchart of an embodiment of the present invention.

FIG. 3 is a diagram illustrating an example of a headline extraction process according to an embodiment of the present invention.

FIG. 4 is an explanatory diagram of an inspection process for the presence or absence of a table of contents according to an embodiment of the present invention.

FIG. 5 is an explanatory diagram of a table of contents page determination process according to an embodiment of the present invention.

FIG. 6 is a diagram illustrating an example of an output result according to an embodiment of the present invention.

[Explanation of symbols]

 REFERENCE SIGNS LIST 10 external storage device 11 digitized manuscript 12 processing device 13 index extraction processing unit 14 table of contents existence inspection processing unit 15 table of contents page determination processing unit 16 table of contents / text comparison inspection processing unit 17 output result

Claims (1)

(57) [Claims] A document processing apparatus for processing a digitized document (11) by a computer, comprising: a header extraction processing unit (13) for extracting a record including a heading from the digitized document (11); It examined whether the record group of heading there is a record of the same heading, for a heading of a particular type
Depending on whether there is a record with the same heading,
A table of contents inspection processing unit (14) that checks whether a table of contents exists in the digitized manuscript, and if there is a table of contents, the first of a certain type of heading
Is the same as the record that appears in
Records that contain the same number in the type heading
As records, check the records in order from the beginning, and find
The records before the record to be checked
The table of contents is regarded as a table of contents, and the record group after the record of the inspection destination is
The table of contents page determination processing unit (15), which separates the table of contents from the body of text as a sentence , performs a comparison check between the headings of the table of contents and the headings of the body, and outputs a check result to the table of contents / text comparison processing (16 And a document processing apparatus for inspecting a table of contents.