JP6568021B2 - Logical relationship recognition apparatus, logical relationship recognition method, and logical relationship recognition program - Google Patents

Description

  The present invention relates to a logical relationship recognition apparatus, a logical relationship recognition method, and a logical relationship recognition program.

  OpS (Operation Support System: Operation Support System) has been introduced for business processing in various companies. On the other hand, for operations that cannot be supported by OpS, each organization has its own operations, and forms are used for information distribution. Since the form is semi-structured data, the meaning (logical relationship) of each value can be understood by a person but difficult for a machine, and a large amount of manual work is required when utilizing the data on the form. There is a burden. On the other hand, a technique for automatically recognizing a logical relationship between item names of a form and a logical relationship between item names and item values based on the arrangement of ruled line frames is known (for example, Patent Documents). 1 and non-patent documents 1 to 3).

JP, 2006-004417, A

Takagi et al., “Examination of Extraction Techniques for Cross-sectional Data Manipulation Techniques for Electronic Forms”, IEICE Technical Report, vol.114, No.150 pp.1-6 (2014) Takagi et al., “Investigation of visual representation of forms for linking electronic form data”, IEICE Communication Society Conference Proceedings (2015) Takagi et al., "Examination of data structure conversion method of electronic forms using visual expression", Shingaku Techniques, vol.115, No.409, pp25-30 (2016)

  However, the conventional technique has a vertical enumeration, horizontal enumeration, vertical enumeration and horizontal enumeration combined type, vertical table, horizontal table, orthogonal table, nested structure type enumeration, nested structure vertical table, nested structure horizontal table or When nested orthogonal tables are included or mixed, there is a problem that the logical relationship between item names in the form and the logical relationship between item names and item values may not be recognized correctly. It was.

  For example, with respect to a table composed of ruled line frames of 4 rows × 3 columns, a vertical table in which the three ruled line frames in the uppermost row are item names and a horizontal table in which the four ruled line frames in the leftmost column are item names And the case where both the three ruled line frames in the uppermost row and the four ruled line frames in the leftmost column are orthogonal tables that are item names. In such a case, it is difficult for the conventional technique to accurately recognize the logical relationship between the item names and the logical relationship between the item names and the item values.

  For example, in the case of vertical enumeration, there may be a plurality of ruled line frames of the same width corresponding to the item names in the vertical direction. Further, each item name has an item value, and the width of the ruled line frame corresponding to the item value is the same as the width of the ruled line frame corresponding to the item name. In such a case, the conventional technology cannot identify which ruled line frame is the item name, so it accurately recognizes the logical relationship between the item names and the logical relationship between the item names and the item values. It is difficult to do.

  Further, for example, when a table represented by an enumeration and a nested structure is included in the nested structure, it is difficult for the conventional technique to accurately recognize the logical relationship between the item name and the item value.

  The logical relationship recognition apparatus of the present invention represents information relating to an area representing an item name or item value of a form as a node, and is set in advance among the nodes based on a graph representing an adjacency relationship between the nodes as an edge. An extraction unit that extracts a node that satisfies the condition as an item name node that is a node in an area that represents an item name, and based on the edge, extracts a starting node that is a node that is a starting point of the table from the item name node The table starting from the origin node is either an orthogonal table in which item names exist at both the top and left ends, a vertical table in which item names exist at the top, and a horizontal table in which item names exist at the left end. A vertical classification logical relationship between the item name nodes in the orthogonal table, a horizontal logical relationship between the item name nodes in the orthogonal table, the table in the orthogonal table A vertical logical relationship between a name node and an item value node that is a node other than the item name node, and a horizontal logical relationship between the item name node and the item value node in the orthogonal table. An orthogonal table acquisition unit to acquire, a vertical logical relationship between the item name nodes in the vertical table, and a vertical logical relationship between the item name node and the item value node in the vertical table are acquired. A horizontal table for acquiring a horizontal logical relationship between the item name nodes in the horizontal table and a horizontal logical relationship between the item name nodes and the item value nodes in the horizontal table An acquisition unit; a first specifying unit that specifies a table to be synthesized by excluding a table that satisfies a predetermined condition indicating that the table is inconsistent from the orthogonal table, the vertical table, and the horizontal table; The given one of the two nodes When a plurality of nodes are adjacent to each other, a first deletion unit that deletes an edge representing an adjacent relationship between the one node and the plurality of nodes, and an item in a predetermined direction among the item name nodes An item name node that is adjacent to an item value node that is a node of an area that represents a value, a second deletion unit that deletes an edge that represents an adjacent relationship with the item value node, the first deletion unit, and the A first acquisition unit that acquires a logical relationship between the item name node and the item value node based on the graph in which an edge is deleted by a second deletion unit; and the first deletion unit Based on the graph in which the edge is deleted by the second acquisition unit that acquires the inclusion relationship between the item name nodes, and the combination target among the logical relationships acquired by the first acquisition unit On the nodes in the table A second specifying unit that specifies a logical relationship regarding enumeration of synthesis targets excluding a logical relationship; and an inclusion relationship acquired by the second acquisition unit, a logical relationship related to the table to be combined and the combination target Creating a first tree structure data based on an inclusive relationship not included in any of the enumerated logical relationships, and creating a tree structure data based on the logical relationship regarding the table to be synthesized A tree structure based on a logical relationship relating to the enumeration of the synthesis target and a second synthesis unit that creates a second tree structure data obtained by synthesizing the tree structure data and the first tree structure data And a third synthesis unit for creating a third tree structure data obtained by synthesizing the tree structure data and the second tree structure data.

  The logical relationship recognition method of the present invention is a logical relationship recognition method executed by a logical relationship recognition device, wherein information relating to an area representing an item name or item value of a form is represented as a node, and the adjacent relationship between the nodes Based on the graph expressed as an edge, an extraction step of extracting a node that satisfies a preset condition from among the nodes as an item name node that is a node of an area representing an item name, and based on the edge, A starting node that is a node that is a starting point of the table is extracted from the item name node, and a table that has the starting node as the starting point is an orthogonal table in which item names exist at both the upper end and the left end, and an item name exists at the upper end. A table classification process for classifying the table into one of a vertical table and a horizontal table having an item name at the left end, a vertical logical relationship between the item name nodes in the orthogonal table, and the orthogonal table A horizontal logical relationship between the item name nodes, a vertical logical relationship between the item name node in the orthogonal table and an item value node that is a node other than the item name node, and the orthogonal table An orthogonal table acquisition step of acquiring a horizontal logical relationship between an item name node and the item value node, a vertical logical relationship between the item name nodes in the vertical table, and the item name in the vertical table A vertical table acquisition step of acquiring a vertical logical relationship between a node and the item value node, a horizontal logical relationship between the item name nodes in the horizontal table, and the item name node in the horizontal table; A table that satisfies a predetermined condition indicating that the table is an inconsistent table from the orthogonal table, the vertical table, and the horizontal table, and a horizontal table acquisition step of acquiring a horizontal logical relationship with the item value node. When a plurality of nodes are adjacent to each other in a predetermined direction of one node and a first specifying step of specifying the excluded synthesis target table, the adjacent relationship between the one node and the plurality of nodes is represented. An item name node adjacent to an item value node that is a node of an area representing an item value in a predetermined direction, and the item value node; The item name node and the item value based on a second deletion step of deleting an edge representing an adjacency, and the graph in which the edge is deleted by the first deletion step and the second deletion step. A first acquisition step of acquiring a logical relationship between the nodes, and a second acquisition step of acquiring an inclusion relationship between the item name nodes based on the graph in which edges are deleted by the first deletion step. And the acquisition step A second specifying step of specifying a logical relationship related to enumeration of synthesis targets, excluding a logical relationship related to a node included in the table to be combined among the logical relationships acquired in one acquisition step; and the second acquisition A first tree-structured data is generated based on an inclusion relation that is not included in any of the inclusion relation acquired by the process and included in either the logical relation relating to the table to be synthesized or the logical relation relating to the enumeration of the synthesis target. The data of the second tree structure in which the data of the tree structure is created based on the synthesis process of 1 and the logical relationship regarding the table to be synthesized, and the data of the tree structure and the data of the first tree structure are synthesized Based on the logical relationship regarding the enumeration of the synthesis target, and the third synthesis step of synthesizing the tree structure data and the second tree structure data. Wooden frame Characterized in a third synthesis step of creating the data, that it contained.

  Further, the logical relationship recognition program of the present invention represents, on a computer, information relating to an area representing an item name or item value of a form as a node, and a graph representing the adjacency relationship between the nodes as an edge. Among these, the extraction step of extracting a node that satisfies a preset condition as an item name node that is a node of an area representing an item name, and a node that is a starting point of the table from the item name node based on the edge Extracts the origin node, and uses the origin node as the origin table, the orthogonal table in which the item name exists at both the upper end and the left end, the vertical table in which the item name exists in the upper end, and the horizontal table in which the item name exists at the left end A table classification step for classifying into any of the above, a vertical logical relationship between the item name nodes in the orthogonal table, and between the item name nodes in the orthogonal table Logical relationship in direction, vertical logical relationship between the item name node in the orthogonal table and an item value node that is a node other than the item name node, and the item name node and the item value node in the orthogonal table An orthogonal table acquisition step for acquiring a horizontal logical relationship between the item name node and the item name node in the vertical table, and the item name node and the item value node in the vertical table A vertical table acquisition step of acquiring a vertical logical relationship between the item name nodes in the horizontal table, and between the item name node and the item value node in the horizontal table A table obtained by excluding a table satisfying a predetermined condition indicating that the table is inconsistent from the orthogonal table, the vertical table, and the horizontal table, and a horizontal table acquisition step of acquiring a logical relationship in the horizontal direction. When a plurality of nodes are adjacent to each other in a first direction for specifying a target table and a predetermined direction of one node, an edge representing an adjacent relationship between the one node and the plurality of nodes is deleted. A first deletion step, and an item name node adjacent to an item value node that is a node of an area representing an item value in a predetermined direction among the item name nodes and an adjacency relationship between the item value nodes Based on the second deletion step of deleting the edge to be represented, and the graph in which the deletion of the edge was performed by the first deletion step and the second deletion step, the item name node and the item value node A first acquisition step of acquiring a logical relationship between the item name nodes, and a second acquisition of acquiring an inclusion relationship between the item name nodes based on the graph from which edges have been deleted by the first deletion step. A second specifying step for specifying a logical relationship related to enumeration of synthesis targets, excluding a logical relationship related to nodes included in the table to be synthesized among the logical relationships obtained in the first obtaining step; The first tree structure based on the inclusion relationship that is not included in either the logical relationship related to the table to be synthesized and the logical relationship related to the enumeration of the synthesis target among the inclusion relationships acquired by the second acquisition step A first synthesizing step for creating the data of the tree, and creating a tree structure data based on the logical relationship relating to the table to be synthesized, and synthesizing the tree structure data and the first tree structure data. A second synthesizing step for creating two tree-structured data, and creating a tree-structured data on the basis of a logical relationship relating to the enumeration of the synthesis target, A third synthesis step of creating the data of the third tree structure obtained by synthesizing the data of the tree structure, characterized in that to the execution.

  According to the present invention, a vertical enumeration, horizontal enumeration, vertical enumeration and horizontal enumeration, vertical table, horizontal table, orthogonal table, nested structure type enumeration, nested structure vertical table, nested structure horizontal table or nested When the structure orthogonal table is included or mixed, it is possible to accurately recognize the logical relationship between the item names of the form and the logical relationship between the item name and the item value.

FIG. 1 is a diagram for explaining the outline of the logical relationship recognition process. FIG. 2 is a diagram illustrating an example of a form including a horizontal table and a vertical table. FIG. 3 is a diagram illustrating an example of a form including a horizontal table and a vertical table. FIG. 4 is a diagram illustrating an example of a form including a vertical table. FIG. 5 is a diagram illustrating an example of a form in which an orthogonal table is included in a nested structure. FIG. 6 is a diagram illustrating an example of the configuration of the logical relationship recognition apparatus according to the first embodiment. FIG. 7 is a diagram for explaining acquisition of adjacent nodes. FIG. 8 is an example of a table list of a horizontal table. FIG. 9 is an example of a table list of an orthogonal table. FIG. 10 is an example of a table list of a horizontal table. FIG. 11 is an example of an enumeration list. FIG. 12 is a diagram for explaining the inclusion graph. FIG. 13 is a diagram for explaining the inclusion graph. FIG. 14 is an example of tree structure data. FIG. 15 is a flowchart showing a flow of processing for generating a form graph. FIG. 16 is a flowchart showing a flow of processing for acquiring a node. FIG. 17 is a flowchart showing a flow of processing for generating a node. FIG. 18 is a flowchart showing a flow of processing for acquiring the adjacency relationship. FIG. 19 is a flowchart showing a flow of processing for acquiring adjacent node groups. FIG. 20 is a flowchart showing a flow of processing for generating a style graph. FIG. 21 is a flowchart showing the flow of processing of the extraction unit. FIG. 22 is a flowchart showing the flow of processing of the analysis unit. FIG. 23 is a flowchart showing the flow of processing of the table classification unit. FIG. 24 is a flowchart showing the flow of processing of the vertical table acquisition unit. FIG. 25 is a flowchart showing the flow of processing of the horizontal table acquisition unit. FIG. 26 is a flowchart showing the flow of processing of the orthogonal table acquisition unit. FIG. 27 is a flowchart illustrating a flow of processing for acquiring a vertical logical relationship. FIG. 28 is a flowchart showing a flow of processing for acquiring a horizontal logical relationship. FIG. 29 is a flowchart showing the flow of processing of the table matching unit. FIG. 30 is a flowchart showing the flow of processing of the first deletion unit. FIG. 31 is a flowchart showing the flow of processing of the second deletion unit. FIG. 32 is a flowchart showing the flow of processing of the enumeration classification unit. FIG. 33 is a flowchart showing the flow of processing of the vertical enumeration acquisition unit. FIG. 34 is a flowchart illustrating a process flow of the horizontal enumeration acquisition unit. FIG. 35 is a flowchart illustrating a process flow of the inclusion relationship acquisition unit. FIG. 36 is a flowchart showing the flow of processing for acquiring the right inclusion relationship. FIG. 37 is a flowchart showing a flow of processing for acquiring the lower inclusion relationship. FIG. 38 is a flowchart showing the flow of processing of the inclusion graph generation unit. FIG. 39 is a flowchart showing the flow of processing of the enumeration matching unit. FIG. 40 is a flowchart showing the flow of processing of the item name synthesizing unit. FIG. 41 is a flowchart showing a process flow of the table synthesis unit. FIG. 42 is a flowchart showing the flow of processing of the enumeration synthesis unit. FIG. 43 is a flowchart showing the flow of processing of the adding unit. FIG. 44 is a diagram for explaining another embodiment. FIG. 45 is a diagram illustrating an example of a computer in which the logical relationship recognition apparatus is realized by executing a program.

  Hereinafter, embodiments of a logical relationship recognition device, a logical relationship recognition method, and a logical relationship recognition program according to the present application will be described in detail with reference to the drawings. In addition, this invention is not limited by this embodiment.

  First, the outline of the logical relationship recognition process by the logical relationship recognition system having the logical relationship recognition device will be described with reference to FIG. FIG. 1 is a diagram for explaining the outline of the logical relationship recognition process. As shown in FIG. 1, first, the logical relationship recognition system reads a tabular form from a PC or the like (step S1). At this time, the data read by the logical relationship recognition system is not limited to a form, and may be semi-structured data such as a Web GUI, a system GUI, and a table structure on an image. Next, the logical relationship recognition system acquires information on the ruled line frame from the read form (step S2). Further, the logical relationship recognition system may acquire the description content from an image of a paper form read by a scanner or the like by OCR (Optical Character Recognition) (step S3).

  Then, the logical relationship recognition system generates a style graph based on the ruled line frame information and the item name definition information (step S4). The ruled line frame information includes visual information such as the coordinates of the ruled line frame, the character string in the ruled line frame, the fill color of the ruled line frame, the type, thickness, and color of the ruled line. Further, the item name definition information includes a condition for determining a ruled line frame as an item name.

  For example, item name definition information whose item name is a ruled frame with a yellow color is described as “if {node: {color: # FFFF00}} then item”. Also, for example, ruled line frame definition information whose item name is a non-empty ruled line frame of a character string is described as “if {node: {! String: null}} then item”. Also, for example, ruled line frame definition information whose item name is a ruled line frame whose color is not white and whose character string is not empty is “if {node: {! Color: white}, {! String: null} then item” Is described. Further, for example, ruled line frame definition information whose item name is a ruled line frame with the character string “y1” is described as “if {node: {string:“ y1 ”}} then item”.

  Further, the style graph is a graph in which a ruled line frame or the like is represented as a node and an adjacent relationship between nodes is represented as an edge for each of a plurality of styles included in the form. In the subsequent processing, the logical relationship recognition system handles the form format as graph format data. The form also includes a table. In the following description, the case where the logical relationship is recognized based on the style graph will be described. However, a form graph in which the entire form is represented by nodes and edges may be used.

  Next, the logical relationship recognition system recognizes the logical relationship from the tabular structure (step S5), converts the recognized result into a predetermined data structure (xml, yaml, json, etc.) (step S6), Store in DB. At this time, the link path of the data structure may be stored in the DB, or the data may be stored in accordance with a predefined schema.

  Next, with reference to FIGS. 2 to 5, the structure of a form that is a target of logical relationship recognition processing by the logical relationship recognition system will be described. 2 to 5 represent nodes, and in the following description, for the sake of explanation, the nodes indicated by these symbols are simply referred to as “nodes” or “item name nodes” or “item value nodes”. There is a case. 2 to 5, shaded portions represent item names, and portions not shaded represent item values.

  Here, the vertical table is a table in which, for example, item names and item names and item values are adjacent in the vertical direction. The horizontal table is a table in which, for example, item names are adjacent to each other and item names and item values are adjacent in the horizontal direction. The orthogonal table is a table in which, for example, item names and item names and item values are adjacent in the vertical direction and the horizontal direction. The vertical enumeration has a plurality of logical relationships in the vertical direction, which are the relationship between item names and the relationship between item names and item values. The horizontal enumeration has a plurality of logical relationships in the horizontal direction.

  FIG. 2 is a diagram illustrating an example of a form including a horizontal table and a vertical table. In FIG. 2, for example, a vertical table starting from the nodes a2, a3, and a4 is included in the horizontal arrangement starting from the node a1.

  FIG. 3 is a diagram illustrating an example of a form including a horizontal table and a vertical table. FIG. 3 includes, for example, a vertical table starting from b8, b9, and b10 in the horizontal arrangement starting from b1. Note that the structure of FIG. 3 is horizontal enumeration because there are a plurality of horizontal logical relationships.

  FIG. 4 is a diagram illustrating an example of a form including a vertical table. In FIG. 4, for example, a vertical table starting from c6, c7, c8 and c9 is included in the vertical arrangement starting from c1. Note that the structure in FIG. 4 is listed vertically because there are a plurality of logical relationships in the vertical direction.

  FIG. 5 is a diagram illustrating an example of a form in which an orthogonal table is included in a nested structure. In FIG. 5, for example, an orthogonal table is included as a nesting in a horizontal arrangement starting from d1, d6, d8, d36, d38, and d40.

[Configuration of First Embodiment]
Next, the configuration of the logical relationship recognition apparatus according to the first embodiment will be described with reference to FIG. FIG. 6 is a diagram illustrating an example of the configuration of the logical relationship recognition apparatus according to the first embodiment. As illustrated in FIG. 6, the logical relationship recognition device 10 includes a control unit 20 and a storage unit 30.

  The control unit 20 controls the entire logical relationship recognition apparatus 10. The control unit 20 is, for example, an electronic circuit such as a CPU (Central Processing Unit) or MPU (Micro Processing Unit), or an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The control unit 20 has an internal memory for storing programs and control data that define various processing procedures, and executes each process using the internal memory. The control unit 20 functions as various processing units when various programs are operated. For example, the control unit 20 includes an extraction unit 201, an analysis unit 202, a table classification unit 203, a vertical table acquisition unit 204, a horizontal table acquisition unit 205, an orthogonal table acquisition unit 206, a table matching unit 207, a first deletion unit 208, Second deletion unit 209, enumeration classification unit 210, vertical enumeration acquisition unit 211, horizontal enumeration acquisition unit 212, enumeration matching unit 213, inclusion relation acquisition unit 214, inclusion graph generation unit 215, item name composition unit 216, table composition A unit 217, an enumeration synthesis unit 218, and an addition unit 219.

  The storage unit 30 is a storage device such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or an optical disk. Note that the storage unit 30 may be a semiconductor memory that can rewrite data, such as a random access memory (RAM), a flash memory, and a non-volatile static random access memory (NVSRAM). The storage unit 30 stores an OS (Operating System) executed by the logical relationship recognition apparatus 10 and various programs. Furthermore, the storage unit 30 stores various information used in executing the program. In addition, the storage unit 30 stores, for example, a table list 301, an enumeration list 302, and an inclusion graph 303.

  Here, the logical relationship recognition processing by the logical relationship recognition device 10 will be described, and details of each part of the logical relationship recognition device 10 will be described. The extraction unit 201 represents information related to the area representing the item name or item value of the form as a node, and a pre-set condition among the nodes based on a graph representing the adjacent relationship between nodes as an edge, that is, a style graph. Nodes satisfying the condition are extracted as item name nodes that are nodes in the area representing the item name.

  FIG. 7 is a diagram for explaining acquisition of adjacent nodes. A case where there are two nodes, node 1 and node 2, will be described. When the side of one node includes the side of the other node, or when the side of one node extends to the side of the other node, the nodes have an adjacency relationship.

  For example, in the example of (a) to (c) of FIG. 7, it is determined that the node 2 is adjacent to the left side of the node 1. Further, in the examples of (d) to (f) in FIG. 7, it is determined that the node 2 is adjacent to the right side of the node 1. Further, in the examples of (g) to (i) in FIG. 7, it is determined that the node 2 is adjacent to the upper side of the node 1. In the example of (j) to (l) in FIG. 7, it is determined that the node 2 is adjacent to the lower side of the node 1.

[Table acquisition processing]
First, the processing of the table classification unit 203, the vertical table acquisition unit 204, the horizontal table acquisition unit 205, the orthogonal table acquisition unit 206, and the table matching unit 207 will be described, and the table acquisition processing will be described. Based on the edge of the style graph, the table classification unit 203 extracts a starting node, which is a node that is the starting point of the table, from the item name node, and converts the table starting from the starting node to the item name at both the upper end and the left end. The table is classified into an orthogonal table that exists, a vertical table that has an item name at the upper end, and a horizontal table that has an item name at the left end.

For example, the table classification unit 203 performs classification based on whether or not a predetermined condition is satisfied when the upper left node, that is, the starting node is the target node. From now on, the relationship with the node obtained by tracing the node adjacent to the right side from the target node is “connected to the right side”, and the relationship with the node obtained by tracing the node adjacent to the lower side from the target node Called “connect to the bottom”. At this time, the table classification unit 203 first defines a node group connected to the right side of the starting node as rights and a node group connected to the lower side of the starting node as bottoms. Then, it is determined whether or not the following conditions 1 and 2 are satisfied for rights and bottoms.
(Condition 1) All rights are item names and the height of rights is the same as the height of the starting node.
(Condition 2) “bottoms” is the item name and the width of “bottoms” is the same as the width of the starting node.

  For example, in the example of FIG. 5, the table classification unit 203 sets the nodes d2 and d4 to rights when the node d1 is the starting node. Further, when the node d9 is the starting node, the table classification unit 203 sets the nodes d10, d11, and d12 as rights, and the nodes d13, d14, d17, d20, d21, d24, d27, d30, and d33 as bottoms.

  When condition 1 and condition 2 are satisfied, table classification unit 203 classifies the table starting from the starting node into an orthogonal table, and when condition 1 is satisfied and condition 2 is not satisfied, the starting node is the starting point. The table is classified into a vertical table, and when the condition 2 is satisfied and the condition 1 is not satisfied, the table starting from the starting node is classified as a horizontal table.

  Next, the orthogonal table acquisition unit 206 includes a vertical logical relationship between item name nodes in the orthogonal table, a horizontal logical relationship between item name nodes in the orthogonal table, and items other than item name nodes and item name nodes in the orthogonal table. A vertical logical relationship between the item value nodes, which are nodes, and a horizontal logical relationship between the item name node and the item value node in the orthogonal table are acquired. Further, the vertical table acquisition unit 204 acquires the vertical logical relationship between the item name nodes in the vertical table and the vertical logical relationship between the item name node and the item value node in the vertical table. In addition, the horizontal table acquisition unit 205 acquires a horizontal logical relationship between item name nodes in the horizontal table and a horizontal logical relationship between item name nodes and item value nodes in the horizontal table.

  The orthogonal table acquisition unit 206 acquires a combination of the item name and the item name or item value on the right side of the item name as a logical relationship in the horizontal direction, and the item name or item value below the item name and the item name. Is obtained as a vertical logical relationship. The horizontal table acquisition unit 205 acquires a combination of an item name and an item name or item value on the right side of the item name as a logical relationship in the horizontal direction. The vertical table acquisition unit 204 acquires a combination of an item name and an item name or item value below the item name as a vertical logical relationship.

  The vertical table acquisition unit 204 acquires the logical relationship as a list, and stores the acquired list in the storage unit 30 as the table list 301. Note that the processing of the extraction unit 201 and the table classification unit 203 knows that the table is a vertical table and that the node a3 and the node a4 are item names, so the vertical table acquisition unit 204 changes the node a4 to the node a4. It does not acquire an erroneous logical relationship as a child in the horizontal direction of the node a3.

  Further, the vertical table acquisition unit 204 acquires the logical relationship from the node group connected to the right side of the upper left node of the table, for example, with the uppermost node group as last_tops and the last_tops as a starting point. Further, the horizontal table acquisition unit 205 acquires the logical relationship as a list, and stores the acquired list in the storage unit 30 as the table list 301.

  In addition, the horizontal table acquisition unit 205 acquires, for example, the leftmost node group among the node groups connected to the lower side of the upper left node of the table as the last_lefts, and acquires the logical relationship using the last_lefts as a starting point. The orthogonal table acquisition unit 206 acquires the logical relationship as a list, and stores the acquired list in the storage unit 30 as the table list 301.

  Further, the orthogonal table acquisition unit 206, for example, designates the uppermost node group among the node groups connected to the right side of the upper left node of the table as least_tops, and is the most of the node group connected to the lower side of the upper left node of the table. Let the node group on the left be the last_lefts. Then, the orthogonal table acquisition unit 206 acquires a logical relationship starting from last_tops and last_lefts.

  FIG. 8 is an example of a table list of a horizontal table. First, as shown in FIG. 8, the table list describes that the table classification is “horizontal table”. Further, for example, as shown in FIG. The row 1 indicates that the horizontal children of the node d8 are nodes d9, d13, d20, d27, d30, and d33. Further, for example, as shown in FIG. The row 2 indicates that the horizontal children of the node d9 are nodes d10 and d11. Further, for example, as shown in FIG. The row 3 indicates that the horizontal child of the node d11 is the node d12. Further, for example, as shown in FIG. The row 4 indicates that the horizontal children of the node d13 are nodes d14 and d17.

  Here, as shown in FIG. 5, since the nodes d10, d11, and d12 are item name nodes, it is necessary to have item value nodes that are children in the vertical direction. However, No. in the table list of FIG. Line 3 describes that the horizontal child of node d11 is node 12. Further, the child of the node d10 is not described in the table list. For this reason, the table matching unit 207 excludes an erroneous table list as shown in FIG.

  FIG. 9 is an example of a table list of an orthogonal table. FIG. 10 is an example of a table list of a horizontal table. Since the table lists of FIGS. 9 and 10 are not in error, they are not excluded by the table matching unit 207.

  The table matching unit 207 identifies a table that excludes a table that satisfies a predetermined condition indicating that the table is inconsistent from the orthogonal table, the vertical table, and the horizontal table. For example, the table matching unit 207 includes an orthogonal table having an item name node having no item value node on the lower side and the right side, and an item having no item value node on the lower side from the orthogonal table, the vertical table, and the horizontal table. A table excluding a vertical table having name nodes and a horizontal table having item name nodes having no item value nodes on the right side is specified. Specifically, the table matching unit 207 deletes from the table list 301 a table list that includes item name nodes that do not have item value nodes as child nodes. Alternatively, the table list in which the item name node having the item name node as a child node exists is deleted. The table matching unit 207 is an example of a first specifying unit. For example, the table matching unit 207 deletes the table list of FIG. 8 and does not delete the table lists of FIGS.

[Enumeration acquisition processing]
Next, processing of the first deletion unit 208, the second deletion unit 209, the enumeration classification unit 210, the vertical enumeration acquisition unit 211, the horizontal enumeration acquisition unit 212, the inclusion relationship acquisition unit 214, and the inclusion graph generation unit 215 will be described. In addition, enumeration acquisition processing will be described.

  When a plurality of nodes are adjacent to each other in a predetermined direction of one node, the first deletion unit 208 deletes an edge representing the adjacency relationship between the one node and the plurality of nodes. The first deletion unit 208 may delete an edge representing the adjacency relationship between one node and a plurality of nodes when a plurality of nodes are adjacent to the left or upper side of the one node. Good.

  Also, the second deletion unit 209 determines the adjacency relationship between the item name node and the item value node adjacent to the item value node that is the node of the area representing the item value in a predetermined direction among the item name nodes. Delete the representing edge. Note that the second deletion unit 209 has an adjacency relationship between an item name node and an item value node adjacent to an item value node that is a node of an area representing an item value on the left or upper side of the item name nodes. You may make it delete the edge showing.

  Then, the vertical enumeration acquisition unit 211 and the horizontal enumeration acquisition unit 212 determine whether the item name node and the item value node are based on the graph in which the edge is deleted by the first deletion unit 208 and the second deletion unit 209. Get logical relationship between. Specifically, the vertical enumeration acquisition unit 211 and the horizontal enumeration acquisition unit 212 acquire the logical relationship as a list as illustrated in FIG. 11 and store the acquired list in the storage unit 30 as the enumeration list 302. FIG. 11 is an example of an enumeration list. The enumeration classification unit 210 classifies the graph into a vertical enumeration and a horizontal enumeration before processing by the vertical enumeration acquisition unit 211 and the horizontal enumeration acquisition unit 212.

  FIG. 11 is an enumeration list showing the logical relationship of the enumeration of FIG. In the list of FIG. The row 1 indicates that the vertical children of the item name node d11 are nodes d15, d18, d22, d25, d28, d31, and d34. In the list of FIG. Line 3 indicates that the horizontal child of the item name node d2 is the node d3.

  Further, for example, in the case of FIG. 5, since the edge indicating the adjacency relationship between the node d1 and the nodes d2 and d4 has been deleted by the first deletion unit 208, the vertical enumeration acquisition unit 211 and the horizontal enumeration acquisition unit 212. However, a logical relationship in which the horizontal children of the item name node d1 are d3 and d5 is not acquired.

  For example, in the case of FIG. 4, since the edge indicating the adjacency relationship between the node c2 and the node c6 is deleted by the second deletion unit 209, the vertical enumeration acquisition unit 211 and the horizontal enumeration acquisition unit 212 A logical relationship in which the vertical child of the item name node c2 is c10 is not acquired.

  The inclusion relationship acquisition unit 214 acquires the inclusion relationship between the item name nodes based on the graph in which the edge is deleted by the first deletion unit 208. Further, the inclusion graph generation unit 215 generates a graph as shown in FIG. 12 based on the inclusion relationship acquired by the inclusion relationship acquisition unit 214 and stores the generated graph in the storage unit 30 as the inclusion graph 303. FIG. 12 is a diagram for explaining the inclusion graph. The broken line in FIG. 12 represents the vertical inclusion relationship. Also, the solid line in FIG. 12 represents the horizontal inclusion relationship. Hereinafter, when the target node includes one or more other nodes in the vertical direction (or horizontal direction), the target node is referred to as a vertical (or horizontal) included node. Among nodes having an inclusion relationship, a node that is not included in another node is referred to as a master node, and another node that is included is referred to as a slave node.

  The inclusion graph of FIG. 12 corresponds to the structure of FIG. As shown in FIGS. 5 and 12, the node d1 includes nodes d2 and d4 in the horizontal direction. The node d8 includes nodes d9 and d10 in the horizontal direction. The node d9 includes nodes d13, d20, d27, d30, and d33 in the vertical direction. The node d13 includes nodes d14 and d17 in the horizontal direction. The node d20 includes nodes d21 and d24 in the horizontal direction. The node d10 includes nodes d11 and d12 in the vertical direction. The node d40 includes nodes d41, d42, and d47 in the horizontal direction. The node d47 includes nodes d48 and d49 in the vertical direction.

  Specifically, the inclusion relationship acquisition unit 214 includes at least one of the first node groups adjacent to the right side of the first item name node as an item name node, and all of the first node groups included in the first node group. If the height of the node is less than or equal to the height of the first item name node, and the vertex of the upper left node of the first node group and the vertex of the first item name node overlap, the first Is determined to include the first node group in the horizontal direction. In addition, the inclusion relationship acquisition unit 214 has at least one of the second node groups adjacent to the lower side of the second item name node as an item name node, and all the nodes included in the second node group The width of the second item name node is less than or equal to the width of the second item name node, and the vertex of the upper left node of the second node group overlaps the vertex of the first item name node, the second item name It is determined that the node includes the second node group in the vertical direction.

  For example, as shown in FIG. 5, at least a node d9 is an item name node in a node group adjacent to the right side of the item name node d8. The heights of all nodes included in the node group adjacent to the right side of the item name node d8 are all equal to or lower than the height of the item name node d8. Also, the upper left node of the node group adjacent to the right side of the item name node d8, that is, the vertex of the node d9 overlaps with the vertex of the item name node d8, and the lower left end of the node group adjacent to the right side of the item name node d8. , That is, the vertex of the node d33 overlaps with the vertex of the item name node d8. Accordingly, the inclusion relationship acquisition unit 214 determines that the item name node d8 includes a node group adjacent to the right side of the item name node d8.

  Note that the format graph to be processed by the inclusion relation acquisition unit 214 has been deleted by the first deletion unit 208 but not by the second deletion unit 209. is there. For this reason, for example, the edge representing the adjacency relationship between the node d40 and the node d47 is not deleted. Therefore, the inclusion relationship acquisition unit 214 determines that the item name node d40 includes the node d47 in the horizontal direction.

  The enumeration matching unit 213 specifies a logical relationship related to enumeration of synthesis targets from which logical relationships related to nodes included in the table list 301 are excluded from the logical relationships acquired by the vertical enumeration acquisition unit 211 and the horizontal enumeration acquisition unit 212. Specifically, the enumeration matching unit 213 deletes the logical relationship included in the table list 301 among the logical relationships of each row of the enumeration list 302. The enumeration matching unit 213 is an example of a second specifying unit. For example, in FIG. The logical relationship of the row 6 (the parent is the node d14, the children are the nodes d15 and d16, the direction is horizontal) is shown in No. 6 of FIG. 6 coincides with the logical relationship of the row 6, the enumeration matching unit 213 displays No. 6 in FIG. Delete the logical relationship of line 6.

[Composition process]
The item name synthesizing unit 216 represents the inclusion graph as tree-structured data as shown in FIG. FIG. 13 is a diagram for explaining the inclusion graph. The data of the tree structure in FIG. 13 is based on the inclusion graph in FIG. Here, since the inclusion graph of FIG. 12 expresses the inclusion relationship of each node as a set, the inclusion graph also includes the logical relationship between the item names in the table. At this time, if the nested structure type includes an orthogonal table, the parent node of the orthogonal table becomes the starting node, and therefore the structure cannot be acquired correctly only from the inclusion graph.

  Therefore, the inter-item name composition unit 216 is based on the inclusion relationships acquired by the inclusion relationship acquisition unit 214 based on the inclusion relationships that are not included in either the logical relationship related to the table to be combined or the logical relationship related to the enumeration of the synthesis target. Create tree structure data. That is, the item name synthesizing unit 216, when forming the inclusion relationship from the main node to the subordinate node as tree structure data, up to the starting node of the table list or the item name node of the enumeration list is a tree node Are synthesized into tree-structured data, and the subsequent nodes are not synthesized.

  For example, the item name synthesizing unit 216 includes, among the nodes included in the synthesis target table and the synthesis target enumeration, a starting node of the orthogonal table, a starting node of the vertical table, and a node group connected to the right side of the starting node. The tree structure data includes the uppermost node group, the leftmost node group of the starting node of the horizontal table and the node group connected to the lower side of the starting node, and the enumerated item name node. . That is, the inter-item name composition unit 216 selects the uppermost node group (least_tops) among the starting node of the vertical table and the right node of the starting node, and the most of the starting node of the horizontal table and the lower node of the starting node. Nodes other than the node group (least_lefts) on the left side, the origin node of the orthogonal table, and the item name node of the enumeration are not synthesized with the tree structure data.

  In the example of FIG. 12, the item name synthesizing unit 216 synthesizes nodes d1, d9, d36, d38, d40, d41, d42, d47, d48, and d49 into tree-structured data. For example, the logical relationship in which the node d9 is a parent and the node d10 is a child is included in the table list of FIG. Similarly, since the logical relationship regarding the nodes included in the orthogonal table starting from the node d9 is included in the table list of FIG. 9, among the nodes included in the table list of FIG. Only the node d9 that is the starting node is synthesized into the tree-structured data.

  Further, in the example of FIG. 10, the item name synthesizing unit 216 synthesizes the nodes d48 and d49 up to tree structure data. For example, the logical relationship in which the node d48 is a parent and the node d50 is a child, and the logical relationship in which the node d49 is a parent and the node d51 is a child are included in the table list of FIG. For this reason, among the nodes included in the table list of FIG. 10, the item name synthesizing unit 216 synthesizes the tree-structured data with the node d48 that is the starting node and the node below the node d48 being the most. Only the node d49 which is the node (group) on the left side.

  The node d8 is a starting node in the table list of FIG. 8, but the table list of FIG. 8 has been deleted by the table matching unit 207 as described above. Therefore, the item name synthesizing unit 216 does not regard the node d8 as a starting node.

  Further, the table synthesis unit 217 creates tree structure data based on the logical relationship regarding the synthesis target table, and synthesizes the tree structure data and the tree structure data created by the item name synthesis unit 216. Create tree structure data. Specifically, the table synthesizing unit 217 represents a logical relationship acquired by the vertical table acquisition unit 204, a logical relationship acquired by the horizontal table acquisition unit 205, and a logical relationship acquired by the orthogonal table acquisition unit 206. Structure data is created for each table and combined with the tree structure data of the item name synthesizing unit 216. For example, the table synthesizing unit 217 creates tree-structured data shown in the branches after the node d9 in FIG. 14 based on the list shown in FIG. FIG. 14 is an example of tree structure data.

  In addition, the enumeration synthesis unit 218 creates tree structure data based on the logical relationship regarding the enumeration of synthesis targets, and synthesizes the tree structure data and the tree structure data created by the table synthesis unit 217. Create data for. For example, the enumeration synthesis unit 218 creates tree structure data shown in the branches after the nodes d2, d4, d6, d36, d38, d41, d42, d48, and d49 in FIG. FIG. 14 is an example of tree structure data.

  For example, the enumeration synthesis unit 218 synthesizes the tree structure data created by the item name synthesis unit 216 and the tree structure data created by the table synthesis unit 217 based on the enumeration list of FIG. Create the tree structure data shown.

  The adding unit 219 adds a root node that defines the tree structure to the tree structure data created by the table synthesis unit 217. For example, the adding unit 219 adds the root node “form1” to the data of the tree structure. In addition, since the root node is added to indicate the logical relationship that forms the format, the adding unit 219 is not essential if the information that configures the format is not necessary.

  In addition, the tree node that is each node of the tree structure data includes, for example, information for identifying child and parent nodes in addition to information acquired from form format information such as item name or item value character strings, Information indicating that the adjacent direction to the parent node, the tree node is included in the table, and the like.

[Process of First Embodiment]
First, processing for generating a form graph will be described. The process for generating the form graph is a process executed as a precondition before the process by the logical relationship recognition apparatus 10. Here, it is assumed that a form graph is generated by a form graph generation unit (not shown). The form graph generation unit may be provided in the logical relationship recognition device 10 or may be provided in another device.

  First, the overall flow of processing of the form graph generation unit will be described with reference to FIG. FIG. 15 is a flowchart showing a flow of processing for generating a form graph. As shown in FIG. 15, the form graph generation unit first reads a form (step S11). Next, the form graph generation unit acquires a node from the read form (step S12). Then, the form graph generation unit acquires the adjacent relationship between the nodes (step S13). Finally, the form graph generation unit outputs the generated form graph (step S14). Note that the form graph is a graph in which information about an area representing an item name or item value of the form is represented as a node, and the adjacent relationship between the nodes is represented as an edge.

  Next, the process for acquiring a node (step S12) will be described with reference to FIG. FIG. 16 is a flowchart showing a flow of processing for acquiring a node. As shown in FIG. 16, the form graph generation unit first reads a form (step S21). Next, the form graph generation unit acquires the range of each ruled line frame (step S22). Then, the form graph generation unit repeats the following processing until all ruled line frames are converted into nodes (steps S23 and S27).

  First, the form graph generation unit generates a node from the range of the ruled line frame (step S24). Next, the form graph generation unit sets the ruled line frame flag of the generated node to true (step S25). Then, the form graph generation unit adds the generated node to the node set N (step S26).

  Then, the form graph generation unit acquires the range of the character string group outside the ruled line frame (step S28). Here, the form graph generation unit repeats the following processing until the range of all character string groups is converted into nodes (steps S29 and S33).

  First, the form graph generation unit generates a node from the range of character string groups when the range of all character string groups is a ruled line frame (step S30). Next, the form graph generation unit sets the ruled line frame flag of the generated node to false (step S31). Then, the form graph generation unit adds the generated node to the node set N (step S32). Finally, the form graph generation unit outputs the node set N (step S34).

  Next, processing (steps S24 and S30) for generating a node from the range of the ruled line frame will be described with reference to FIG. FIG. 17 is a flowchart showing a flow of processing for generating a node. As illustrated in FIG. 17, the form graph generation unit reads the range of the form graph and the ruled line frame, and sets the range of the ruled line frame as area (step S41). Next, the form graph generation unit generates a new node in the form graph (step S42), and sets the area range information and the holding information in the new node (step S43). Then, the form graph generation unit indexes the new node (step S44) and outputs the new node (step S45).

  Next, the process for acquiring the adjacency relationship (step S13) will be described with reference to FIG. FIG. 18 is a flowchart showing a flow of processing for acquiring the adjacency relationship. As shown in FIG. 18, the form graph generation unit first reads a node set (step S51). Next, the form graph generation unit repeats the following processing for node 1 which is each node of the read node set (steps S52 and S55).

  First, the form graph generation unit acquires a group of nodes adjacent vertically and horizontally (step S53). Next, the upper, lower, left and right adjacent edges of the node are stored in the adjacent graph (step S54). Finally, the form graph generation unit outputs an adjacent graph (step S56).

  Next, the process (step S53) for acquiring adjacent node groups will be described with reference to FIG. FIG. 19 is a flowchart showing a flow of processing for acquiring adjacent node groups. As shown in FIG. 19, first, the form graph generation unit reads a node set and node 1. The form graph generation unit sets the node set to N (step S61). Here, the following processing is repeated for the node 2 which is each node of the node set N (steps S62 and S73).

  First, when the node 1 and the node 2 are the same (step S63, true), the form graph generation unit sets the node 2 as a node not adjacent to the node 1 (step S72). When the node 1 and the node 2 are not the same (step S63, false), the form graph generation unit determines that the node 2 is adjacent to the left of the node 1 when the left adjacent condition is satisfied (step S64, true). If the right adjacent condition is satisfied (step S66, true), it is determined that the node 2 is adjacent to the right of the node 1 (step S67), and the upper adjacent condition is satisfied. Is satisfied (step S68, true), it is determined that the node 2 is adjacent to the node 1 (step S69), and the lower adjacent condition is satisfied (step S70, true). Determines that node 2 is adjacent to node 1 (step S71). Finally, the form graph generation unit outputs an adjacent edge representing the adjacent relationship between each node of node 1 (step S74).

  Here, an example of the adjacency condition in each direction will be described. The coordinates of the upper left, upper right, lower left, and lower right vertices of node 1 are node 1 (x1, y1), node 1 (x2, y1), node 1 (x1, y2), and node 1 (x2, y2), respectively. . In addition, the coordinates of the upper left, upper right, lower left, and lower right vertices of node 2 are respectively node 2 (x1, y1), node 2 (x2, y1), node 2 (x1, y2), and node 2 (x2, y2). And Further, margin is a constant (for example, 0) representing a margin between ruled line frames set in advance.

At this time, each adjacency condition is expressed as follows as an example.
(Adjacent condition on the left)
Node 2. y1 ≧ node 1. y1
& Node 2. y2 ≧ node 1. y2
& 0 ≦ node 1. x1-node2. x2 ≦ margin
(Adjacent condition on the right)
Node 2. y1 ≧ node 1. y1
& Node 2. y2 ≧ node 1. y2
& 0 ≦ node 2. x1-node1. x2 ≦ margin
(Upper adjacent condition)
Node 2. x1 ≧ node 1. x1
& Node 2. x2 ≧ node 1. x2
& 0 ≦ node 1. y1-node2. y2 ≦ margin
(Lower adjacent condition)
Node 2. x1 ≧ node 1. x1
& Node 2. x2 ≧ node 1. x2
& 0 ≦ node 2. y1-node1. y2 ≦ margin

  In addition, since the form may include a plurality of forms, the form graph is divided into form graphs. Generation of the style graph will be described with reference to FIG. FIG. 20 is a flowchart showing a flow of processing for generating a style graph. As shown in FIG. 20, the form graph generation unit first reads a form graph (step S81). Next, the form graph generation unit deletes all inclusion relationships whose nodes are item values from the form graph (step S82).

  Then, the form graph generation unit designates an arbitrary node among the unclassified nodes as the node X (step S83). The form graph generation unit obtains a connected graph starting from the node X (step S84). If there is a node that is not classified in the connected graph (step S85, true), the form graph generation unit further designates a node that is not classified as the node X (step S83). When there is no node that is not classified into the connected graph (step S85, false), the form graph generation unit outputs the obtained connected graph as a style graph group (step S86).

  Next, the processing flow of the logical relationship recognition apparatus 10 will be described. First, processing of the extraction unit 201 will be described with reference to FIG. FIG. 21 is a flowchart showing the flow of processing of the extraction unit. As shown in FIG. 21, first, the extraction unit 201 reads a style graph and a visual expression rule after analysis (step S91). Next, the extraction unit 201 selects a node group that satisfies the condition of the post-analysis visual expression rule from the style graph (step S92). Next, the extraction unit 201 sets the item name attribute of the selected node group to the item name, that is, sets the value of the flag indicating the item name to true (step S93). Finally, the extraction unit 201 returns the style graph as G (step S94).

  Next, processing of the analysis unit 202 will be described with reference to FIG. FIG. 22 is a flowchart showing the flow of processing of the analysis unit. The analysis unit 202 creates a post-analysis visual expression rule based on the visual expression rule. As shown in FIG. 22, first, the analysis unit 202 reads a visual expression rule group (step S101). Next, the analysis unit 202 generates an Array or Hash type variable rule_list (step S102). Thereafter, the analysis unit 202 processes the read visual expression rules one by one (steps S103 and S107).

  First, the analysis unit 202 analyzes the condition of the visual expression rule (step S104). Next, the analysis unit 202 analyzes the action of the visual expression rule (step S105). Then, the analysis unit 202 stores the analyzed conditions and actions in the rule_list as post-analysis visual expression rules (step S106). The analysis unit 202 processes all visual expression rules, and then outputs a rule_list in which the post-analysis visual expression rule group is stored (step S108).

  Next, the process of the table classification | category part 203 is demonstrated using FIG. FIG. 23 is a flowchart showing the flow of processing of the table classification unit. As shown in FIG. 23, first, the table classification unit 203 reads the style graph, the target node that is the upper left node of the style graph, and the table list 301 (step S31a). Here, when the target node is not an item name or the target node is already described in the table list 301 as a parent or child (step S32a, true), the table classification unit 203 sets the table flag to “do not process”. (Step S42a). The table flag is a variable indicating the table classification as one of “vertical table”, “horizontal table”, “orthogonal table”, and “not processed”.

  When the target node is an item name and the target node is not described in the table list 301 as a parent or a child (step S32a, false), the table classification unit 203 connects to the lower side of the target node. The node group is stored in bottoms, and the node group connected to the right side of the target node is stored in rights (step S33a).

  In the example of FIG. 5, when the target node is the node d9, the nodes d10, d11, and d12 are stored in the rights, and the nodes d13, d14, d17, d20, d21, d24, d27, d30, and d33 are stored in the bottoms. Stored.

  Then, the table classification unit 203 sets flg1 to true when all the rights are item names, and sets flg1 to false otherwise. Also, the table classification unit 203 sets flg2 to true when the rights region is the same height as the target node, and sets flg2 to false otherwise (step S34a). The table classification unit 203 sets flg_r to true when both flg1 and flg2 are true, and sets flg_r to false otherwise.

  Also, the table classification unit 203 sets flg3 to true if bottoms are all item names, and sets flg3 to false otherwise. Also, the table classification unit 203 sets flg4 to true when the bottoms region has the same width as the target node, and sets flg4 to false otherwise (step S35a). Also, the table classification unit 203 sets flg_b to true when both flg3 and flg4 are true, and sets flg_b to false otherwise.

  When both flg_r and flg_b are true (step S36a is true and step S37a is true), the table classification unit 203 sets the table flag to “orthogonal table” (step S38a). When flg_r is true and flg_b is false (step S36a is true and step S37a is false), the table classification unit 203 sets the table flag to “vertical table” (step S39a). When flg_r is false and flg_b is true (step S36a is false and step S40a is true), the table classification unit 203 sets the table flag to “horizontal table” (step S41a). When both flg_r and flg_b are false (step S36a is false and step S40a is false), the table classification unit 203 sets the table flag to “not process” (step S42a). Finally, the table classification unit 203 outputs a table flag (step S43a).

  For example, in the example starting from the node d9 in FIG. 5, all the rights (nodes d10, d11, and d12) have the same item name and the height of the area of the rights is the same as that of the node d9. , Flg1 and flg2 are set to true. Further, since bottoms (nodes d13, d14, d17, d20, d21, d24, d27, d30, and d33) are all item names and the width of the bottoms area is the same as that of the node d9, the table classification unit 203 Set flg3 and flg4 to true. Thus, since both flg_r and flg_b are true, the table classification unit 203 sets the table flag to “orthogonal table”. As a result, the table starting from the node d9 in FIG. 5 is classified as an orthogonal table.

  Next, processing of the vertical table acquisition unit 204 will be described with reference to FIG. FIG. 24 is a flowchart showing the flow of processing of the vertical table acquisition unit. As shown in FIG. 24, first, the vertical table acquisition unit 204 reads a starting point node and a style graph (step S51a). Next, the vertical table acquisition unit 204 creates a new table list (step S52a). Next, the vertical table acquisition unit 204 stores, in last_tops, the uppermost node group among the node groups connected to the starting node and the right side of the starting node (step S53a).

  Here, the vertical table acquisition unit 204 performs the following processing for each of the nodes included in the least_tops (steps S54a and S57a). First, the vertical table acquisition unit 204 acquires a vertical table logical relationship (step S55a). Details of the process of acquiring the vertical table logical relationship will be described later. Next, the vertical table acquisition unit 204 adds a logical relationship to the table list (step S56a). The vertical table acquisition unit 204 outputs the table list after processing all the nodes (step S58a).

  Next, the process of the horizontal table acquisition unit 205 will be described with reference to FIG. FIG. 25 is a flowchart showing the flow of processing of the horizontal table acquisition unit. As shown in FIG. 25, first, the horizontal table acquisition unit 205 reads a starting point node and a style graph (step S61a). Next, the horizontal table acquisition unit 205 creates a new table list (step S62a). Next, the horizontal table acquisition unit 205 stores, in last_lefts, the leftmost node group among the node groups connected to the origin node and the origin node (step S63a).

  Here, the horizontal table acquisition unit 205 performs the following processing for each of the nodes included in the last_lefts (steps S64a and S67a). First, the horizontal table acquisition unit 205 acquires a horizontal table logical relationship (step S65a). Details of the process of acquiring the horizontal and logical relation will be described later. Next, the horizontal table acquisition unit 205 adds a logical relationship to the table list (step S66a). The horizontal table acquisition unit 205 outputs the table list after processing all the nodes (step S68a).

  For example, the horizontal table acquisition unit 205 outputs a table list as shown in FIG. 10 based on the style graph corresponding to the table starting from the node d48 in FIG. At this time, the horizontal table acquisition unit 205 stores the node d49 in the last_lefts.

  Next, the processing of the orthogonal table acquisition unit 206 will be described with reference to FIG. FIG. 26 is a flowchart showing the flow of processing of the orthogonal table acquisition unit. As shown in FIG. 26, first, the orthogonal table acquisition unit 206 reads the start node and the style graph (step S71a). Next, the orthogonal table acquisition unit 206 creates a new table list (step S72a).

  Next, the orthogonal table acquisition unit 206 stores the uppermost node group among the node groups connected to the right side of the starting node in last_tops (step S73a). Here, the orthogonal table acquisition unit 206 performs the following process for each of the nodes included in the least_tops (steps S74a and S77a). First, the orthogonal table acquisition unit 206 acquires a vertical table logical relationship (step S75a). Next, the orthogonal table acquisition unit 206 adds a logical relationship to the table list (step S76a).

  Next, the orthogonal table acquisition unit 206 stores the leftmost node group among the node groups connected to the lower side of the origin node in last_lefts (step S78a). Here, the orthogonal table acquisition unit 206 performs the following processing for each of the nodes included in the least_lefts (steps S79a and S82a). First, the orthogonal table acquisition unit 206 acquires a horizontal table logical relationship (step S80a). Next, the orthogonal table acquisition unit 206 adds a logical relationship to the table list (step S81a). The orthogonal table acquisition unit 206 outputs the table list after processing all the nodes (step S83a).

  For example, the orthogonal table acquisition unit 206 outputs a table list as shown in FIG. 9 based on the style graph corresponding to the table starting from the node d9 in FIG. At this time, the orthogonal table acquisition unit 206 stores the node d10 in the least_tops. Further, the orthogonal table acquisition unit 206 stores the nodes d13, d20, d27, d30, and d33 in last_lefts.

  Next, processing for acquiring a vertical logical relationship, that is, vertical table logical relationship acquisition processing (step S55a in FIG. 24 and step S75a in FIG. 26) will be described with reference to FIG. FIG. 27 is a flowchart illustrating a flow of processing for acquiring a vertical logical relationship. The vertical table logical relationship acquisition processing is performed by the vertical table acquisition unit 204 or the orthogonal table acquisition unit 206. Here, an example in which the orthogonal table acquisition unit 206 performs vertical table logical relationship acquisition processing will be described. In addition, when the vertical table acquisition unit 204 performs the vertical table logical relationship acquisition processing, the processing content is the same as when the orthogonal table acquisition unit 206 performs the vertical table logical relationship acquisition processing.

  As shown in FIG. 27, first, the orthogonal table acquisition unit 206 reads the target node and the style graph (step S101a). And the orthogonal table acquisition part 206 produces | generates the variable pair showing a new logical relationship (step S102a). Next, the orthogonal table acquisition unit 206 stores the node set next to the target node in next_b (step S103a).

  Here, when the number of nodes of next_b is two or more or next_b is an item name (step S104a, true), the orthogonal table acquisition unit 206 sets a parent as a target node, a child as a next_b, and a direction. Is added to the vertical (step S105a).

  Then, the orthogonal table acquisition unit 206 recursively performs vertical table logical relationship acquisition processing for each node (crt) included in next_b (steps S106a and S108a). The orthogonal table acquisition unit 206 stores the logical relationship group of the item name and the item value in the pair, stores the style graph in G, and replaces the target node with crt (step S107a). Then, the orthogonal table acquisition unit 206 returns to step S103a, and stores the node set adjacent to the target node in next_b.

  Further, when the number of nodes of next_b is not two or more and any of the next_b is not an item name (step S104a, false), the orthogonal table acquisition unit 206 stores the lower node set of the target node in bottoms. (Step S109a). Then, the orthogonal table acquisition unit 206 adds a logical relationship in which the parent is the target node, the child is bottoms, and the direction is vertical (step S110a). The orthogonal table acquisition unit 206 outputs the logical relation set (pair) after performing processing for all target nodes (step S111a).

  Next, processing for acquiring a horizontal logical relationship, that is, horizontal table logical relationship acquisition processing (step S65a in FIG. 25 and step S80a in FIG. 26) will be described with reference to FIG. FIG. 28 is a flowchart showing a flow of processing for acquiring a horizontal logical relationship. The horizontal table logical relationship acquisition processing is performed by the horizontal table acquisition unit 205 or the orthogonal table acquisition unit 206. Here, an example in which the orthogonal table acquisition unit 206 performs a horizontal table logical relationship acquisition process will be described. Note that the processing contents are the same when the horizontal table acquisition unit 205 performs the horizontal table logical relationship acquisition processing as when the orthogonal table acquisition unit 206 performs the horizontal table logical relationship acquisition processing.

  As shown in FIG. 28, first, the orthogonal table acquisition unit 206 reads the target node and the style graph (step S201a). And the orthogonal table acquisition part 206 produces | generates the variable pair showing a new logical relationship (step S202a). Next, the orthogonal table acquisition unit 206 stores the node set on the right side of the target node in next_r (step S203a).

  Here, when the number of nodes of next_r is two or more, or when next_r is all item names (step S204a, true), the orthogonal table acquisition unit 206 sets the parent as the target node, the child as the next_r, and the direction. Is added to the side (step S205a).

  Then, the orthogonal table acquisition unit 206 recursively performs a horizontal table logical relationship acquisition process for each node (crt) included in next_r (steps S206a and S208a). The orthogonal table acquisition unit 206 stores the logical relationship group of the item name and item value in the pair, stores the style graph in G, and replaces the target node with crt (step S207a). Then, the orthogonal table acquisition unit 206 returns to step S203a, and stores the node set on the right side of the target node in next_r.

  If the number of nodes in next_r is not two or more and any of the next_r is not an item name (step S204a, false), the orthogonal table acquisition unit 206 stores the right node set of the target node in rights. (Step S209a). Then, the orthogonal table acquisition unit 206 adds a logical relationship in which the parent is the target node, the child is rights, and the direction is horizontal (step S210a). The orthogonal table acquisition unit 206 outputs the logical relation set (pair) after performing processing for all target nodes (step S211a).

  Here, the processing of the table matching unit 207 will be described with reference to FIG. FIG. 29 is a flowchart showing the flow of processing of the table matching unit. The table matching unit 207 first reads a style graph and a table list (step S111). Next, the table matching unit 207 acquires a node group children having no children from the table list (step S112). When the item name is included in the children (step S113, true), the table matching unit 207 deletes the table from the table list (step S114). If the item name is not included in the children (step S113, false), the table matching unit 207 leaves the table in the table list.

  Next, the processing of the first deletion unit 208 will be described with reference to FIG. FIG. 30 is a flowchart showing the flow of processing of the first deletion unit. First, the first deletion unit 208 reads a style graph (step S31b).

  Next, the first deletion unit 208 performs the following process for each node of the style graph (steps S32b and S37b). First, when the number of nodes adjacent in the left direction is two or more (step S33b, true), the first deletion unit 208 deletes the adjacent relationship between the node adjacent in the left direction and the target node (step S34b). ). If the number of nodes adjacent in the left direction is not two or more (step S33b, false), the first deletion unit 208 does not delete the adjacent relationship.

  Next, when the number of nodes adjacent in the upward direction is two or more (step S35b, true), the first deletion unit 208 deletes the adjacent relationship between the node adjacent in the upward direction and the target node (step S35b). S36b). If the number of nodes adjacent in the upward direction is not two or more (step S35b, false), the first deletion unit 208 does not delete the adjacent relationship. The 1st deletion part 208 outputs a style graph, after processing about all object nodes (Step S38b).

  Next, the process of the second deletion unit 209 will be described with reference to FIG. FIG. 31 is a flowchart showing the flow of processing of the second deletion unit. First, the second deletion unit 209 reads the style graph and the item name node list (step S41b). The item name node list is a list of item name nodes extracted by the extraction unit 201.

  Next, the second deletion unit 209 performs the following processing for each item name node in the list of item name nodes (steps S42b and S47b). First, when the item value node is adjacent in the left direction (step S43b, true), the second deletion unit 209 deletes the adjacent relationship between the node adjacent in the left direction and the target node (step S44b). When the item value node is not adjacent in the left direction (step S43b, false), the second deletion unit 209 does not delete the adjacent relationship.

  Next, when the item value node is adjacent in the upward direction (step S45b, true), the second deletion unit 209 deletes the adjacent relationship between the node adjacent in the upward direction and the target node (step S46b). If the item value node is not adjacent in the upward direction (step S45b, false), the second deletion unit 209 does not delete the adjacent relationship. The 2nd deletion part 209 outputs a style graph, after processing about all object nodes (Step S48b).

  Next, processing of the enumeration classification unit 210 will be described with reference to FIG. FIG. 32 is a flowchart showing the flow of processing of the enumeration classification unit. From now on, the relationship with the node obtained by tracing the node adjacent to the right side from the target node is “connected to the right side”, and the relationship with the node obtained by tracing the node adjacent to the lower side from the target node Called “connect to the bottom”. As shown in FIG. 32, first, the enumeration classification unit 210 reads a style graph and a target node (step S51b). If the target node is not an item name (step S52b, true), the enumeration classification unit 210 sets the enumeration flag to “no enumeration” (step S58b). The enumeration flag is a variable indicating the classification of each target node as one of “vertical enumeration”, “horizontal enumeration”, and “no enumeration”. In addition, the enumeration includes a composite type of vertical enumeration and horizontal enumeration, and an enumeration nested structure. These enumerations can be expressed by a combination of vertical enumeration and horizontal enumeration.

  When the target node is an item name (step S51b, false), the enumeration classification unit 210 stores a node group connected to the lower side of the target node in bottoms, and sets the node group connected to the right side of the target node to the rights. (Step S53b).

  Here, when the item name is not included in rights and the number of rights is 1 (step S54b, true), the enumeration classification unit 210 sets the enumeration flag to “horizontal enumeration” (step S55b). When the item name is included in rights, or when the number of rights is not 1 (step S54b, false), the enumeration classification unit 210 performs the following processing.

  When the item name is not included in bottoms and the number of bottoms is 1 (step S56b, true), the enumeration classification unit 210 sets the enumeration flag to “vertical enumeration” (step S57b). When the item name is included in bottoms, or when the number of bottoms is not 1 (step S56b, false), the enumeration classification unit 210 sets the enumeration flag to “no enumeration” (step S58b). Finally, the enumeration classification unit 210 outputs an enumeration flag (step S59b).

  Next, the processing of the vertical enumeration acquisition unit 211 will be described with reference to FIG. FIG. 33 is a flowchart showing the flow of processing of the vertical enumeration acquisition unit. As shown in FIG. 33, first, the vertical enumeration acquiring unit 211 reads a style graph and a target node (step S61b). Next, the vertical enumeration acquiring unit 211 stores a node group connected to the lower side of the target node in bottoms (step S62b). Here, when the item name is included in bottoms (step S63b, true), the vertical enumeration acquiring unit 211 ends the process. If the item name is not included in bottoms (step S63b, false), the vertical enumeration acquiring unit 211 acquires a logical relationship in which the parent is the target node, the child is bottoms, and the direction is vertical (step S64b). Then, the vertical enumeration acquisition unit 211 adds the acquired logical relationship to the enumeration list 302 (step S65b).

  Next, the process of the horizontal enumeration acquisition unit 212 will be described with reference to FIG. FIG. 34 is a flowchart illustrating a process flow of the horizontal enumeration acquisition unit. As shown in FIG. 34, the horizontal enumeration obtaining unit 212 first reads the style graph and the target node (step S71b). Next, the horizontal enumeration acquisition unit 212 stores a node group connected to the right side of the target node in rights (step S72b). If the item name is included in rights (step S73, true), the horizontal enumeration acquisition unit 212 ends the process. If the item name is not included in rights (step S73b, false), the horizontal enumeration acquisition unit 212b acquires a logical relationship in which the parent is the target node, the child is rights, and the direction is horizontal (step S74b). Then, the horizontal enumeration acquisition unit 212 adds the acquired logical relationship to the enumeration list 302 (step S75b).

  Next, processing of the inclusion relationship acquisition unit 214 will be described using FIG. FIG. 35 is a flowchart illustrating a process flow of the inclusion relationship acquisition unit. The inclusion relationship acquisition unit 214 first reads a style graph (step S81b).

  Here, the inclusion relationship acquisition unit 214 performs the following processing for each node included in the style graph (steps S82b and S87b). First, when the node itself includes the item name and the node of the item name in the node adjacent to the right (step S83b, true), the inclusion relationship acquisition unit 214 acquires the right inclusion relationship (step S84b). Details of the process of acquiring the right inclusion relationship will be described later. Also, if the node itself is not an item name, or if the node adjacent to the right direction does not include the item name node (step S83b, false), the inclusion relationship acquisition unit 214 does not acquire the right inclusion relationship.

  Next, when the node itself includes the item name and the node of the item name in the node adjacent in the downward direction (step S85b, true), the inclusion relationship acquisition unit 214 acquires the lower inclusion relationship (step S86b). . Details of the process of acquiring the lower inclusion relationship will be described later. If the node itself is not an item name, or if the node adjacent to the downward direction does not include the item name node (step S85b, false), the inclusion relationship acquisition unit 214 does not acquire the lower inclusion relationship. The inclusion relationship acquisition unit 214 processes all the nodes, and then outputs the acquired inclusion relationship as an inclusion relationship list (step S88b).

  Next, processing for acquiring the right inclusion relationship will be described with reference to FIG. FIG. 36 is a flowchart showing the flow of processing for acquiring the right inclusion relationship. As shown in FIG. 36, first, the inclusion relationship acquisition unit 214 reads the target node and the style graph (step S101b). Next, the inclusion relationship acquisition unit 214 stores the minimum value of the y coordinate of the node group on the right side of the target node in min_y, and stores the maximum value of the y coordinate of the node group on the right side of the target node in min_y. (Step S102b).

  If the y-coordinate range of the target node matches the y-coordinate range of the right node group of the target node (step S103b, true), the inclusion relationship acquisition unit 214 sets the list to the right node of the target node. The group is stored (step S107b). The inclusion relationship acquisition unit 214 calculates the y-coordinate range of the node group on the right side of the target node using min_y and max_y. Then, the inclusion relationship acquisition unit 214 performs the following processing for each node included in the list (steps S108b and S110b). The inclusion relationship acquisition unit 214 returns to step S102b and executes recursion processing every time each node r is set as a target node (step S109b).

  Further, when the y-coordinate range of the target node does not match the y-coordinate range of the right node group of the target node (step S103b, false), and the item value is not included in the included node ( In step S104b, false), the inclusion relationship acquisition unit 214 initializes the slave node set (list) included on the right side (step S105b). In addition, when the item value is included in the included node (step S104b, true), the inclusion relationship acquisition unit 214 does not perform initialization. Finally, the inclusion relationship acquisition unit 214 outputs list (step S106b).

  Next, processing for obtaining the lower inclusion relationship will be described with reference to FIG. FIG. 37 is a flowchart showing a flow of processing for acquiring the lower inclusion relationship. As shown in FIG. 37, first, the inclusion relationship acquisition unit 214 reads the target node and the style graph (step S151b). Next, the inclusion relationship acquisition unit 214 stores the minimum value of the x coordinate of the node group below the target node in min_x, and the maximum value of the x coordinate of the node group below the target node in max_x. Is stored (step S152b).

  Here, when the x-coordinate range of the target node matches the x-coordinate range of the lower node group of the target node (step S153b, true), the inclusion relationship acquisition unit 214 sets the lower side of the target node to the list. Are stored (step S157b). The inclusion relationship acquisition unit 214 calculates the x-coordinate range of the lower node group of the target node using min_x and max_x. Then, the inclusion relationship acquisition unit 214 performs the following processing for each node included in the list (steps S158b and S160b). The inclusion relationship acquisition unit 214 returns to step S152b and executes recursion processing every time each node b is set as a target node (step S159b).

  Also, when the x-coordinate range of the target node does not match the x-coordinate range of the lower node group of the target node (step S153b, false), and the item value is not included in the included node (Step S154b, false), the inclusion relationship acquisition unit 214 initializes a slave node set (list) included in the lower side (Step S155b). In addition, when the item value is included in the included node (step S154b, true), the inclusion relationship acquisition unit 214 does not perform initialization. Finally, the inclusion relationship acquisition unit 214 outputs list (step S156b).

  Next, processing of the inclusion graph generation unit 215 will be described with reference to FIG. FIG. 38 is a flowchart showing the flow of processing of the inclusion graph generation unit. As shown in FIG. 38, first, the inclusion graph generation unit 215 reads the style graph and the inclusion relation list (step S201b). Next, the inclusion graph generation unit 215 stores the vertical inclusion node acquired from the inclusion relation list in Nv, stores the horizontal inclusion node acquired from the inclusion relation list in Nh, and stores the new inclusion in IG. The inclusion graph set is stored (step S202b).

  Here, the inclusion graph generation unit 215 performs the following processing for each inclusion node i included in Nv and Nh (steps S203b and S213b). Hereinafter, when the target node is already assigned to the inclusion graph due to the inclusion relationship of other nodes, the target node is referred to as a “divided node”. First, when the inclusion node i has been divided (step S204b, true), the inclusion graph generation unit 215 proceeds to processing of the next inclusion node. If the inclusion node i has not been divided (step S204b, false), the inclusion graph generation unit 215 stores the inclusion node i and the subordinate item name node set in inc (step S205b).

  Here, when there is a divided node in inc (step S206b, true), the inclusion graph generation unit 215 searches the generated inclusion graph for a graph in which a slave node overlaps, and sets the inclusion node to n. Store (step S207b). Next, the inclusion graph generation unit 215 adds a node group that has not been added to the nodes in the inclusion graph starting from the inclusion node n (step S208b). Then, the inclusion graph generation unit 215 sets the inclusion direction from the inclusion node's inclusion direction or the list of item names (step S209b).

  On the other hand, when there is no divided node in inc (step S206b, false), the inclusion graph generation unit 215 generates a new inclusion graph (step S210b). Then, the inclusion graph generation unit 215 sets the inclusion direction from the direction included in the main node or the list of item names (step S211b), and adds a new inclusion graph to the inclusion graph set IG (step S212b). The inclusion graph generation unit 215 outputs the inclusion graph set IG after processing all i (Step S214b).

  As described above, the inclusion graph generation unit 215 preferentially processes a node having a large number of included nodes among nodes having an inclusion relationship. The inclusion graph generation unit 215 first acquires all inclusion relations in the style graph and searches for nodes in the inclusion direction.

  Here, the processing of the enumeration matching unit 213 will be described with reference to FIG. FIG. 39 is a flowchart showing the flow of processing of the enumeration matching unit. As shown in FIG. 39, the enumeration matching unit 213 first reads a style graph and an enumeration list (step S121). Next, when the logical relationship of the enumeration list is included in the table list (step S122, true), the enumeration list is deleted (step S123).

  Next, processing of the item name synthesizing unit 216 will be described with reference to FIG. FIG. 40 is a flowchart showing the flow of processing of the item name synthesizing unit. As shown in FIG. 40, first, the item name synthesizing unit 216 reads the style graph, the inclusion graph 303, the table list 301, the enumeration list 302, and the start point node group (step S301b). Next, the item name synthesizing unit 216 generates a new tree structure T (step S302b). The starting point node group is a set of nodes that are the starting points of the inclusion graph.

  Here, the item name synthesizing unit 216 performs the following processing for each start point node included in the start point node group (steps S303b and S319b). First, the item name synthesizing unit 216 obtains a node group whose node is the item name from the inclusion graph, and stores it in the children (step S304b). Next, the item name synthesizing unit 216 generates a new tree node t for the start point node whose child is child and whose parent is none (step S305b). The item name synthesizing unit 216 sets the target tree node to t (step S306b). Then, the item name synthesizing unit 216 adds to the tree structure without setting the type of the target tree node (step S307b).

  Here, when the target tree node is in the table list or the enumeration list (step S308b, false), the item name synthesizing unit 216 proceeds to the processing of the next start point node. When the target tree node is not in the enumeration list (step S308b, true) and there is a node set that the target tree node includes in the horizontal direction (step S309b, true), the inter-item name composition unit 216 The type of the target tree node is set to inclusion (step S310b). Then, the item name synthesizing unit 216 sets the next node in the horizontally included node set as the target tree node, returns to step S307b, and performs recursive processing (steps S311b, S312b, and S313b).

  If the target tree node is not in the table list or the enumeration list (step S308b, true), and there is no node set that the target tree node includes in the horizontal direction (step S309b, false), the target tree node If there is a node set included in the vertical direction (step S314b, true), the item name name synthesizing unit 216 sets the type of the target tree node to include (step S315b). Then, the item name synthesizing unit 216 sets the next node of the vertically included node set as the target tree node, returns to step S307b, and performs recursive processing (steps S316b, S317b, and S318b). The item name synthesizing unit 216 performs processing on all the start point nodes, and then outputs tree structure data (step S320b).

  When the target tree node is not in the enumeration list (step S308b, true), there is no node set that the target tree node includes in the horizontal direction (step S309b, false), and the target tree node is vertical. If there is no node set included in the direction (step S314b, false), the inter-name-name composition unit 216 proceeds to processing of the next start point node.

  Next, processing of the table synthesis unit 217 will be described with reference to FIG. FIG. 41 is a flowchart showing a process flow of the table synthesis unit. First, the table synthesis unit 217 reads the style graph and the table list 301 (step S301a). Next, the table composition unit 217 generates a new tree structure (step S302a). Then, the table synthesizing unit 217 acquires an ancestor (parentless node) from the table list (step S303a).

  Here, the table synthesizing unit 217 performs the following processing for each node included in the table list (steps S304a and S310a). First, the table synthesis unit 217 generates a new tree node (step S305a). Next, the table synthesizing unit 217 searches for the parent of the node from the table list and sets it as the parent of the tree node. If there is no parent, the table node is emptied (step S306a). Then, the table composition unit 217 sets the child of the tree node as the child of the table list (step S307a). Then, the table composition unit 217 sets the tree node type in the table (step S308a). The table synthesis unit 217 then synthesizes the tree node into a tree structure (step S309a). Finally, when processing is completed for each node in the table list, the table synthesis unit 217 outputs tree structure data (step S311a).

  Next, processing of the enumeration synthesis unit 218 will be described with reference to FIG. FIG. 42 is a flowchart showing the flow of processing of the enumeration synthesis unit. As shown in FIG. 42, the enumeration synthesis unit 218 first reads the style graph, the enumeration list 302, and the tree structure data (step S341b).

  Then, the enumeration synthesis unit 218 performs the following processing for each node of the enumeration list 302 (steps S342b and S347b). First, the enumeration synthesis unit 218 generates a new tree node (step S343b). Next, the enumeration composition unit 218 sets the children of the tree node as children of the enumeration list (step S344b). Then, the enumeration synthesis unit 218 sets the tree node type to enumeration (step S345b), and synthesizes the tree node data with the tree structure data (step S346b). The enumeration synthesis unit 218 outputs the tree structure data after performing the above processing for all the nodes (step S348b).

  Next, processing of the adding unit 219 will be described with reference to FIG. FIG. 43 is a flowchart showing the flow of processing of the adding unit. As shown in FIG. 43, the adding unit 219 first reads the style graph and tree structure data, stores the tree structure data in S, and stores the style graph in G (step S131). Next, the adding unit 219 stores a parentless node set in nds and an arbitrary character string in str (step S132). Here, if the character string outside the ruled line frame in the left or upward direction of nds can be acquired, the adding unit 219 stores the acquired character string in str (step S133).

  When a character string outside the ruled line frame is not found (step S134, false), the adding unit 219 uses the tree node having an arbitrary character string stored in str and having a type as a style as a parent of nds. (Step S135). When a character string outside the ruled line frame is found (step S134, true), the adding unit 219 has an arbitrary character string stored in str and uses a tree node whose type is a style as a parent of nds. It adds to data (step S136). Finally, the adding unit 219 outputs tree structure data (step S137).

  In addition, as an arbitrary character string, the character string may be changed in accordance with the count as in form # (i). In this case, since the part (i) changes in accordance with the count, the character string of the root node becomes “form1”, “form2”. Further, if there is a description such as “XX orthogonal table” outside the left or upper ruled line frame of the table, the character string of the root node may be set as “XX orthogonal table”.

[Effect of the first embodiment]
The extraction unit 201 represents information related to the area representing the item name or item value of the form as a node, and based on the graph representing the adjacency relationship between the nodes as an edge, the node satisfying a preset condition is selected. , It is extracted as an item name node which is a node of the area representing the item name. Based on the edge, the table classification unit 203 extracts a starting node, which is a node that is a starting point of the table, from the item name node, and the table having the starting node as the starting point is orthogonal with the item names existing at both the upper end and the left end. The table is classified into one of a table, a vertical table in which the item name exists at the upper end, and a horizontal table in which the item name exists at the left end.

  The orthogonal table acquisition unit 206 is a vertical logical relationship between item name nodes in the orthogonal table, a horizontal logical relationship between item name nodes in the orthogonal table, and a node other than the item name node and the item name node in the orthogonal table. The vertical logical relationship between the item value nodes and the horizontal logical relationship between the item name node and the item value node in the orthogonal table are acquired. The vertical table acquisition unit 204 acquires a vertical logical relationship between item name nodes in the vertical table and a vertical logical relationship between item name nodes and item value nodes in the vertical table. The horizontal table acquisition unit 205 acquires a horizontal logical relationship between item name nodes in the horizontal table and a horizontal logical relationship between item name nodes and item value nodes in the horizontal table. In addition, the table matching unit 207 specifies a table that excludes a table that satisfies a predetermined condition indicating that it is an inconsistent table from the orthogonal table, the vertical table, and the horizontal table.

  In addition, when a plurality of nodes are adjacent to each other in a predetermined direction of one node, the first deletion unit 208 deletes an edge representing the adjacency relationship between the one node and the plurality of nodes. Also, the second deletion unit 209 determines the adjacency relationship between the item name node and the item value node adjacent to the item value node that is the node of the area representing the item value in a predetermined direction among the item name nodes. Delete the representing edge.

  In addition, the vertical enumeration acquisition unit 211 and the horizontal enumeration acquisition unit 212, based on the graph in which the edge deletion is performed by the first deletion unit 208 and the second deletion unit 209, is performed between the item name node and the item value node. Get logical relationship between. The inclusion relationship acquisition unit 214 acquires the inclusion relationship between the item name nodes based on the graph in which the edge is deleted by the first deletion unit 208.

  In addition, the enumeration matching unit 213 identifies the logical relationship related to the enumeration of the synthesis targets excluding the logical relationship related to the nodes included in the table list 301 among the logical relationships acquired by the vertical enumeration acquiring unit 211 and the horizontal enumeration acquiring unit 212. To do.

  Further, the inter-item name composition unit 216 is based on the inclusion relationships that are not included in either the logical relationship related to the synthesis target table or the logical relationship related to the synthesis target enumeration among the inclusion relationships acquired by the inclusion relationship acquisition unit 214. Create tree structure data. Further, the table synthesis unit 217 creates tree structure data based on the logical relationship regarding the synthesis target table, and synthesizes the tree structure data and the tree structure data created by the item name synthesis unit 216. Create tree structure data. In addition, the enumeration synthesis unit 218 creates tree structure data based on the logical relationship regarding the enumeration of synthesis targets, and synthesizes the tree structure data and the tree structure data created by the table synthesis unit 217. Create data for.

  For this reason, according to this embodiment, even if the form includes a vertical table, a horizontal table, an orthogonal table, or a nested orthogonal table, the logical relationship between the item names of the form, and the item names and item values It becomes possible to accurately recognize the logical relationship between the two. Further, in the present embodiment, since the logical relationship can be acquired semi-automatically, it is possible to efficiently acquire and utilize the semi-structured data of the form.

  Further, according to the present embodiment, even if the form includes a vertical enumeration, a horizontal enumeration, a combined type of vertical enumeration and horizontal enumeration, or an enumerated type nested structure, the logical relationship between the item names of the forms, and The logical relationship between the item name and the item value can be accurately recognized. Further, in the present embodiment, since the logical relationship can be acquired semi-automatically, it is possible to efficiently acquire and utilize the semi-structured data of the form.

  Further, according to the present embodiment, the logical relationship included in the tree structure data is specified by the table matching unit 207 and the enumeration matching unit 213, so that the vertical enumeration, horizontal enumeration, combined vertical enumeration and horizontal enumeration, Or when the nested structure type includes vertical enumeration, horizontal enumeration, vertical enumeration and horizontal enumeration, vertical table, horizontal table, orthogonal table, nested vertical table, nested horizontal table, nested orthogonal table Even so, it is possible to accurately recognize the logical relationship between the item names of the form and the logical relationship between the item name and the item value.

  The table classification unit 203 includes a first condition that all the first node groups connected to the right side of the starting node are item names, and the height of the first node group is the same as the height of the starting node, and the starting point When the second condition that all the second node groups connected to the lower side of the node are item names and the width of the second node group is the same as the width of the starting node is satisfied, the starting node is set as the starting point If the table is classified as an orthogonal table and the first condition is satisfied and the second condition is not satisfied, the table starting from the starting node is classified into a horizontal table, the second condition is satisfied, and the first condition is satisfied. If the condition is not satisfied, the table starting from the starting node is classified into a vertical table.

  Further, the table matching unit 207 includes an orthogonal table having an item name node having no item value node on the lower side and the right side, and an item having no item value node on the lower side from the orthogonal table, the vertical table, and the horizontal table. A table excluding a vertical table having name nodes and a horizontal table having item name nodes having no item value nodes on the right side is specified.

  Thus, by using the adjacency, height and width of nodes, not only can the vertical table, horizontal table, and orthogonal table be classified correctly, but the item name node is erroneously treated as an item value node. Inconsistent tables such as those listed can be excluded.

  Among the nodes included in the synthesis target table and the synthesis target enumeration, the item name synthesizing unit 216 is the first of the orthogonal table starting node, the vertical table starting node, and the node group connected to the right side of the starting node. The node group on the upper side, the leftmost node group of the starting node in the horizontal table and the node group connected to the lower side of the starting node, and the item name node of the enumeration as the data of the first tree structure include. Thereby, it is possible to prevent the logical relationship from being redundantly included in the finally output tree-structured data.

  The orthogonal table acquisition unit 206 acquires a combination of the item name and the item name or item value on the right side of the item name as a logical relationship in the horizontal direction, and the item name or item value below the item name and the item name. Is obtained as a vertical logical relationship. The horizontal table acquisition unit 205 acquires a combination of an item name and an item name or item value on the right side of the item name as a logical relationship in the horizontal direction. The vertical table acquisition unit 204 acquires a combination of an item name and an item name or item value below the item name as a vertical logical relationship. Thus, the logical relationship can be accurately acquired by referring to the lower node in the vertical direction and the right node in the horizontal direction.

  In addition, when a plurality of nodes are adjacent to the left side or the upper side of one node, the first deletion unit 208 may delete an edge representing the adjacent relationship between the one node and the plurality of nodes. In addition, the second deletion unit 209 has an adjacency relationship between the item name node and the item value node adjacent to the item value node that is the node of the area representing the item value on the left or upper side of the item name nodes. You may delete the edge showing. In general, the positional relationship between item names in a form and between item names and item values is often left to right or top to bottom. For this reason, by setting the direction of the adjacent relationship to be deleted on the left side and the upper side, it is possible to deal with many forms.

  The inclusion relationship acquisition unit 214 has at least one of the first node groups adjacent to the right side of the first item name node as the item name node, and the heights of all the nodes included in the first node group. Is less than or equal to the height of the first item name node, and the vertex of the upper left node of the first node group and the vertex of the first item name node overlap, the first item name node Are included in the first node group. In addition, the inclusion relationship acquisition unit 214 has at least one of the second node groups adjacent to the lower side of the second item name node as an item name node, and all the nodes included in the second node group The width of the second item name node is less than or equal to the width of the second item name node, and the vertex of the upper left node of the second node group overlaps the vertex of the first item name node, the second item name It is determined that the node includes the second node group. Thus, the inclusion relation can be accurately recognized by utilizing the adjacent relation, height, and width of the nodes.

[Other Embodiments]
The target of the logical relationship recognition may be a Web screen or a system GUI as long as it can be formed into a form format. For example, an item name and an item value are acquired from a Web screen that makes a reservation for an air ticket on the Web as shown in FIG. 44, and the Web screen is subjected to logical relationship recognition processing by formatting it into a form format. It can be. FIG. 44 is a diagram for explaining another embodiment.

[System configuration, etc.]
Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. Furthermore, all or a part of each processing function performed in each device may be realized by a CPU and a program that is analyzed and executed by the CPU, or may be realized as hardware by wired logic.

  Also, among the processes described in this embodiment, all or part of the processes described as being performed automatically can be performed manually, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedure, control procedure, specific name, and information including various data and parameters shown in the above-described document and drawings can be arbitrarily changed unless otherwise specified.

[program]
As an embodiment, the logical relationship recognition apparatus 10 can be implemented by installing a logical relationship recognition program for executing the logical relationship recognition as package software or online software on a desired computer. For example, the information processing apparatus can function as the logical relation recognition apparatus 10 by causing the information processing apparatus to execute the above logical relation recognition program. The information processing apparatus referred to here includes a desktop or notebook personal computer. In addition, the information processing apparatus includes mobile communication terminals such as smartphones, mobile phones and PHS (Personal Handyphone System), and slate terminals such as PDA (Personal Digital Assistant).

  The logical relationship recognition apparatus 10 can also be implemented as a logical relationship recognition server device that uses a terminal device used by a user as a client and provides the client with the above-described service related to logical relationship recognition. For example, the logical relationship recognition server device is implemented as a server device that provides a logical relationship recognition service that takes a form as input and outputs tree structure data. In this case, the logical relationship recognition server device may be implemented as a Web server, or may be implemented as a cloud that provides the above-described service related to logical relationship recognition by outsourcing.

  FIG. 45 is a diagram illustrating an example of a computer in which the logical relationship recognition apparatus is realized by executing a program. The computer 1000 includes a memory 1010 and a CPU 1020, for example. The computer 1000 also includes a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These units are connected by a bus 1080.

  The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM 1012. The ROM 1011 stores a boot program such as BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1090. The disk drive interface 1040 is connected to the disk drive 1100. For example, a removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to a mouse 1110 and a keyboard 1120, for example. The video adapter 1060 is connected to the display 1130, for example.

  The hard disk drive 1090 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. That is, a program that defines each process of the logical relationship recognition apparatus 10 is implemented as a program module 1093 in which a code executable by a computer is described. The program module 1093 is stored in the hard disk drive 1090, for example. For example, a program module 1093 for executing processing similar to the functional configuration in the logical relationship recognition apparatus 10 is stored in the hard disk drive 1090. Note that the hard disk drive 1090 may be replaced by an SSD.

  The setting data used in the processing of the above-described embodiment is stored as program data 1094 in, for example, the memory 1010 or the hard disk drive 1090. Then, the CPU 1020 reads the program module 1093 and the program data 1094 stored in the memory 1010 and the hard disk drive 1090 to the RAM 1012 and executes them as necessary.

  The program module 1093 and the program data 1094 are not limited to being stored in the hard disk drive 1090, but may be stored in, for example, a removable storage medium and read out by the CPU 1020 via the disk drive 1100 or the like. Alternatively, the program module 1093 and the program data 1094 may be stored in another computer connected via a network (LAN, WAN (Wide Area Network), etc.). The program module 1093 and the program data 1094 may be read by the CPU 1020 from another computer via the network interface 1070.

DESCRIPTION OF SYMBOLS 10 Logical relationship recognition apparatus 20 Control part 30 Storage part 201 Extraction part 202 Analysis part 203 Table classification part 204 Vertical table acquisition part 205 Horizontal table acquisition part 206 Orthogonal table acquisition part 207 Table matching part 208 1st deletion part 209 2nd deletion part 209 2nd Deletion unit 210 Enumeration classification unit 211 Vertical enumeration acquisition unit 212 Horizontal enumeration acquisition unit 213 Enumeration matching unit 214 Inclusion relation acquisition unit 215 Inclusion graph generation unit 216 Inter-name name synthesis unit 217 Table synthesis unit 218 Enumeration synthesis unit 219 Addition unit 301 Table list 302 Enumeration list 303 Inclusion graph

Claims (8)

Based on a graph that represents information related to the area representing the item name or item value of the form as a node and the adjacent relationship between the nodes as an edge, a node that satisfies a preset condition is selected as the item name. An extraction unit that extracts as an item name node that is a node of an area that represents
Based on the edge, a starting node that is a node that is a starting point of the table is extracted from the item name node, and a table starting from the starting node is an orthogonal table having item names on both the upper end and the left end, and the upper end A table classification part that classifies the table into one of a vertical table in which the item name is present and a horizontal table in which the item name is present on the left end;
A vertical logical relationship between the item name nodes in the orthogonal table, a horizontal logical relationship between the item name nodes in the orthogonal table, and a node other than the item name node and the item name node in the orthogonal table. An orthogonal table acquisition unit for acquiring a vertical logical relationship between item value nodes, and a horizontal logical relationship between the item name node and the item value node in the orthogonal table;
A vertical table acquisition unit that acquires a vertical logical relationship between the item name nodes in the vertical table, and a vertical logical relationship between the item name node and the item value node in the vertical table;
A horizontal table acquisition unit for acquiring a horizontal logical relationship between the item name nodes in the horizontal table, and a horizontal logical relationship between the item name node and the item value node in the horizontal table;
A first specifying unit that specifies a table to be synthesized excluding a table that satisfies a predetermined condition indicating that the table is inconsistent from the orthogonal table, the vertical table, and the horizontal table;
A first deletion unit that deletes an edge representing an adjacency relationship between the one node and the plurality of nodes when a plurality of nodes are adjacent to each other in a predetermined direction of the one node;
Among the item name nodes, an item name node that is adjacent to an item value node that is a node of an area that represents an item value in a predetermined direction, and an edge that represents an adjacency relationship between the item value node are deleted Delete part,
A first acquisition unit that acquires a logical relationship between the item name node and the item value node based on the graph in which an edge is deleted by the first deletion unit and the second deletion unit. When,
A second acquisition unit that acquires an inclusion relationship between the item name nodes based on the graph in which an edge is deleted by the first deletion unit;
A second specifying unit for specifying a logical relationship related to enumeration of synthesis targets excluding a logical relationship related to a node included in the synthesis target table among the logical relationships acquired by the first acquisition unit;
Of the inclusion relationships acquired by the second acquisition unit, the first tree structure based on the inclusion relationship that is not included in either the logical relationship related to the table to be combined and the logical relationship related to the enumeration of the combination target. A first synthesis unit for creating data;
A tree structure data is created based on a logical relationship regarding the table to be synthesized, and a second tree structure data is created by synthesizing the tree structure data and the first tree structure data. A synthesis unit;
A tree structure data is created based on the logical relationship relating to the synthesis target enumeration, and a third tree structure data is created by synthesizing the tree structure data and the second tree structure data. A synthesis unit;
A logical relationship recognition apparatus comprising: The table classification unit includes a first condition that all the first node groups connected to the right side of the starting node are item names and the height of the first node group is the same as the height of the starting node. And the second condition is that all the second node groups connected to the lower side of the starting node are item names and the width of the second node group is the same as the width of the starting node. In this case, the table starting from the starting node is classified as an orthogonal table, and when the first condition is satisfied and the second condition is not satisfied, the table starting from the starting node is classified as a horizontal table. If the second condition is satisfied and the first condition is not satisfied, the table starting from the starting node is classified into a vertical table,
The first specifying unit includes, from the orthogonal table, the vertical table, and the horizontal table, an orthogonal table having the item name node where the item value node does not exist on either the lower side or the right side, and the item on the lower side. The logical relationship according to claim 1, wherein a vertical table having the item name node for which no value node exists and a horizontal table having the item name node for which the item value node does not exist on the right side are excluded. Recognition device.   The first synthesizing unit includes a node connected to the origin node of the orthogonal table, the origin node of the vertical table, and the right side of the origin node among the nodes included in the synthesis target table and the synthesis target enumeration A node group on the uppermost side of the group, a node group on the leftmost side of the node group connected to the lower side of the starting node of the horizontal table and the starting node, and an item name node of enumeration The logical relationship recognition apparatus according to claim 1, wherein the logical relation recognition device is included in data having a tree structure. The orthogonal table acquisition unit acquires a combination of an item name and an item name on the right side of the item name or an item value as a logical relationship in the horizontal direction, and an item name or item below the item name and the item name. A combination of values is acquired as the logical relationship in the vertical direction,
The vertical table acquisition unit acquires a combination of an item name and an item name or item value below the item name as a logical relationship in the vertical direction,
The said horizontal table acquisition part acquires the combination of the item name and the item name on the right side of the said item name, or item value as the said horizontal direction logical relationship, The any one of Claim 1 to 3 characterized by the above-mentioned. The logical relationship recognition device described. The first deletion unit, when a plurality of nodes are adjacent to the left or upper side of one node, deletes an edge representing an adjacency relationship between the one node and the plurality of nodes;
The second deletion unit includes an item name node adjacent to an item value node that is a node of an area representing an item value on the left or upper side of the item name nodes, and an adjacency relationship between the item value nodes. The logical relationship recognition apparatus according to claim 1, wherein an edge representing the information is deleted.   In the second acquisition unit, at least one of the first node groups adjacent to the right side of the first item name node is the item name node, and all the nodes included in the first node group The height of the first item name node is equal to or less than the height of the first item name node, and the vertex of the upper left node of the first node group overlaps the vertex of the first item name node, It is determined that the first item name node includes the first node group, and at least one of the second node groups adjacent to the lower side of the second item name node is the item name node. And the width of all the nodes included in the second node group is equal to or less than the width of the second item name node, and the vertex of the upper left node of the second node group, and If the vertices of the first item name node overlap, the second term Logical relationship recognition apparatus according to claim 5 in which the name node and determines that encompasses the second node group. A logical relationship recognition method executed by a logical relationship recognition device,
Based on a graph that represents information related to the area representing the item name or item value of the form as a node and the adjacent relationship between the nodes as an edge, a node that satisfies a preset condition is selected as the item name. An extraction step of extracting as an item name node that is a node of an area representing
Based on the edge, a starting node that is a node that is a starting point of the table is extracted from the item name node, and a table starting from the starting node is an orthogonal table having item names on both the upper end and the left end, and the upper end A table classification process for classifying the table into one of a vertical table in which the item name exists in the table and a horizontal table in which the item name exists on the left end;
A vertical logical relationship between the item name nodes in the orthogonal table, a horizontal logical relationship between the item name nodes in the orthogonal table, and a node other than the item name node and the item name node in the orthogonal table. An orthogonal table acquisition step of acquiring a vertical logical relationship between the item value nodes and a horizontal logical relationship between the item name node and the item value node in the orthogonal table;
A vertical table acquisition step of acquiring a vertical logical relationship between the item name nodes in the vertical table, and a vertical logical relationship between the item name node and the item value node in the vertical table;
A horizontal table acquisition step of acquiring a horizontal logical relationship between the item name nodes in the horizontal table and a horizontal logical relationship between the item name node and the item value node in the horizontal table;
A first specifying step of specifying a table to be synthesized by excluding a table satisfying a predetermined condition indicating that the table is inconsistent from the orthogonal table, the vertical table, and the horizontal table;
When a plurality of nodes are adjacent to each other in a predetermined direction of one node, a first deletion step of deleting an edge representing an adjacent relationship between the one node and the plurality of nodes;
Among the item name nodes, an item name node that is adjacent to an item value node that is a node of an area that represents an item value in a predetermined direction, and an edge that represents an adjacency relationship between the item value node are deleted. Delete process,
A first acquisition step of acquiring a logical relationship between the item name node and the item value node based on the graph in which an edge is deleted by the first deletion step and the second deletion step. When,
A second acquisition step of acquiring an inclusion relationship between the item name nodes based on the graph in which an edge is deleted by the first deletion step;
A second specifying step of specifying a logical relationship related to enumeration of synthesis targets, excluding a logical relationship related to nodes included in the synthesis target table, among the logical relationships acquired in the first acquisition step;
Of the inclusion relationships acquired by the second acquisition step, the first tree structure based on the inclusion relationship that is not included in either the logical relationship related to the table to be combined and the logical relationship related to the enumeration of the combination target. A first synthesis step for creating data;
A tree structure data is created based on a logical relationship regarding the table to be synthesized, and a second tree structure data is created by synthesizing the tree structure data and the first tree structure data. A synthesis process;
A tree structure data is created based on the logical relationship relating to the synthesis target enumeration, and a third tree structure data is created by synthesizing the tree structure data and the second tree structure data. A synthesis process;
The logical relationship recognition method characterized by including. On the computer,
Based on a graph that represents information related to the area representing the item name or item value of the form as a node and the adjacent relationship between the nodes as an edge, a node that satisfies a preset condition is selected as the item name. An extraction step of extracting as an item name node that is a node of an area representing
Based on the edge, a starting node that is a node that is a starting point of the table is extracted from the item name node, and a table starting from the starting node is an orthogonal table having item names on both the upper end and the left end, and the upper end A table classification step for classifying the table into one of a vertical table in which the item name is present and a horizontal table in which the item name is present on the left end;
A vertical logical relationship between the item name nodes in the orthogonal table, a horizontal logical relationship between the item name nodes in the orthogonal table, and a node other than the item name node and the item name node in the orthogonal table. An orthogonal table acquisition step for acquiring a vertical logical relationship between item value nodes and a horizontal logical relationship between the item name node and the item value node in the orthogonal table;
A vertical table acquisition step of acquiring a vertical logical relationship between the item name nodes in the vertical table and a vertical logical relationship between the item name node and the item value node in the vertical table;
A horizontal table acquisition step for acquiring a horizontal logical relationship between the item name nodes in the horizontal table and a horizontal logical relationship between the item name node and the item value node in the horizontal table;
A first specifying step of specifying a table to be synthesized excluding a table satisfying a predetermined condition indicating that the table is inconsistent from the orthogonal table, the vertical table, and the horizontal table;
A first deletion step of deleting an edge representing an adjacency relationship between the one node and the plurality of nodes when a plurality of nodes are adjacent to each other in a predetermined direction of the one node;
Among the item name nodes, an item name node that is adjacent to an item value node that is a node of an area that represents an item value in a predetermined direction, and an edge that represents an adjacency relationship between the item value node are deleted Delete step,
A first acquisition step of acquiring a logical relationship between the item name node and the item value node based on the graph in which an edge is deleted by the first deletion step and the second deletion step. When,
A second acquisition step of acquiring an inclusion relationship between the item name nodes based on the graph in which an edge is deleted by the first deletion step;
A second specifying step of specifying a logical relationship regarding enumeration of synthesis targets excluding a logical relationship regarding nodes included in the synthesis target table among the logical relationships acquired by the first acquisition step;
Of the inclusion relationships acquired by the second acquisition step, the first tree structure based on the inclusion relationship not included in any of the logical relationship related to the table to be combined and the logical relationship related to the enumeration of the combination target. A first synthesis step for creating data;
A tree structure data is created based on a logical relationship regarding the table to be synthesized, and a second tree structure data is created by synthesizing the tree structure data and the first tree structure data. A synthesis step;
A tree structure data is created based on the logical relationship relating to the synthesis target enumeration, and a third tree structure data is created by synthesizing the tree structure data and the second tree structure data. A synthesis step;
A logical relationship recognition program characterized in that

Images