JP2009110405A - Document data processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To appropriately determine the correctness of description in document data, and to facilitate the management of a dictionary for the determination. <P>SOLUTION: A concept extraction part 13 extracts a concept name associated with a surface expression corresponding to an expression in a first column of input document data from a surface expression dictionary. A written concept extraction part 15 extracts a concept name associated with a surface expression corresponding to an expression in a second column of the input document data from the surface expression dictionary. A determination part 17 determines that an expression to be described in the second column related to the expression in the first column in the document data is missing unless the concept name extracted from the written concept extraction part 15 is not included in a related concept list. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a document data processing apparatus that analyzes a description of document data.

  Conventionally, to support the creation and proofreading of document data, a field-dependent dictionary is used to preferentially present words in the field corresponding to the description in the input document data, and grammar checking is performed on a sentence-by-sentence basis through syntax analysis. There is technology to do.

  Further, as disclosed in, for example, Patent Document 1, there is a technique for presenting a portion lacking in concreteness among descriptions in input document data. In this technique, when a predetermined characteristic part is included in the description in the input document data, a sentence having this characteristic part is included, and 5W1H (when, where, etc.) that should not be omitted is included. There is a technique for determining whether or not there is.

Further, as disclosed in, for example, Patent Document 2, there is a technique that supports automatic construction of a dictionary unique to a field by referring to an ontology dictionary that describes a conceptual relationship between words.
JP 2002-183117 A JP 2001-67356 A

  However, as described above, the technology that preferentially displays the words in the field related to the description in the document data or performs the grammar check cannot determine the semantic deficiency of the description content. Further, as described above, if it is determined whether or not 5W1H is included in a sentence having predetermined characteristics in document data, it is possible to check a sentence having no concreteness, but 5W1H is a conceptual element. Therefore, when the document data is, for example, a product repair report, this is not pointed out when there is no specific word to be described in the repair report.

  Further, in the technique disclosed in Patent Document 2, a hierarchical dictionary is used to refer to an ontology dictionary that describes the conceptual relationship of words and to support field-specific automatic dictionary construction. The reference regarding the relationship between the two was unclear, and the realization as a device was not certain.

  SUMMARY OF THE INVENTION An object of the present invention is to provide an in-document data processing apparatus capable of accurately determining the validity of a description in document data and facilitating management of a dictionary for this determination. .

  That is, the document data processing apparatus according to the present invention includes a surface expression storage means for storing a surface expression dictionary in which a concept name and a surface expression of the concept name are associated with each other, system information of the concept name, and the concept name. A first concept name and a second concept name corresponding to an expression to be included in the document data together with an expression corresponding to the first concept name, not an upper concept name or a lower concept name of the first concept name Hierarchical concept storage means for storing a hierarchical concept dictionary including the link information, and first extraction means for extracting a concept name associated with the surface expression related to the first expression in the input document data from the surface expression dictionary And a second extraction means for extracting a concept name associated with the surface expression related to the second expression described together with the first expression in the input document data from the surface expression dictionary, and system information of the hierarchical concept dictionary Based on the link information, the related concept name extracted by the first extraction means is extracted, and input when the related concept name does not include the concept name extracted by the second extraction means. It is determined that the expression to be described together with the first expression is insufficient in the document data.

  According to the present invention, it is possible to accurately determine the legitimacy of the description in the document data, and to easily manage the dictionary for this determination.

Embodiments of the present invention will be described below with reference to the drawings.
(First embodiment)
First, a first embodiment of the present invention will be described.
FIG. 1 is a block diagram illustrating a configuration example of a document data processing apparatus according to an embodiment of the present invention.
As shown in FIG. 1, a document data processing apparatus according to an embodiment of the present invention includes a control unit 11 that controls the entire apparatus, a storage device 12, a concept extraction unit 13, a concept selection unit 14, and a written concept extraction unit 15. The input / output interface 16 and the determination unit 17 are connected to each other via a bus 18.

  The storage device 12 is configured by hardware such as a hard disk drive or a nonvolatile memory device. The storage device 12 stores a control program to be executed by the control unit 11, the concept extraction unit 13, the concept selection unit 14, and the written concept extraction unit 15, in addition to a determination rule storage unit 21, a surface layer expression dictionary storage unit 22, and a hierarchy A concept dictionary storage unit 23 is included.

Whether or not the determination rule storage unit 21 is insufficiently described in another column, for example, the “report content” column, with respect to the concept requested to be described in a column of the input document data, for example, the “symptom” column. A determination rule for determining is stored.
For example, for the concept A that is determined to be described in the “symptom” column, the determination rule is that the description in the “report content” column is (1) the concept A is described. (2) the concept Either A or a superordinate concept of concept A may be described. (3) It is only necessary that either concept A, a superordinate concept of concept A, or a subordinate concept of concept A is described. 4) It is sufficient that either concept A or concept A linked concept is described. (5) concept A, concept A higher or lower concept, concept A linked concept, or concept A Any one of the upper and lower concepts of the link destination concept may be described, and one of them is set.

FIG. 2 is a diagram showing a configuration example of the surface layer expression dictionary stored in the storage device 12 of the document data processing apparatus 1 according to the first embodiment of the present invention in the form of a table. This surface expression dictionary is an example based on a failure diagnosis report of home appliances.
The surface expression dictionary storage unit 22 stores the surface expression dictionary having the configuration shown in FIG. This surface expression dictionary is used to determine whether or not the input document data contains expressions corresponding to a predetermined concept, and the surface expression associated with the concept name and the target device It is a list of models.
The surface representation is an actual representation in the input document. The device, model, and surface expression in the surface expression dictionary can be described using regular expressions.

The concept name column of the surface layer expression dictionary describes the names of concepts that may be described in the input document. In the target device / model column, a device name and a model name in which the concept described in the “concept name” column may be described.
In the surface expression column, an expression when the concept described in the concept name column is actually described in the input document is described. The column of the surface layer expression at the time of report content determination describes the surface layer expression to be added when it is desired to make a determination by adding another surface layer expression to the description of the surface layer expression column at the time of the determination of “report content” in the input document.

For concepts that should be described as concepts in the conceptual system but should not be extracted as concepts from the “symptoms” and “report contents” columns of the input document, the “surface expression” column and “report content judgment” The “surface expression” column may be a specific description, for example, “none”.
Furthermore, while it is not preferable to extract the concept as a concept when determining the description concept in the “symptom” column of the input document data, the “surface expression” column indicates the concept to be extracted as a superordinate concept / subordinate concept in the “report content” column. May be “none”, and the “surface expression at the time of report content determination” field may be an expression when the concept is actually described.

  In the document data processing apparatus 1 according to the embodiment of the present invention, when an expression in the input document data includes the same expression as the surface expression managed by the surface expression dictionary, the document data processing apparatus 1 associates the expression with the expression in the surface expression dictionary. It is determined that an expression corresponding to the name is included in the input document data.

FIG. 3 is a diagram showing an example of a hierarchical concept managed by the document data processing apparatus according to the first embodiment of the present invention.
The document data processing apparatus according to the first embodiment of the present invention manages the hierarchical concept as shown in FIG. The hierarchical concept shown in FIG. 3 is an independent multi-level concept such as “component” and “symptom”, that is, a category. The concept name of each hierarchy of the hierarchy concept shown in FIG. 3 corresponds to each concept name in the column of “concept name” on the surface expression dictionary shown in FIG.

Specifically, in this hierarchical concept, “cooling component”, “filter”, “hose”, “rotating component”, “rotating component”, “ “Turntable” and “Heater” are associated.
Further, “compressor” and “refrigerant” are associated as the concept names of the third hierarchy subordinate to the concept name “cooling component” of the second hierarchy, and the second subordinate of the concept name “rotating parts” of the second hierarchy. “Motor” and “belt” are associated as concept names of the third hierarchy.
Further, “abnormal symptom during operation”, “abnormality when power is turned off”, and “visible damage” are associated as concept names in the second hierarchy lower than the concept name “symptom” in the first hierarchy. In addition, “bad operation”, “abnormal termination”, and “other abnormalities” are associated as conceptual names of the third hierarchy subordinate to the concept name “abnormal symptom during operation” of the second hierarchy.
In addition, “not cool”, “not warm”, and “do not rotate” are associated as concept names of the fourth hierarchy lower than the concept name “malfunction” of the third hierarchy. Also, “does not move” and “stops halfway” are associated as the concept names of the fourth hierarchy lower than the concept name “abnormal termination” of the third hierarchy. Further, “water leak” and “abnormal sound” are associated as the concept names of the fourth layer lower than the concept name “other abnormality” of the third layer.

  Further, in this hierarchical concept, not only the hierarchical relationship within a category but also link information to another category is managed. Link information across categories does not have a direct relationship as a concept, but is information for associating related concepts in determination of description in a document. In other words, the link information is a combination of a concept name in the hierarchical concept and a system different from the system to which the concept name belongs, that is, a concept name that is neither a superordinate concept nor a subordinate concept.

In the example shown in FIGS. 3 and 4, four pieces of link information are set. A dotted line in FIG. 3 indicates a link with another category. The first link information includes the concept name “cooling component” in the second hierarchy lower than the concept name “component” in the first hierarchy and the fourth hierarchy lower than the concept name “symptom” in the first hierarchy. It is a combination with the concept name “I do not get cold”.
The second link information includes a concept name “hose” in the second hierarchy lower than the concept name “component” in the first hierarchy and a fourth hierarchy lower than the concept name “symptom” in the first hierarchy. It is a combination with the concept name “water leak”.
Further, the third link information includes a second hierarchical concept name “rotating part” lower than the first hierarchical concept name “component” and a fourth hierarchical subordinate to the first hierarchical concept name “symptom”. This is a combination with the concept name “do not turn” in the hierarchy.
Further, the fourth link information includes a second hierarchical concept name “turn table” lower than the first hierarchical concept name “component” and a fourth hierarchical subordinate to the first hierarchical concept name “symptom”. This is a combination with the concept name “do not turn” in the hierarchy.

  If the link information is managed, if the description of the input document data is a description of the component “turntable” with the device name “microwave oven”, the document is based on the fourth link information described above. , Can be considered to be described with respect to a symptom “does not turn” specific to the part “turntable”, and can associate a part and a symptom specific to the part.

  A creation rule exists for this link information. For example, this creation rule is: “Do not create links in the same category”, “Do not create multiple links from one branch to the same branch in another category”, “When deleting an item with a link, The link that has been moved to a higher concept ".

  Further, in the example shown in FIG. 3, apart from the link information, the concept name “water leak” in the fourth hierarchy lower than the concept name “symptom” in the first hierarchy and the concept name “symptom” in the first hierarchy. The concept name “water leakage” in the third hierarchy below “” is managed to be the same item. With this management, these concept names can share dictionary information such as surface expression.

FIG. 4 is a diagram showing a configuration example of the hierarchical concept dictionary stored in the storage device 12 of the document data processing device according to the first embodiment of the present invention in the form of a table.
As shown in FIG. 4, a hierarchical concept dictionary is stored in the hierarchical concept dictionary storage unit 23 of the document data processing apparatus according to the first embodiment of the present invention.

This hierarchical concept dictionary embodies the hierarchical structure shown in FIG. In this hierarchical concept dictionary, row labels (ID), concept names, link information with row labels of other concepts, and hierarchy configuration information are associated and managed.
The hierarchy configuration information is a column for enumerating the concept system to which the concept name associated in the hierarchy concept dictionary belongs from the highest level to the corresponding concept name. This column corresponds to “hierarchy 01”, “hierarchy 02”, “hierarchy 03”, and “hierarchy 04” on the hierarchical concept dictionary shown in FIG.
A different character string is assigned to each line of the label of the hierarchical concept dictionary. The same concept name may appear multiple times in the hierarchical concept dictionary as long as the concept system is different.
In the example shown in FIG. 4, the hierarchical concept dictionary is managed side by side in the order of the superordinate concepts for convenience. However, if the concept system can be extracted in order from the superordinate concepts, it is actually stored in the hierarchical concept dictionary storage unit 23. The order is not limited to this.

The concept extraction unit 13 extracts the concept name corresponding to the expression in the predetermined column of the input document from the surface expression dictionary stored in the surface expression dictionary storage unit 22, so that the description is requested in the input document data. To extract the concept.
The concept selection unit 14 selects a concept name that can be described in the report of the input document from the concept names registered in the surface expression dictionary according to the notation in a predetermined column of the input document.
The description concept extraction unit 15 extracts a concept name corresponding to the expression in a predetermined column of the input document from the report content column check dictionary. The report content column check dictionary is obtained by adding a column of essential information flags to be described later to the surface layer expression dictionary.

The input / output interface 16 exchanges data with the user terminal device 2. The input / output interface 16 can be connected to an external storage device (not shown) via a cable, and inputs / outputs document data to / from the external storage device.
The determination unit 17 follows the concept name extracted by the concept extraction unit 13, the concept name extracted by the description concept extraction unit 15, and the determination rule stored in the determination rule storage unit 21. It is determined whether or not the description of the column referred to by 15 is insufficient.

FIG. 5 is a block diagram showing a configuration example of a user terminal device connected to the document data processing device according to the first embodiment of the present invention.
The user terminal device 2 is a terminal device handled by a user who creates a document to be checked for consistency. As shown in FIG. 5, the user terminal device 2 includes a control unit 31 that controls the processing of the entire device, a storage device 32, a display device 33, an input device 34, and a communication interface 35, which are mutually connected via a bus 36. Connected.

The storage device 32 is a storage device configured by hardware such as a hard disk drive or a nonvolatile memory device. The storage device 32 stores a control program. The storage device 32 also functions as a work memory for various processes performed by the control unit 31.
The display device 33 is a liquid crystal display, for example, and outputs the determination result of the determination unit 17 of the document data processing device 1 to the user. For example, the message “Symptom A is required to be described, but there is no description related to the report contents.” “There is no concept lacking description for the symptom”.

The input device 34 is, for example, a keyboard or a mouse, and accepts an operation related to document creation for performing intra-document consistency check. The communication interface 35 is connected to the document data processing apparatus 1 via a cable, and exchanges an input document created in the user terminal apparatus 2 and a processing result of the document data processing apparatus 1.
The communication interface 35 can be connected to an external storage device via a cable, and input documents can be input / output to / from the external storage device.

  Next, the operation of the document data processing apparatus having the configuration shown in FIG. 1 will be described. FIG. 6 is a flowchart showing an example of the processing operation of the document data processing apparatus 1 according to the first embodiment of the present invention.

  First, input document data is created by an operation on the input device 34 of the user terminal device 2, and this document data is input from the communication interface 35 to the input / output interface 16 of the document data processing device 1, or an external storage device. Assume that the input document data that has already been stored is input to the input / output interface 16 of the document data processing apparatus 1. The control unit 11 of the document data processing apparatus 1 stores the document data input to the input / output interface 16 in the storage device 12 (step S1).

FIG. 7 is a diagram showing a configuration example of input document data input to the document data processing apparatus according to the first embodiment of the present invention.
The input document data is a refrigerator failure diagnosis report in which the target device name, model name, symptom, and report content are described in a table format.
In the example shown in FIG. 7, the description in the device name column is “refrigerator”, the description in the model name column is “RFxx-yyy”, and the description in the symptom column is “the cabinet does not cool”. Yes, and the description in the report content column is "Compressor defective. Replaceable."
Note that the input document data is not limited to the table format shown in FIG. 7, but may be a bullet format or a normal text format.

  The concept selection unit 14 reads the surface layer expression dictionary stored in the surface layer expression dictionary storage unit 22 (step S2), and executes determination on each row of the surface layer expression in conjunction with the concept extraction unit 13 (step S3).

The concept selection unit 14 extracts one line of the surface expression dictionary, and the expression in the “device name” and “model name” fields of the input document is expressed in the “target device / model” field in the selected line of the surface expression dictionary. It is determined whether or not they match as a column (step S4).
When the expression in the “target device / model” column of the selected row of the surface expression dictionary does not match the expression in the “device name” and “model name” column of the input document (NO in step S4), the concept selection unit 14 Move on to the next line of the surface expression dictionary.
On the other hand, when the concept selection unit 14 determines that the expression in the “target device / model” column in the selected row of the surface expression dictionary matches the expression in the “device name” and “model name” columns of the input document (step S4). YES), the concept extraction unit 13 determines whether the expression in the “surface expression” column of the selected line of the surface expression dictionary matches the expression in the “symptom” field of the input document shown in FIG. 7 as a character string. Determine (step S5).

The concept extraction unit 13 sets an essential information flag when the contents of the “surface expression” field in the selected line of the surface expression dictionary match the description in the “symptom” field of the input document as a character string (YES in step S5). Newly added and the value of the essential information flag in the selected row of the surface expression dictionary is set to “1” (step S6).
In addition, if the contents of the “surface expression” column of the selected line of the surface expression dictionary do not match the description in the “symptom” field of the input document as a character string (NO in step S5), the concept extraction unit 13 is required information A flag is newly added, and the value of the essential information flag of the selected row is set to “0”.

The concept extraction unit 13 outputs information obtained by adding the essential information flag to the selected line of the surface expression dictionary to the storage device 12 as an entry of a newly generated report content column check dictionary (step S7).
The concept extraction unit 13 collects the report content column check dictionary entries output to the storage device 12 after the processing for all the rows in the surface expression dictionary is completed, and stores it as a report content column check dictionary (steps S8 → S9). ).

FIG. 8 is a diagram showing an example of a report content column check dictionary that is an intermediate output of the document data processing apparatus according to the first embodiment of the present invention.
In the report content column check dictionary shown in FIG. 8, the “target device / model” column in each row of the surface expression dictionary shown in FIG. 2 includes “device name” and “model name” of the input document shown in FIG. Lines that do not match the column representation are not included.
In addition, the report content column check dictionary shown in FIG. 8 has a column of essential information flags added to the surface layer expression dictionary shown in FIG. “1” is entered in the column of the essential information flag in the line of the matching surface expression “not cool”.

The description concept extraction unit 15 reads the report content column check dictionary stored in the storage device 12, and performs the determination process for each line (step S10).
The description concept extraction unit 15 extracts one line of the report content column check dictionary, and the description of the “surface expression” column of the selected row of the report content column check dictionary is a character in the description of the “report content” column of the input document. It is determined whether or not they match as a column (step S11).
However, if the “surface expression at the time of report content determination” field in the selected row of the report content field check dictionary is not blank, the description concept extraction unit 15 sets the contents as described in the “surface expression” field of the dictionary. At the same time, it is determined whether or not to match.

The description concept extracting unit 15 includes a description in the “report content” column of the input document, a description in the “surface expression” column of the selected row of the report content column check dictionary, and a blank in the “surface expression at the time of report content determination” column. If it matches any of the descriptions that are not (YES in step S11), a new report content column presence / absence flag column is set, and the value of the report content column presence / absence flag in the selected row of the report content column check dictionary is set to “1”. (Step S12).
On the other hand, the description concept extraction unit 15 includes the description in the “report content” column of the input document, the description in the “surface expression” column of the selected row of the report content column check dictionary, and the “surface expression at the time of report content determination” column. If it does not match any of the non-blank descriptions (NO in step S11), the column of the report content column presence / absence flag is newly set, and the value of the report content column presence / absence flag of the selected row is set to “0”.

FIG. 9 is a diagram showing an example of a report content column check result that is an intermediate output of the document data processing apparatus according to the first embodiment of the present invention.
The report content column check result shown in FIG. 9 is the row where the essential information flag is “1” among the rows of the report content column check dictionary shown in FIG. 8, that is, the concept name is “not cool”. Or the report content column description presence flag is “1”, that is, the concept name is “compressor”, and among these lines, the concept name, required information flag, report content column description presence flag It is what left the column.

After setting the value of the report content column presence / absence flag, the description concept extraction unit 15 determines whether both the value of the mandatory information flag and the value of the report content column presence / absence flag in the selected row of the report content column check dictionary are “0”. It is checked whether or not (step S13).
If either of the flags is “1” (NO in step S13), the written concept extraction unit 15 uses the concept name and the value of the essential information flag as an entry in the newly generated report content check result list. Then, the value of the report content column presence / absence flag is output to the storage device 12 (step S14).

The description concept extraction unit 15 proceeds to the process of the next line in the report content column check dictionary after the process of step S14 or when “YES” is determined in the process of step S13 (step S15).
The description concept extraction unit 15 collects the report content column check result list entries stored in the storage device 12 after the processing for all the lines in the report content column check dictionary and stores it as a report content column check result (step S16). .

  The determination unit 17 checks the hierarchical concept dictionary stored in the hierarchical concept dictionary storage unit 23 using the report content column check result output from the description concept extraction unit 15 and the determination rule stored in the determination rule storage unit 21. (Step S17).

Here, details of the process of step S17 which is the process of the determination unit 17 will be described. FIG. 10 is a flowchart showing an example of the processing operation of the determination unit 17 of the document data processing apparatus 1 according to the first embodiment of the present invention.
The determination unit 17 reads the report content column check result list stored in the storage device 12 (step S21). Next, the determination unit 17 reads the hierarchical concept dictionary stored in the hierarchical concept dictionary storage unit 23 (step S22).

The determination unit 17 extracts the report content column check result list line by line, and checks the value of the report content column description presence flag in the selected row of the report content column check result list (step S24).
If the value of the report content column description presence flag is “1” (YES in step S24), the determination unit 17 searches the hierarchical concept dictionary for a row having the same concept name as the concept name of the selected row in the report column check result list. A column of “description” is newly added to the row of the same concept name in the hierarchical concept dictionary, and this value is set to “present” (step S25).

  The determination unit 17 checks the value of the essential information flag in the selected row of the report content column check result list after the process of step S25 or when determining “NO” in the process of step S24 (step S26). If the value of the mandatory information flag is “1” in the selected row of the report content column check result list (YES in step S26), the determination unit 17 newly generates the concept name of the selected row in the hierarchical concept dictionary. It is output to the storage device 12 as an entry in the target concept list (step S27).

  The determination part 17 moves to the process of the next line of the said report content column check result list after the process of step S27, or when it determines with "NO" by the process of step S26 (step S28).

  The determination unit 17 collects the entries of the check target concept list output to the storage device 12 after the processing of all the rows in the report content column check result list is completed, and stores the check target concept list in the storage device 12 as a check target concept list (step S29). ).

FIG. 11 is a diagram showing an example of a check target concept list that is an intermediate output of the document data processing apparatus according to the first embodiment of the present invention.
In the example illustrated in FIG. 11, in the concept list to be checked, the concept name “not cooled” of the column whose essential information flag value is “1” in the report content column check result list is managed. In the example shown in FIG. 11, the check target concept list has only one row, but the list is not limited to one row.

  FIG. 12 is a diagram showing an example of a hierarchical concept dictionary to which description presence / absence information, which is an intermediate output of the document data processing apparatus according to the first embodiment of the present invention, is given. In the example shown in FIG. 12, a column “description” is added to the hierarchical concept dictionary shown in FIG. 4, and the selected row of the hierarchical concept dictionary at the time when “YES” is determined in the process of step S24 in this column. "Yes" is set in the column of.

The determination unit 17 reads the check target concept list stored in the storage device 12, and selects the list line by line (step S30).
The determining unit 17 searches the hierarchical concept dictionary shown in FIG. 12 for the concept name of the selected row in the check target concept list and follows the determination rule stored in the determination rule storage unit 21 to select the concept of the selected row in the check target concept list. Name related concepts are extracted from the hierarchical concept dictionary shown in FIG. 12 and listed (step S31).

  Here, for example, (5) “Concept A, or a concept that is a superordinate / subordinate concept of concept A, a concept linked to another category of concept A, or a superordinate / subordinate concept of a linked concept, for example, ”Is sufficient if it is described” is set.

FIG. 13 is a diagram showing an example of a related concept list which is an intermediate output of the document data processing apparatus according to the first embodiment of the present invention.
The related concept list shown in FIG. 13 is a related concept list of the concept name “not cool” in the selected row of the check target concept list.

  First, the determination unit 17 extracts, from the hierarchical concept dictionary illustrated in FIG. 12, a row including the values of the columns of the concept name “not cool”, ID “s4”, and “description” of the selected row of the check target concept list. Add to an empty list. Next, the determination unit 17 includes the concept name “symptom”, “abnormal symptom during operation”, “operation failure”, the ID of the concept name, and “description”, which are the superordinate concepts of the concept name “not cool” in the selected row. ”Is extracted from the hierarchical concept dictionary shown in FIG. 12 and added to the list.

The superordinate concept is a concept having a lower number of layers than the concept name “not cooled” in the selected row of the check target concept list, and the entire hierarchy configuration name matches the name of the corresponding hierarchy of the concept name “not cooled”. Since the concept name “does not cool” has no subordinate concept, a row related to the subordinate concept of the concept name “does not cool” is not added to the related concept list.
The subordinate concept has a higher number of layers than the concept name “not cool” in the selected row of the check target concept list, and the concept name “not cool” matches the name of the corresponding hierarchy. Indicates. When the concept name “not cool” appears in a plurality of locations on the hierarchical concept dictionary shown in FIG. 12, the determination unit 17 performs the same process on all the rows in the hierarchical concept dictionary.

  In addition, the determination unit 17 sets “p2_cooling component” as the link destination in the concept name “not cool” with the ID “s4” on the hierarchical concept dictionary illustrated in FIG. A row including values of the columns of ID “p2”, concept name “cooling component” and “description” on the hierarchical concept dictionary shown is extracted from the hierarchical concept dictionary shown in FIG. 12 and added to the list.

  Furthermore, the determination unit 17 includes a column of a higher concept name “component”, a lower concept name “compressor”, “refrigerant”, an ID of the concept name, and “description” of the concept name “cooling component” whose ID is “p2”. Are extracted from the hierarchical concept dictionary shown in FIG. 12 and added to the related concept list.

The determination unit 17 selects the created related concept list line by line (step S32). The determination unit 17 checks the presence / absence of “present” in the “description” column in the selected row of the related concept list (step S33).
When the “description” column of the selected row in the related concept list is “present” (YES in step S33), the determination unit 17 gives information indicating that there is a description in the selected row of the check target concept list being processed. (Step S34).

The determination unit 17 proceeds to the process of the next row in the related concept list after the process of step S34 or when determining “NO” in the process of step S33 (step S35).
The determination unit 17 checks whether or not information indicating that there is a description is given to the selected row of the check target concept list after the processing for all the rows of the related concept list is completed (step S36).

  When the information indicating that there is a description in the selected row of the check target concept list is not given (NO in step S36), the determination unit 17 displays the expression corresponding to the concept name of the selected row of the check target concept list in the input document. The information indicating this is output to the storage device 12 as an alarm list entry (step S37).

  The determination unit 17 proceeds to the process of the next row in the check target concept list after the process of step S37 or when “YES” is determined in the process of step S36 (step S38).

The determination unit 17 collects the alarm list entries stored in the storage device 12 after completing the processing for all the rows of the check target concept list (step S38), and sends a message to the user via the input / output interface 16 to the user terminal device. 2 (step S38).
When the communication interface 35 of the user terminal device 2 inputs a message, the control unit 31 causes the display device 33 to display the message as an in-document consistency check result (step S18).
Further, when there is no alarm list entry, the determination unit 17 may output a message “There is no concept of insufficient description”.

  As described above, the document data processing apparatus according to the first embodiment of the present invention has link information for the concept determined to be described in, for example, the “symptom” column in the input document data. By checking whether the hierarchical concept dictionary and concepts recognized as related concepts from the hierarchical concept dictionary are described, not only the concepts that are required to be described, but also their higher-level concepts, lower-level concepts, and other categories are linked by links. It can be determined that the necessary information has also been described. Therefore, even if necessary information is described without using the completely same expression as the “symptom” column described above, it is not erroneously determined that the description is insufficient.

  In addition, since it can be determined that the information associated with the hierarchical concept dictionary is not described at all, it can be determined that the description is insufficient, so that it is possible to support correction of deficiencies in document data creation, and the burden of document data creation by the user is reduced. Can be reduced.

  Further, since the hierarchical concept dictionary and the surface expression dictionary are managed separately, the surface layer expression of a concept that appears multiple times in the hierarchical concept dictionary such as the concept name “water leak” shown in FIG. The correction of the expression is only one line on the surface expression dictionary, and the management load is reduced.

In the present embodiment, an example is shown in which there are two concept systems (categories) managed by the hierarchical concept management unit. However, the number of categories is not limited to two, and may be three or more.
In addition, the present invention is not limited to a report related to repair of a home appliance failure. For example, for a hospital test report, the hierarchy concept is composed of categories such as “body part”, “disease”, and “examination / treatment”, and the part related to the disease described in the test purpose column It is also possible to configure so that it is determined that necessary information is described by describing the inspection result.

(Second Embodiment)
Next, a second embodiment of the present invention will be described. Note that the description of the same part of the configuration of the document data processing apparatus according to the present embodiment as shown in FIG. 1 is omitted.
FIG. 14 is a block diagram showing a configuration example of a document data processing apparatus according to the second embodiment of the present invention.
As shown in FIG. 14, the document data processing apparatus according to the second embodiment of the present invention has a hierarchical concept for storing the configuration rules of the hierarchical conceptual system in addition to the configuration of the first embodiment. A rule storage unit 41 is included. In addition, a hierarchical concept analysis unit 42 for analyzing the hierarchical conceptual system according to the hierarchical conceptual system configuration rules is provided.

  The hierarchical concept system configuration rules stored in the hierarchical concept rule storage unit 41 include, for example, “no link is given in the same category” or “a concept belonging to the same branch in another category from a concept belonging to the same branch in the hierarchical concept” No link is given to ". For example, in the hierarchical concept dictionary shown in FIG. 4, when the configuration rule “do not give a link in the same category” is set with respect to the link assignment on the hierarchical concept dictionary, between the lines where the highest concept is “part” Then, even if the user terminal device 2 is operated, the link information cannot be given.

Similarly, with respect to link assignment in the hierarchical concept dictionary, a configuration rule is set that “a link is not given from a concept belonging to the same branch in the hierarchical concept to a concept belonging to the same branch in another category”, as shown in FIG. When link information is set between a concept name “cooling component” whose ID is “p2” and a concept name “not cool” whose ID is “s4” on the hierarchical concept dictionary, the ID is “p2”. A subordinate concept “compressor” of a certain concept name “cooling component”, and a superordinate concept name of a concept name “not cool” with ID “s4”, that is, a concept name “abnormal symptom during operation” with ID “s2”. No link can be set between
Here, the “concept belonging to the same branch on the hierarchical concept” indicates a group composed of a plurality of concepts that are in a relationship between a superordinate concept and a subordinate concept.

  The hierarchical concept analysis unit 42 checks the hierarchical concept dictionary in accordance with the configuration rules of the hierarchical concept system stored in the hierarchical concept rule storage unit 41, and when a portion that violates the configuration rules is found, the display device 33 of the user terminal device 2. Presents an alarm for the user.

  The hierarchical concept analysis unit 42, for example, between the concept name “cooling component” whose ID is “p2” and the concept name “not cool” whose ID is “s4” and the concept name compressor whose ID is “p3” ”And“ abnormal symptom at the time of operation of the concept name whose ID is “s2” ”, the link information is given to“ the same branch of the symptom category from two branches on the component category ”. "Please delete one of them." And output the information to be presented to the user terminal device 2 via the input / output interface 16 to present to the user the relevant part of the hierarchical concept. To do.

Further, the hierarchical concept analysis unit 42 has a function of analyzing link candidates to be further set from link information that has already been set and presenting them to the user.
For example, when the link information between the concept name “hose” with ID “p6” and the concept name “water leak” with ID “s11” is given on the hierarchical concept dictionary in FIG. The concept name “water leak” of ID “s11”, that is, the concept name “water leak” of ID “s14” is extracted from the hierarchical concept dictionary.

  The hierarchical concept analysis unit 42 does not give a link to the above-described concept name “hose” of the ID “p6” or the higher-order concept / lower-order concept of the concept name. In order to present a combination of the concept name “water leak” with ID “s14” and the concept name “hose” with ID “p6” to the user as a new link grant candidate. The information for presentation is output to the user terminal device 2 via the input / output interface 16.

Further, the hierarchical concept analysis unit 42 analyzes the concept name and its surface layer expression, and presents a combination of concepts including the same word to the user as a new link grant candidate.
For example, the hierarchical concept analysis unit 42 displays the surface name “not rotate” of the concept name “does not rotate” on the surface expression dictionary shown in FIG. If the combination does not violate the composition rule of the hierarchical concept system managed by the hierarchical concept rule storage unit 41, input and output information for presentation is presented to the user as a new link candidate. The data is output to the user terminal device 2 via the interface 16. In addition, the process which determines the word contained in a concept name or surface expression can be implemented by the morphological analysis which is a well-known technique.

  As described above, in the document data processing apparatus according to the second embodiment of the present invention, the hierarchical concept analysis unit 42 checks the created hierarchical concept dictionary and points out an error. It is reduced. Further, since the hierarchical concept analysis unit 42 presents new link candidates to the user, the load of creating the hierarchical concept dictionary is reduced.

(Third embodiment)
Next, a third embodiment of the present invention will be described. Note that the description of the same part of the configuration of the document data processing apparatus according to the present embodiment as shown in FIG. 1 is omitted.
FIG. 15 is a block diagram showing a configuration example of a document data processing apparatus according to the third embodiment of the present invention.
As shown in FIG. 15, the document data processing apparatus according to the third embodiment of the present invention, in addition to the configuration of the first embodiment, performs syntax analysis related to the concept name associated with the surface representation. A determination item extraction unit 53 that extracts a determination item from the output of the unit 52 and the concept extraction unit 13. Further, the document data processing apparatus according to the third embodiment does not include the concept selection unit 14 and the description concept extraction unit 15 as compared with the configuration of the first embodiment.

In the third embodiment, the validity of the description of the RCM (Risk Control Matrix) is checked.
Checking the description of the RCM is to confirm that the business process is executed correctly, who executes what at each business step, who is responsible for the business, what documents are relevant, This is to check whether there is any deficiency in the RCM that sorts out the risks that may occur in the business step.

When the description content of the RCM is deficient, the document data processing apparatus 1 performs processing such as displaying an alarm indicating the deficiency in the description.
In the present embodiment, an example described as a business process for applying and approving leave in a company is used as the RCM.

FIG. 16 is a diagram showing an example of a vacation handling standard managed by the document data processing apparatus according to the third embodiment of the present invention.
The storage device 12 of the document data processing apparatus 1 stores vacation handling standard information shown in FIG. This vacation handling standard is a kind of regulations existing in a company, and rules concerning leave application and approval are described in the company.

FIG. 17 is a diagram showing a determination rule for the relationship between the applicant and the approver, which is derived from the vacation handling standard managed by the document data processing apparatus according to the third embodiment of the present invention.
In the present embodiment, information on the relationship between the applicant and the approver shown in FIG. 17 is stored in the determination rule storage unit 21 as the determination rule.

  In Company A, leave handling standards as shown in FIG. 16 are defined as regulations. The description contents of the vacation handling standard are arranged from the viewpoint of the vacation applicant and the approver as shown in FIG. When vacation application processing is performed in company A, the conditions shown in FIG. 17 must be satisfied.

FIG. 18 is a diagram showing an example of a hierarchical concept managed by the document data processing apparatus according to the third embodiment of the present invention.
The document data processing apparatus according to the third embodiment of the present invention manages the hierarchical concept as shown in FIG. The hierarchical concept shown in FIG. 18 is a hierarchical representation of three conceptual systems in the company: “Regulations”, “Organization / Position”, and “In-house system”.
In addition, as shown in FIG. 18, the concept name “in charge” of the concept name “organization structure” of the second hierarchy is the same item, not the superordinate concept or the subordinate concept. Link information with the concept name “in charge” subordinate to the concept name “position” is stored in the storage device 12 and managed.

FIG. 19 is a diagram showing a configuration example of the hierarchical concept dictionary stored in the storage device 12 of the document data processing device according to the third embodiment of the present invention in a table format.
The hierarchical concept dictionary storage unit 23 of the document data processing apparatus according to the third embodiment of the present invention stores the hierarchical concept dictionary having the configuration shown in FIG. This hierarchical concept dictionary is an example in which the relationship of “rules”, “organization / post”, and “in-house system” in the hierarchical concept shown in FIG. 18 is expressed as a hierarchical dictionary.

FIG. 20 is a diagram showing a configuration example of the surface layer expression dictionary stored in the storage device 12 of the document data processing apparatus 1 according to the third embodiment of the present invention in a table format.
The surface expression dictionary storage unit 22 stores the surface expression dictionary having the configuration shown in FIG. This surface expression dictionary is managed by associating concept names, concept classifications, and surface expression in “rules”, “organization / post”, and “in-house system”.

Next, the processing operation of the document data processing apparatus 1 according to the third embodiment of the present invention will be described. FIG. 21 is a flowchart showing an example of the processing operation of the document data processing apparatus according to the third embodiment of the present invention.
FIG. 22 is a diagram showing an example of the RCM input by the document data processing apparatus according to the third embodiment of the present invention. Here, the RCM describes a leave application and approval method.

  In the laboratory B in company A, an RCM having the configuration shown in FIG. 22 is created by operating the input device 34 of the user terminal device 2 for risk management of its own department. This RCM is input from the communication interface 35 to the input / output interface 16 of the document data processing apparatus 1. The control unit 11 of the document data processing apparatus 1 stores the document data input by the input / output interface 16 in the storage device 12 (step S41).

  The concept extraction unit 13 reads the surface layer expression dictionary stored in the surface layer expression dictionary storage unit 22 (step S42). Moreover, the determination item extraction part 53 and the determination part 17 each read the determination rule memorize | stored in the determination rule memory | storage part 21 (step S43). The syntax analysis unit 52 parses the sentence in the business content sequence of the RCM, which is input document data, in sentence units (step S44).

  FIG. 23 is an example of a result of a syntactic analysis of the third sentence in the sentence described in the column of “work contents” in the row related to the sub step number “Step 1-01” of the RCM shown in FIG. . In FIG. 23, the third sentence in the sentence described in the column of “work contents” in the row related to the sub step number “Step 1-01” is divided into clauses.

In FIG. 23, combinations of clauses connected by a solid line are determined to have a dependency relationship. The concept extraction unit 13 analyzes the syntax analysis result by using the surface expression dictionary read in the process of step S42, so that, for example, the “business” of the row related to the substep number “Step1-01” of the RCM shown in FIG. From the parsing result of the third sentence in the sentence described in the “Content” column, the “Research Director”, “Lab Director”, “Lab Director” and “SM” that match the surface expression of the surface expression dictionary are assigned to the positions related to the work. Names and expressions corresponding to the names are extracted, and corresponding surface names are given to these surface layer expressions in the surface layer expression dictionary (step S45).
In FIG. 24, a concept name is given to the parsing result of the third sentence in the sentence described in the column of “work contents” in the row related to the sub step number “Step1-01” of the RCM shown in FIG. It is an example of a result.
The determination item extraction unit 53 extracts the determination item from the output of the concept extraction unit 13 according to the determination rule read in the process of step S43 (step S46). In the determination rule shown in FIG. 17, the relationship between “approver” and “applicant” is determined by the two registered rules. In accordance with this determination rule, the determination item extraction unit 53 acquires the concept name associated with the surface expression on which the “approver” or “applicant” is dependent on the output of the concept extraction unit 13.

  From the concept extraction result shown in FIG. 24, that is, from the concept extraction result of the third sentence in the sentence described in the column of “work contents” in the row related to the sub step number “Step1-01” of RCM shown in FIG. , Surface expression “Research Director” and concept name “Principal” as first applicant, surface expression “Lab Director” and concept name “Lab Director” as first approver, surface expression “Lab” as second applicant "Super chief" and concept name "Laboratory head", and surface expression "SM" and concept name "Research manager" are extracted as the second approver.

  The determination item extraction unit 53 further determines the combination of the extracted applicant and approver. For example, the approver combined with the first applicant is the first one that can be reached with the shortest distance between phrases in the concept extraction result shown in FIG. 24, that is, the number of solid lines connecting the phrases is small. This is done by determining the approver.

  By this combination determination, two sets of determination items shown in the first row and the second row in FIG. 25 are extracted from the concept extraction result shown in FIG. Further, by performing similar syntax analysis, concept extraction, and judgment item extraction processing on the column of business contents in the row of the RCM substep number “Step 1-02” shown in FIG. 22, the third row of FIG. The judgment items shown in are extracted.

The determination unit 17 reads the hierarchical concept dictionary shown in FIG. 19 (step S47). Further, the determination unit 17 extracts the determination item extraction result list shown in FIG. 25 line by line, and checks the relationship between the applicant concept name and the approver concept name in the selected row of the determination item extraction result list (step S48). ).
In the processing of the first row of the determination matter extraction result list in FIG. 25, the determination unit 17 includes the concept names “chief” and “lab manager” related to the applicant and the approver among the concept names in the hierarchical concept dictionary. Focus on the line.

In the determination unit 17, the concept name “chief” of the ID “s25” on the hierarchical concept dictionary shown in FIG. 19 is a hierarchy definition regarding the concept name “position” of the higher-level ID “s15”, and the concept name “chief” Since the superordinate concept name one above is “in charge” with ID “s24”, it is determined that the chief is “in charge”.
In addition, the determination unit 17 pays attention to the concept name “in charge” of the ID “s14”, which is the same item name associated with the above-described link information in order to know who the administrator of the organization to which the chief belongs. This is a hierarchical definition related to the concept name “organization structure” of the higher-level ID “s2”, and the higher-level concept name immediately above the concept name “in charge” is the concept name “laboratory” of the ID “s13”. Determine that there is.

The determination unit 17 follows the determination rule stored in the determination rule storage unit 21 read in the process of step S43, and the concept name “chief” related to the applicant and the approver described in the analysis location in the input RCM. Then, it is determined whether or not the expression relationship in the RCM corresponding to the “laboratory manager” matches the determination rule (step S49).
Specifically, in step S49, the determination unit 17 uses the concept name “chief” of the ID “s25” on the hierarchical concept dictionary shown in FIG. 19 as the subordinate concept of the concept name “laboratory” of the ID “s13”. The expression in the RCM corresponding to the concept name “chief” is the applicant “Research Director”, and the expression in the RCM corresponding to the concept name “Laboratory” is the applicant “Lab Director” To do.

  Then, the determination unit 17 determines that the approver “lab director”, who is the expression of the analysis location in the RCM, is an administrator of the organization to which the applicant “research supervisor” belongs and matches the determination rule. Therefore, the determination unit 17 determines that the descriptions of the first applicant “Research Director” and the applicant's approver “Lab Director”, which are expressions of the analysis location in the RCM, are correct (YES in Step S49). In this case, the process ends without outputting an alarm, and the process proceeds to the next line in the determination item extraction result list (step S51).

  The determination unit 17 includes the second applicant “Lab Director” and the approver, which is another combination of the expression of the analysis part in the RCM described in the second line of the determination item extraction result list shown in FIG. Similarly, the relationship of “SM” can be determined to be a correct description by using the surface expression dictionary and the hierarchical concept dictionary.

  On the other hand, in the column of the business content in the row of the RCM substep number “Step1-02” shown in FIG. 22, which is another business content described in the third row of the judgment item extraction result list shown in FIG. 25. Describes the combination of the applicant “SM” and the approver “lab director”. The determination unit 17 replaces these descriptions with the applicant “Research Department Manager” and the approver “Research Department Manager” based on the contents of the determination item extraction result list.

  The determination unit 17 refers to the concept name “Research Manager” of the ID “s12”, which is the concept name obtained by replacing the surface expression as described above, among the concept names in the hierarchical concept dictionary shown in FIG. Is the “laboratory head” of ID “s9”, and the ID “s13” concept name “laboratory head” read as described above is determined to be a concept name one level below.

  Therefore, since the applicant “SM”, which is the description content of the RCM, is an administrator of the organization to which the approver “lab manager” belongs, it is determined that it does not match the determination rule. Therefore, the determination unit 17 sends an instruction signal for displaying an incomplete description alarm in the column of business contents in the row of the substep number “Step 1-02” of the RCM shown in FIG. To the user terminal device 2.

FIG. 25 is a diagram showing an example in which an alarm is added to the RCM input by the document data processing apparatus according to the third embodiment of the present invention.
When the communication interface 35 of the user terminal device 2 inputs this instruction signal, the control unit 31 displays an incompletely described alarm window in the business content column in the line of the sub step number “Step 1-02” of the RCM that is the input document data. Is displayed on the display device 33 (step S50).

  As described above, in the document data processing apparatus according to the third embodiment of the present invention, when the input document data is the RCM, the description contents of the RCM are incomplete by utilizing the surface expression dictionary and the hierarchical concept dictionary. The presence or absence can be checked.

  The present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. Various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be omitted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

1 is a block diagram showing a configuration example of a document data processing apparatus according to a first embodiment of the present invention. The figure which shows the structural example of the surface layer expression dictionary memorize | stored in the memory | storage device of the document data processing apparatus according to the 1st Embodiment of this invention in a table format. The figure which shows an example of the hierarchy concept which the document data processing apparatus according to the 1st Embodiment of this invention manages. The figure which shows an example of the hierarchy concept dictionary memorize | stored in the memory | storage device of the document data processing apparatus according to the 1st Embodiment of this invention. The block diagram which shows the structural example of the user terminal device according to the 1st Embodiment of this invention. 6 is a flowchart illustrating an example of a processing operation of the document data processing apparatus according to the first embodiment of the present invention. FIG. 5 is a diagram showing a configuration example of input document data input to the document data processing apparatus according to the first embodiment of the present invention. The figure which shows an example of the dictionary for the report content column check which is the middle output of the document data processing apparatus according to the 1st Embodiment of this invention. The figure which shows an example of the report content column check result which is the middle output of the document data processing apparatus according to the 1st Embodiment of this invention. 6 is a flowchart illustrating an example of a processing operation of a determination unit of the document data processing apparatus according to the first embodiment of the present invention. The figure which shows an example of the check object concept list | wrist which is the middle output of the document data processing apparatus according to the 1st Embodiment of this invention. The figure which shows an example of the hierarchy concept dictionary which provided the description presence information which is the middle output of the document data processing apparatus according to embodiment of this invention. The figure which shows an example of the related concept list | wrist which is the middle output of the document data processing apparatus according to the 1st Embodiment of this invention. The block diagram which shows the structural example of the document data processing apparatus according to the 2nd Embodiment of this invention. The block diagram which shows the structural example of the document data processing apparatus according to the 3rd Embodiment of this invention. The figure which shows an example of the vacation handling standard which the document data processing apparatus according to the 3rd Embodiment of this invention manages. The figure which showed the determination rule of the relationship between an applicant and an approver derived | led-out from the vacation handling reference | standard managed by the document data processing apparatus according to the 3rd Embodiment of this invention. FIG. 10 is a diagram showing an example of an RCM input by a document data processing apparatus according to the third embodiment of the present invention. The figure which shows the structural example of the hierarchy concept dictionary memorize | stored in the memory | storage device of the document data processing apparatus according to the 3rd Embodiment of this invention with a table format. The figure which shows the structural example of the surface layer expression dictionary memorize | stored in the memory | storage device of the document data processing apparatus according to the 3rd Embodiment of this invention in a table format. 10 is a flowchart showing an example of processing operation of the document data processing apparatus according to the third embodiment of the present invention. FIG. 10 is a diagram showing an example of an RCM input by a document data processing apparatus according to the third embodiment of the present invention. The figure which shows an example of the halfway output of the syntax analysis part which is the halfway output of the document data processing apparatus according to the 3rd Embodiment of this invention. The figure which shows an example of the output of the syntax analysis part which is the middle output of the document data processing apparatus according to the 3rd Embodiment of this invention. The figure which shows an example of the determination matter extraction result which is the middle output of the document data processing apparatus according to the 3rd Embodiment of this invention. The figure which shows the example which attached | subjected the alarm to RCM which the document data processing apparatus according to the 3rd Embodiment of this invention input.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Document data processing apparatus, 2 ... User terminal device, 11, 31 ... Control part, 12, 32 ... Memory | storage device, 13 ... Concept extraction part, 14 ... Concept selection part, 15 ... Description concept extraction part, 16 ... Input / output Interface: 17: Determination unit, 18, 36: Bus, 21: Determination rule storage unit, 22: Surface expression dictionary storage unit, 23: Hierarchical concept dictionary storage unit, 33: Display device, 34: Input device, 35: Communication interface 52 ... Syntax analysis unit, 53 ... Determination item extraction unit.

Claims (4)

An input means for inputting document data;
A surface expression storage unit for storing a surface expression dictionary in which a concept name and a surface expression of the concept name are associated with each other;
In the document data, the system information of the concept name and the expression corresponding to the first concept name, not the first concept name and the upper concept name or the lower concept name of the first concept name. Hierarchical concept storage means for storing a hierarchical concept dictionary including link information with a second concept name corresponding to an expression to be included;
First extraction means for extracting from the surface expression dictionary a concept name associated with the surface expression related to the first expression in the document data input by the input means;
Second extraction means for extracting a concept name associated with a surface expression related to a second expression described together with the first expression in the input document data from the surface expression dictionary;
Third extraction means for extracting a related concept name of the concept name extracted by the first extraction means based on system information and link information of the hierarchical concept dictionary;
When the related concept name does not include the concept name extracted by the second extraction means, it is determined that the input document data lacks an expression to be described together with the first expression A document data processing apparatus comprising: a determination unit. A configuration condition defining means for defining a configuration condition of the link information;
The document data processing apparatus according to claim 1, further comprising: a link information determination unit that determines the validity of the configuration of link information of the hierarchical concept dictionary based on the configuration condition. A configuration condition defining means for defining a configuration condition of the link information;
Link information determination means for determining whether or not the concept of the hierarchical concept dictionary stored in the hierarchical concept storage means is a new concept related to the link information based on the configuration condition. The document data processing apparatus according to claim 1. An input means for inputting document data including management contents for risk reduction of business activities;
A surface expression storage unit for storing a surface expression dictionary in which a concept name and a surface expression of the concept name are associated with each other;
A hierarchical concept dictionary including system information of concept names and link information between a first concept name of the concept names and a second concept name that is not a higher concept name or a lower concept name of the first concept name is stored. Hierarchical concept storage means;
A syntax analysis unit that performs syntax analysis of sentences in the input document data;
Extraction means for extracting from the surface expression dictionary a concept name associated with a surface expression related to a plurality of expressions that serve as a material for determining the validity of the description of the document data being output from the parsing unit;
A determination item extraction unit that extracts a relationship between concepts extracted by the extraction unit as a determination item;
Based on system information and link information of the hierarchical concept dictionary, it is determined that the description in the input document data is not correct when the relationship between the concept names extracted by the determination item extraction unit does not satisfy a predetermined condition And a document data processing apparatus.

Images