JP3352709B2 - Document shaping device and processing method of document shaping device - 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 shaping apparatus for extracting a relationship between a plurality of document element data and shaping a document into an easy-to-view form based on the relationship, and a processing method of the document shaping apparatus.

[0002]

2. Description of the Related Art In a system for providing information to a user, how to arrange and shape a plurality of pieces of information is an important point in order to make the information easy to understand.

For example, in a system for formatting a document,
When laying out text, charts, image data, etc. on a page, it is necessary to take into account the clarity of the document. In particular, charts and images need to be allocated near the relevant text, so each time the text is edited or changed to a separate format, the user must specify the optimal layout of the chart. Was.

In order to solve such a complicated work, if a user defines in advance the relationship between a text and a referenced chart, the layout of the chart is semi-automatically determined according to the position where the text is poured. There are systems with functions. In such a system, for example, the user sets a chart area called an anchored frame, anchors the reference point in the text with a anchor, and always places the chart area below the reference point. Is specified.
If the text moves to another page due to some editing or formatting change, the anchored frame automatically moves with the text. However, in a document that refers to the same diagram from multiple places, the anchor for one figure cannot be placed in multiple places in the text. Work such as reconnecting the anchor is required.

On the other hand, a system for automatically extracting a reference relationship between figures and tables by using words such as “FIG. **” and “table **” referring to the figures and tables in the text and assigning them to the same page ( Japanese Patent Application Laid-Open No. 61-21570) has also been proposed. In this system, when there are a plurality of locations that refer to the same chart, a sibling relationship is established between the reference locations, and at the time of shaping, the chart is automatically arranged near the first printed reference location. Therefore, even if the sentence including the first reference point is deleted, if the next reference point remains, the chart can be automatically allocated in the vicinity. However, in this method, since there is no priority between reference locations, when formatting text containing a list of tables and tables in the table of contents, all the tables and tables are allocated to the table of contents, and the tables and tables are allocated to the reference points in the text. The problem that it cannot be performed arises.

[0006]

As described above, in the conventional system, since there is no relation between data (reference locations) referring to data such as charts, processing is performed over a plurality of reference locations. It was difficult. Also, when there is some relationship between the reference parts, the order is only according to the flow of the sentence, and the priority is not set in consideration of the strength and meaning of the reference. It was not possible to automatically perform the assignment within.

Accordingly, the present invention provides a document processing apparatus that automatically determines the optimal arrangement of a chart referenced from a plurality of data based on the logical structure between the referenced data and reduces the burden on the user. The purpose is to:

[0008]

According to the present invention, document element data which is a component of a document is input, and a document structure including a logical structure and a reference structure between the input document element data is extracted and extracted. A document shaping apparatus that allocates document element data based on a document structure that has been formed, shapes the document into a document form, and outputs the formatted document. A document shaping apparatus characterized by having an allocation availability determining unit that determines whether to allocate data near element data to be referenced.

[0009]

According to the present invention, when determining the arrangement of data referenced from a plurality of reference locations, the optimum arrangement of the referenced data is automatically determined based on the logical structure between the referenced document element data. In addition, by presenting and selecting the arrangement of element data to the user according to the priority determined by the system, the arrangement can be easily changed according to the user's intention. Can be reduced.

[0010]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing one embodiment of the present invention.

[0011] The character data, chart data, image data, etc., which are the element data constituting the document, are input from the input unit 10 composed of, for example, a keyboard, a mouse, an image scanner, other storage devices, a communication device, and the like. It is sent to the extraction unit 20. The document structure extracting unit 20 includes a logical structure extracting unit 21 and a reference structure extracting unit 22. First, the logical structure extracting unit 21 extracts a logical structure between element data. Next, the reference structure extraction unit 22 extracts a reference relationship (reference structure) between the element data.

The document shaping section 30 shapes the input document element data using the extracted document structure. The allocation position generation unit 31 allocates the referenced element data to the vicinity of the referenced element data based on the reference structure extracted in the previous stage. At this time, with respect to the element data referred to from the plurality of reference points, the priority order determination unit 32 determines the priority order among the allocation positions for each reference point based on the logical structure between the element data to be referred to. The assigned element data is the output unit 40. For example, a formatted document is output from a display or a printer. The operation of the present invention will be described below with reference to the example of FIG.

The input unit 10 receives document element data composed of text data A and chart data B. For example, text data A is input from the keyboard, and chart data B is input from the scanner. The document structure extraction unit 20 extracts a logical structure and a reference structure from the relationship between the respective document element data.

In the case of the sentence data A, the logical structure is represented by a hierarchical relationship between sentences such as a title, an author, an affiliation, a chapter, and a section. The logical structure extraction unit 21
If the input data includes the structure information added in advance by the user, the structure information is read. Alternatively, when no structure information is added, a logical structure analysis rule may be used to automatically extract the information. The technology for automatically analyzing the logical structure is described in detail in Japanese Patent Application Laid-Open No. 61-190563. The logical structure of the input text data extracted by the logical structure extraction unit 21 has, for example, a hierarchical structure as shown in FIG.

The reference structure is a reference relationship between each element data,
For example, it refers to each figure / table, other chapters, references, etc. from the text. Similarly to the logical structure, the reference structure is read from the input data or is automatically extracted by, for example, the following procedure. For technical documentation,
Reference relations are specified for each figure and table from the text using reference words and the like. In the example of FIG. 2, each sentence representing the description of the drawing from the fifteenth sentence, the “seen in FIG.
0 sentence "This figure", 50th sentence "Function comparison table", 55th sentence
The “next table” in the sentence indicates a reference to the chart. The reference structure extraction unit 22 has, for example, a reference phrase dictionary and a reference description extraction rule as shown in FIG. 4, and extracts expressions referring to figures and tables from sentences. Next, the referenced chart data is analyzed, and the chart corresponding to each reference expression is specified.

Here, "FIG. 1" refers to the sixteenth sentence and the thirty-sixth sentence.
The sentence is referred to from three places: sentence 40. "Table 1" is from the 21st and 50th sentences, and "Table 2" is from the 22nd sentence.
It is referenced from each of the sentence and the 55th sentence. Sixteenth
The correspondence between the reference expression such as the sentence and the 21st sentence and the figure and table is determined by, for example, having a rule specified by matching the combination of words such as “figure” and “table” with the figure and table number. . Also, the 40th sentence, 50th sentence, 55th sentence
For a reference location such as a sentence where a figure / table number is not specified, for example, there is a mechanism for specifying the corresponding figure / table from the anatomical relationship of pronouns by context analysis or the correlation of description of the contents of the figure / table. In addition, for charts that have already been assigned in the text, the distance from the charts and the "
It is also possible to analyze a description designating a position such as “left figure”. Through these analyses, for example, a reference structure as shown in FIG. 5 is extracted. Here, since the priority order is not yet determined, it is left blank.

The shaping section 30 allocates document element data according to, for example, format data (area) as shown in FIG. 6 using the extracted logical structure and reference structure. Here, the dotted frame indicates a text frame into which a text is poured. In the format data, the size of each page, the size of the text frame included in the page, the logical attributes to be poured into the text frame, the order of the text frames, and the like are specified in advance. In the shaping unit 30, sentences having the designated logical attribute are sequentially poured into each sentence frame according to the extracted logical structure. Here, if the format data describes formats such as font size and indent for each logical attribute (title, author name,..., Chapter heading, chapter paragraph),
Each sentence can be formatted according to the format of its logical attributes.

At this time, if a chart is referred to from the sentence, the chart is allocated near the sentence. However, when the same chart is referred to from a plurality of sentences as in the example of FIG. 2, it is detected according to the flowchart shown in FIG. 7, and the processing is passed to the priority order determination unit 32. That is,
First, a sentence is input (step 1), and it is determined whether there is another chart to be referred to (step 2). If there is no sentence, the next sentence is input. A determination is made (step 3). If not, it is next determined whether or not it is referenced from a plurality of locations (step 4). If it is referenced, processing for determining the priority is performed (step 5). Set to 1 (Step 6) and return to Step 2 for each. The priority is determined as described above.

In the priority order determining unit 32, for example, as shown in FIG.
A table is created with sentence numbers referring to the same chart and attribute names in the logical structure of each sentence. The already determined priorities are written in the priority order column in the reference structure of FIG. 5 as needed, and the information is also used in the process of determining the priorities in the next chart. For example, it has a priority determination rule as shown in FIG. 9 and determines the priority.

FIG. 8 (a) lists the reference portions of "FIG. 1." From the priority determination rule 1-1, FIG.
As shown in (b), the priority of the sixteenth sentence drops to third place. According to Rule 2-1 or 2-7, FIG.
The priority is determined as shown in FIG. Similarly, the priority of “Table 1” is determined by Rule 1-1, and “Table 2” is determined by Rule 1-1 or 2-2, 2-5, and the like. Based on the priority determined as described above and surrounded by the reference structure, the allocation position generation unit 31 determines the first priority.
Assign a chart near the rank sentence.

In this example, after all the priorities have been determined, they are allocated to the positions considered to be optimal. However, the priorities can be determined at the time of allocation. In this case, it is possible to generate a plurality of allocation position candidates having specific coordinate values for each reference location, and determine the priority by comparing the actual state of allocation. In either case, it is also possible to have an interface that automatically assigns the user's judgment in accordance with the priority order and then asks the user for confirmation. In this case, for example, when the user selects “to another reference position” using a menu or a command as shown in FIG. 10, the chart is re-allocated to the reference position having the second highest priority. In addition, a menu item having a name such as “to the next reference position” reassigns the chart to the next reference position along the text flow regardless of the priority order. Further, by selecting a menu item such as “reference list”, it is possible to prepare a screen for the user to change the priority order by viewing a list of reference structures as shown in FIG. 11, for example. For example, the user can easily change the priority of the reference relation by selecting a line representing an arbitrary reference relation with a mouse or the like and issuing an instruction to change the thickness of the line. In the example of FIG. 11, the chart is represented by a reduced image icon, but may be represented by a character string such as a chart heading.

If the priority determining unit 32 determines whether or not allocation is possible by describing a rule for determining whether or not allocation is possible, for example, "in the case of a technical document, a table is not allocated to the table of contents", the priority is determined. Even the first place can realize a function such as not allocating a chart.

Further, since the relationship between the reference locations of the same chart is extracted, in the case of a document having the chart description in the table of contents as in the example of FIG. 2, the page number actually assigned to the chart is replaced by It can be added to the table captions in the table of contents.

[0024]

As described above in detail, according to the present invention, when determining the arrangement of data to be referred to from a plurality of reference points, it is possible to automatically allocate the data near an optimum reference point. Become. As a result, troublesome layout work at the time of displaying, shaping, and editing a document can be reduced. As the document has a more complicated reference structure, it becomes more difficult to generate and manage the assignment that matches the reference structure, and thus the effect of the present invention is increased.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment of the present invention.

FIG. 2 is a diagram showing an example of input document data.

FIG. 3 is a diagram showing an example of a logical structure extracted by a logical structure extraction unit.

FIG. 4 is a diagram showing an example of a reference phrase dictionary and a reference description extraction rule used in a reference structure extraction unit.

FIG. 5 is a diagram showing an example of a reference structure extracted by a reference structure extraction unit.

FIG. 6 is a diagram illustrating an example of format data used in a shaping unit.

FIG. 7 is a diagram showing a detection algorithm of a case where the same chart is referred to from a plurality of sentences.

FIG. 8 is a diagram showing a priority order determination table used by a priority order determination unit.

FIG. 9 is a diagram illustrating an example of a priority determination rule used by a priority determination unit.

FIG. 10 is a diagram showing an example of a menu for inputting a user's instruction.

FIG. 11 is a diagram showing an example of a screen for changing a reference structure.

[Explanation of symbols]

10 input unit 20 document structure extraction unit 30 shaping unit
40 ... output unit

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-61-21570 (JP, A) JP-A-63-245556 (JP, A) JP-A-1-230168 (JP, A) JP-A-2- 3859 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06F 17/21-17/24

Claims (2)

(57) [Claims] 1. A logical structure extracting means for extracting a hierarchical logical structure between a plurality of document element data which are constituent elements of a document; a document element data referred to by the document element data; Reference structure extraction means for extracting a reference structure indicating a relationship between reference locations of document element data to be referenced, and priority determination for determining a priority of an allocation position of document element data referenced from the logical structure and the reference structure A rule storage unit for storing rules, and a reference structure extracted by the reference structure extraction unit,
When there are a plurality of document element data that refer to the same referenced document element data, based on the logical structure and the reference structure of the referenced document element data, based on the priority determination rules stored in the rule storage unit, Allocation position priority determining means for determining the priority of the document element data to which the referenced document element data is allocated; and allocating the referenced document element data to the vicinity of the document element data determined by the position priority determining means. A document shaping device comprising: a document shaping unit for shaping a document by using the document shaping unit; and a document output unit for outputting the document shaped by the document shaping unit. 2. A logical structure extracting means, a reference structure extracting means, a rule storing means for storing a priority determining rule for determining a priority of an allocation position of document element data referenced from the logical structure and the reference structure, A processing method of a document shaping device including an allocation position priority determining unit, a document shaping unit, and a document output unit. The logical structure extracting means for extracting a hierarchical logical structure among a plurality of document element data which are constituent elements of the document; and the reference structure extracting means, wherein the document element data referred to by the document element data and Extracting a reference structure indicating a relationship between reference locations of the document element data referring to the document element data; and the allocating position priority determining means assigns the same reference to the reference structure extracted by the reference structure extracting means. If there are a plurality of document element data that refer to the document element data to be referred to, the document element data is referenced based on the logical structure and the reference structure of the referenced document element data based on the priority determination rules stored in the rule storage unit. Determining the priority order of the document element data to which the document element data is to be allocated; Allocating a document in the vicinity of the document element data to be determined determined by the position priority determining means, and shaping the document; and outputting the document formatted by the document shaping means by the document output means. A processing method of a document shaping device, characterized by the following.