Document creation support device, method, and program

Abstract

To provide a support technique for creating a document form related to application without manual intervention, based on many related books on a facility.SOLUTION: A form creation support apparatus 10 includes: a first selection unit 11 which selects at least one of multiple related books 15 of a facility for which a document form 16 is to be created; a second selection unit 12 which selects one of multiple pieces of format data 18 of the document form 16 with fixed entry fields; an extraction unit 24 which extracts related information 29 related to each of the entry fields 13 from the related books 15 on the basis of an extraction rule 19; an allocation unit 25 which allocates sentences created on the basis of the related information 29 to each of the entry fields 13; and an output unit 26 which outputs the document form 16 with the sentences allocated to the entry fields 13 in a predetermined style.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technique for supporting the creation of forms. [Background technology]

[0002] When constructing a plant or facility, you may be required to submit an application for approval to confirm the conformity of the construction. Such application forms are often standardized forms with blank spaces for information, and are designed to be compatible with equipment common to different facilities. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Nuclear Regulation Authority website "Laws and Standards" Summary of the Invention [Problem to be solved by the invention]

[0004] In addition to design documents, documents related to plants and equipment also include documents on regulations, required specifications, etc. These documents contain complex text and a wide range of information that is interrelated (linked) with each other.

[0005] The above-mentioned application forms for approval have been prepared by manually extracting information related to the target facilities from the above-mentioned multiple documents. This has required considerable effort to prepare the application forms, and if the documents are revised, the relevant information has to be extracted again by hand.

[0006] The embodiment of the present invention has been made taking these circumstances into consideration, and aims to provide a support technology for creating application-related forms without human labor based on a large number of related documents related to equipment. [Means for solving the problem]

[0007] In the document creation support device according to the embodiment, when a facility for which a document is to be created is designated by a user operation via an information input interface, a document related to the facility is linked to the document in advance. By doing so, or From the constructed database The name of the equipment specified by the user operation search To do The first selection section selects one or more from a plurality of items, and format data of the form in which entry fields are set in a standard format is created and saved in advance. The equipment To check the suitability of the construction work To the licensing authority The information that matches what is required to be submitted is input by the user through the information input interface. depending on a second selection section for selecting one of the related documents, and the selected related documents are sorted into setting categories corresponding to each of the entry fields; The selected an extraction unit that extracts related information related to each of the entry fields from the related book based on an extraction rule; of Changed the wording By doing an allocation unit that allocates the generated sentences to each of the entry columns in accordance with the allocated expression pattern models; and a data processing unit that stores the form in which the sentences are allocated to the entry columns in an electronic medium or a paper medium. With your body and an output unit for outputting the extraction rules, the extraction rules being based on know-how accumulated by the user up to now. By fixed To do , and machine learning based on past examples To , and, The characteristics of the text used in the related information Signs Based on By at least one of the things The related documents are created and stored as data, and include rule documents describing the rules that apply to the facility, design documents created for constructing the facility, and documents related to the facility. Created by Other existing documents that were affected 、 At least one of the following applies, and the extraction rules refer to a technical term dictionary file. Teshoa function of extracting the related information from at least the existing documents, a function of extracting the related information corresponding to the classification items from at least the regulation documents by referring to a classification model having a plurality of classification items, and a function of extracting the related information corresponding to the classification items from at least the design documents by referring to a keyword file. Subject to the construction work Items in the equipment specifications table 、 and search the information in the manual By narrowing down a function of extracting the related information, and the first selection unit, the second selection unit, the extraction unit, the allocation unit, and the output unit are integrated into a processor. More realization It is characterized by being [Effects of the Invention]

[0008] An object of the present invention is to provide a support technology for creating application forms without manual labor based on a large number of related documents related to facilities. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a block diagram showing a form creation support device according to each embodiment of the present invention; [Figure 2] FIG. 4 is an explanatory diagram of a process for creating a form from related documents in the first embodiment. [Figure 3] FIG. 10 is an explanatory diagram of a process for creating a form from related documents (existing forms) in the second embodiment. [Figure 4] Table illustrating a model of expression patterns. [Figure 5] FIG. 10 is an explanatory diagram of a process for creating a form from related documents (regulation documents) in the second embodiment. [Figure 6] FIG. 10 is an explanatory diagram of a process for creating a form from related documents (design documents) in the second embodiment. [Figure 7] 3 is a flowchart illustrating steps of a form creation support method and an algorithm of a form creation support program according to each embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] (First embodiment) Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. Fig. 1 is a block diagram showing a form creation support device 10 according to a first embodiment of the present invention. As described above, the form creation support device 10 includes a first selection unit 11 that selects one or more related documents 15 for equipment for which a form 16 is to be created from a plurality of documents; a second selection unit 12 that selects one form data 18 for the form 16, each of which has a fixed set of entry fields 13 (Fig. 2); an extraction unit 24 that extracts related information 29 related to each entry field 13 from the related documents 15 based on extraction rules 19; an allocation unit 25 that allocates text generated based on the related information 29 to each entry field 13; and an output unit 26 that outputs the form 16, with the text allocated to the entry fields 13, in a predetermined format.

[0011] The related documents 15 here include rule documents 15a describing the rules that apply to the equipment, design documents 15b created for constructing the equipment, and other existing documents 15c for which the equipment was created. These related documents 15 are stored as electronic data in a database DB constructed on a server of a communication network 28. The document creation support device 10 can access the database DB via this communication network 28 to search for and acquire the desired related documents 15.

[0012] These related documents 15 are used for quality control of the equipment, and the report 16 created from them is a list for confirming the conformity between the equipment design requirements and the design results. The report creation support device 10 has the function of automatically extracting the necessary parts of these related documents 15 and consolidating them into the report 16 so that they can be viewed at a glance.

[0013] Although not shown in the figure, this report creation support device 10 is composed of a CPU that executes an operating system and application programs, a main memory unit consisting of ROM and RAM, an auxiliary memory unit consisting of a hard disk, a communication control unit consisting of a network card, input units such as a keyboard and mouse, and an output unit such as a monitor.

[0014] Each function of the form creation support device 10 is realized by loading predetermined software onto the CPU or main memory, operating the communication control unit, input unit, output unit, etc. under the control of the CPU, and reading and writing data from and to the main memory or auxiliary memory. Data required for processing (format data 18 of the form 16, extraction rules 19) is stored in the main memory or auxiliary memory. In this embodiment, the database of related books 15 is constructed on a server on the communication network 28, but it may also be constructed inside the form creation support device 10.

[0015] The process of creating a form 16 from related books 15 in the first embodiment will be described with reference to Figure 2. The related books 15 selected by the first selection unit 11 (Figure 1) are assigned to the setting categories 14 associated with each of the entry fields 13. The text assigned to each of the entry fields 13 by the assignment unit 25 (Figure 1) is generated based on the related information 29 included in the related books 15 assigned to the associated setting categories 14.

[0016] Returning to Figure 1, the explanation will continue. The input means 23 is responsible for the information input interface between the user and the form creation support device 10. The input means 23 is made up of a first designation unit 21 that designates the equipment for which the form 16 is to be created, and a second designation unit 22 that designates the format of the form 16 to be created.

[0017] The first selection unit 11 selects from the database DB the related books 15 for the equipment specified by the first specification unit 21. The related books 15 related to the equipment for which the report 16 is to be created are linked in advance, or the related books 15 for data processing can be selected using an automatic search function or the like.

[0018] The second selection unit 12 selects one of the plurality of stored form data 18 as the format of the form 16 specified by the second specification unit 22. The form 16 is configured as a collection of a plurality of electronic forms that correspond to the attributes of the entry fields 13 (Fig. 2). The form data 18 for a plurality of forms 16 that have different attributes as a collection of the entry fields 13 are created and stored in advance.

[0019] The extraction unit 24 extracts related information 29 related to each entry field 13 from the related book 15 based on the extraction rules 19. The extraction rules 19 can be defined by compiling know-how accumulated by the user or can be created by machine learning past examples. Alternatively, the extraction rules 19 can be created based on the characteristics of sentences used in the related information 29, the numbering rules for facilities (equipment), etc.

[0020] The allocation unit 25 regenerates the related information 29 extracted from the related books 15 into sentences that fit the entry fields 13, and then electronically allocates the sentences to each entry field 13 (electronic form). The allocation unit 25 generates sentences by separating parts from the extracted related information 29 and changing the sentence expressions.

[0021] The output unit 26 outputs a form 16 in which the entry fields 13 (electronic form) to which text has been assigned have been edited in a predetermined format. The form 16 output may be in the form of an electronic medium or a paper medium. The form 16 output from the output unit 26 can then be manually corrected as needed by the user to become a corrected form 17.

[0022] Furthermore, the form creation support device 10 is equipped with an update unit 27 that updates the extraction rules 19 based on the revised form 17. In this update unit 27, machine learning is performed so that the revised form 17 is output as training data in response to the input of the selected related books 15 and form format data 18. This not only improves the creation accuracy of the form 16, but also makes it easy to respond to revisions to the related books 15 and form format data 18.

[0023] Incidentally, for example, when some of the many pieces of equipment in a facility are changed, some of the many related documents 15 may be revised. In response to such revisions to the related documents 15, it is desirable that the text in the relevant sections of the forms 16 be revised in synchronization.

[0024] Therefore, the form creation support device 10 includes an acquisition unit 35 that acquires difference information 37 between the pre-revision related book 15x and the revised related book 15y, and a detection unit 36 ​​that detects entry fields 13 to which related information 29 affected by this difference information 37 is assigned. This makes it possible to present entry fields 13 that require revision and prompt the user to take action on the relevant form 16. Then, the text in the detected entry fields 13 is revised, and the form 16 is re-output from the output unit 26. This revision of the text in the entry fields 13 may be automatically generated based on the revised related information 29 affected by the difference information 37, or may be performed at the user's discretion.

[0025] This allows only the affected forms 16 to be revised each time, even if many related documents 15 are revised individually. In other words, it is no longer necessary to revise or consider revising all forms 16 at once, even those that are not affected by the revision of the related documents 15. Furthermore, the revision records of many existing forms 16 can be managed individually.

[0026] Furthermore, a verification unit (not shown) can be provided that verifies the validity of a form in which the text in all or part of the entry fields 13 has been manually created by the user, in light of the extraction rules 19. For entry fields 13 that lack validity, the device can also be provided with a function to suggest corrections to the text. In this way, the form creation support device 10 has not only the function of automatically creating all of the form 16, but also the function of further refining a manually created form.

[0027] (Second embodiment) Next, a second embodiment of the present invention will be described with reference to Figures 3 to 6. Note that a form creation support device 10 according to the second embodiment is shown in the same block diagram as Figure 1, and overlapping explanations will be omitted.

[0028] Referring to Figure 3, the process of creating a form 16 from a related document (existing form 15c) in the second embodiment is explained. Note that Figure 4 is a table illustrating an expression pattern model. The extraction unit 24 of the form creation support device 10 (Figure 1) according to the second embodiment extracts related information 29 (29a, 29b) from the existing form 15c by referring to a technical term dictionary file (not shown). The allocation unit 25 then allocates the related information 29a, 29b to the entry columns 13 of the form 16 in accordance with the expression pattern model (Figure 4).

[0029] The extraction unit 24 analyzes the relationships between words registered in the technical terminology dictionary among those contained in the related book 15 in accordance with grammatical theory, and extracts related information 29. The expression pattern model (FIG. 4) is associated in advance with the entry fields 13 of the form 16. The allocation unit 25 then allocates the extracted related information 29 to the corresponding entry fields 13 in accordance with the expression pattern model.

[0030] As a specific example, let us look at the following sentence contained in Related Document 15: "Class 2 components, Class 2 support structures, Class 3 components, and Class 4 piping shall use materials with appropriate mechanical strength and chemical composition for the pressure, temperature, load, and other operating conditions under which they are used. (Sentence 1) Class 1 components (excluding Class 1 vessels), Class 1 support structures (excluding those supporting Class 1 piping and Class 1 valves), Class 2 components, Class 3 components (limited to those belonging to engineered safety facilities), steel pressure sections, steel linings of concrete reactor containment vessels, and core support structures shall use materials with appropriate fracture toughness for their minimum operating temperatures. (Sentence 2) Furthermore, fracture toughness shall be confirmed by dimension, material, or fracture toughness testing. (Sentence 3)"

[0031] The results of analyzing the relationships between the words contained in this sentence that are registered in the technical terminology dictionary are as follows: Furthermore, based on the expression pattern model (Figure 4), the first sentence is assigned to the first form (not shown) of the entry field 13, and the second and third sentences are assigned to the second form (not shown) of the entry field 13. In this manner, related information 29 is assigned to each of the entry fields 13 of the form 16.

[0032] <1st sentence> Main section: [Class 2 equipment, Class 2 support structure, Class 3 equipment, Class 4 pipe] D (for ~): [pressure, temperature, load] o (~): [mechanical strength, chemical composition, material] Predicate: [Use]

[0033] <2nd sentence> Main parts: [Class 1 components, Class 1 support structures, Class 2 components, Class 3 components, steel pressure-resistant parts, concrete reactor containment vessel, steel lining, etc., core support structures] Type 2 (for ~): [Minimum operating temperature] o (~): [fracture toughness, materials] Predicate: [Use]

[0034] <3rd sentence> Main part: [Fracture toughness] Determination (by...): [dimensions, material, fracture toughness, test] Predicate: [Check]

[0035] The process of creating a form 16 from a related document (rule document 15a) in the second embodiment will be described with reference to Figure 5. The extraction unit 24 of the form creation support device 10 (Figure 1) according to the second embodiment extracts related information 29 from at least the rule document 15a (Figure 1) by referring to a classification model (not shown). Then, the allocation unit 25 allocates the related information 29 to the entry columns 13 of the form 16 in accordance with the expression pattern model (Figure 4).

[0036] This classification model (not shown) has multiple classification items such as "definition," "initial declaration," "design conditions" ("installation requirements," "functional requirements," "evaluation requirements"), and "operational requirements" as "requirements types 31." Related information 29 corresponding to the "basic design policy 32" for each classification item is extracted from the rules book 15a (Figure 1). This extraction is performed using a model that uses templates and machine learning.

[0037] The process of creating a form 16 from related documents (design documents 15b) in the second embodiment will be described with reference to Figure 6. The extraction unit 24 of the form creation support device 10 (Figure 1) according to the second embodiment references a keyword file (not shown) to extract related information 29 from at least the design documents 15b (Figure 1). The allocation unit 25 then allocates the related information 29 to the entry columns 13 of the form 16 in accordance with the expression pattern model (Figure 4).

[0038] The extraction of related information 29 by referencing this keyword file (not shown) is performed by searching and narrowing down the items in the specification sheet and the description of the manuals for the related construction and equipment from the database of the design results for construction approval. Furthermore, the names of the manuals, etc. are extracted from the database of the design results for construction approval, and the confirmation methods are also extracted.

[0039] Furthermore, the extracted related information 29 is allocated to the entry columns 13 of the form 16, such as the items in the specifications table of the construction approval design results, the items to be written in the instructions, the name of the instructions for the construction approval design results, etc., the items in the specifications table of the equipment design results, the items to be written in the instructions, the name of the instructions for the equipment design results, etc., and the confirmation method.

[0040] The steps of the form creation support method and the algorithm of the form creation support program according to each embodiment will be described with reference to the flowchart in Figure 7 (see Figure 1 as appropriate). First, one or more related documents 15 for the equipment for which a form 16 is to be created are selected from among a plurality of documents (S11). Next, one form data 18 for the form 16, in which entry fields 13 (Figure 2) are set in a standard format, is selected from among a plurality of documents (S12).

[0041] Furthermore, related information 29 relating to each entry field 13 is extracted from the related book 15 based on the extraction rules 19 (S13). Then, text generated based on the related information 29 is assigned to each entry field 13 (S14). Finally, a form 16 in which text is assigned to the entry fields 13 is output in a predetermined format (S15, END).

[0042] According to at least one of the embodiments of the document creation support device described above, by assigning related information extracted from related documents based on extraction rules to each entry field, it becomes possible to create application documents based on a large number of related documents related to equipment without manual labor.

[0043] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as the inventions described in the claims and their equivalents.

[0044] The form creation support device described above includes a control device with a highly integrated processor such as a dedicated chip, FPGA (Field Programmable Gate Array), GPU (Graphics Processing Unit), or CPU (Central Processing Unit), a storage device such as ROM (Read Only Memory) or RAM (Random Access Memory), an external storage device such as HDD (Hard Disk Drive) or SSD (Solid State Drive), a display device such as a monitor, input devices such as a mouse and keyboard, and a communication I / F, and can be realized with a hardware configuration using a normal computer. In other words, the components of the form creation support device can also be realized by a computer processor.

[0045] The form creation support program may be provided by being pre-installed in a ROM, etc. Alternatively, the program may be provided by being stored in an installable or executable file format on a computer-readable storage medium such as a CD-ROM, CD-R, memory card, DVD, or flexible disk (FD).

[0046] The form creation support program according to this embodiment may be stored on a computer connected to a network such as the Internet and provided by downloading it via the network. The form creation support device may also be configured by combining separate modules that independently perform the functions of the components and are interconnected via a network or dedicated lines. [Explanation of symbols]

[0047] 10... Report creation support device, 11... Related document selection unit (first selection unit), 12... Form data selection unit (second selection unit), 13... Entry field, 14... Setting classification, 15... Related documents, 15a (15)... Regulation documents (related documents), 15b (15)... Design documents (related documents), 15c (15)... Existing reports (related documents), 15x... Related documents before revision, 15y... Related documents after revision, 16... Report, 17... Corrected report, 18... Form data data, 19...extraction rules, 21...creation target designation unit (first designation unit), 22...report designation unit (second designation unit), 23...input means, 24...related information extraction unit (extraction unit), 25...report output unit (allocation unit), 26...report output unit (output unit), 27...extraction rule update unit (update unit), 28...communication network, 29 (29a, 29b)...related information, 35...difference information acquisition unit (acquisition unit), 36...fill-in field detection unit (detection unit).

Claims

1. a first selection unit that, when a piece of equipment for which a report is to be created is designated by a user operation via an information input interface, selects one or more documents related to the equipment from among a plurality of documents by linking the documents to the report in advance or by searching a constructed database for the name of the equipment designated by the user operation; a second selection unit that selects, from among a plurality of form data in which format data for the form with standardized entry fields is created and stored in advance, one that matches the one that is required to be submitted to a licensing authority to confirm the conformity of the construction work for the facility in response to a user operation via the information input interface; an extraction unit that sorts the selected related books into setting categories corresponding to the respective entry fields, and extracts related information related to each of the entry fields from the selected related books based on extraction rules; an allocation unit that allocates sentences generated by changing the sentence expressions of the related information to each of the entry fields in accordance with the allocated expression pattern models; an output unit that outputs the form in which the text is allocated to the entry columns as an electronic medium or a paper medium, The extraction rules are created by at least one of specifying know-how accumulated by a user, learning from past examples, and based on the characteristics of sentences used in the related information, and are stored as data; The related documents correspond to at least one of a rule book describing the rules applicable to the equipment, a design document prepared for constructing the equipment, and other existing documents for which the equipment was created; The extraction rules prescribe a function of extracting the related information from at least the existing forms by referring to a technical term dictionary file, a function of extracting the related information corresponding to the classification items from at least the regulation documents by referring to a classification model having a plurality of classification items, and a function of extracting the related information by searching and narrowing down the items in the specification sheet related to the related construction work, the equipment that is the subject of the construction work, and the descriptions in the instructions from at least the design documents by referring to a keyword file, The form creation support device, wherein the first selection unit, the second selection unit, the extraction unit, the allocation unit, and the output unit are realized by an integrated processor.

2. 2. The document creation support device according to claim 1, A document creation support device that includes an update unit that updates the extraction rules created by machine learning the previous examples based on a modified document that has been modified from the document output on the electronic medium or the paper medium.

3. 3. The document creation support device according to claim 1, an acquisition unit that acquires difference information between the related book before revision and the related book after revision; a detection unit that detects the entry field to which the related information affected by the difference information is assigned, The output unit is a form creation support device that outputs the form in which the detected text in the entry field has been revised based on post-revision related information that is affected by the difference information.

4. a first selection unit, when a user specifies a facility for which a report is to be created by operating an information input interface, selecting one or more of the related documents of the facility by linking the related documents to the facility in advance with the report or by searching a constructed database for the name of the facility specified by the user operation; a step in which a second selection unit selects, from among a plurality of form data created and stored in advance, a form that matches the form that is required to be submitted to a licensing authority to confirm the conformity of the construction work of the facility, in response to a user operation via the information input interface, with the form format data having standardized entry fields; The selected related books are sorted into setting categories corresponding to the respective entry fields, and an extraction unit extracts related information related to each of the entry fields from the selected related books based on an extraction rule; an allocating step in which an allocating unit allocates sentences generated by changing the sentence expressions of the related information to each of the entry fields in accordance with the allocated expression pattern model; an output unit outputting the form in which the text is allocated to the entry fields as an electronic medium or a paper medium; The extraction rules are created by at least one of specifying know-how accumulated by a user, learning from past examples, and based on the characteristics of sentences used in the related information, and are stored as data; The related documents include at least one of a rule book describing the rules applicable to the equipment, a design document prepared for constructing the equipment, and other existing documents for which the equipment was created; The extraction rules prescribe a function of extracting the related information from at least the existing forms by referring to a technical term dictionary file, a function of extracting the related information corresponding to the classification items from at least the regulation documents by referring to a classification model having a plurality of classification items, and a function of extracting the related information by searching and narrowing down the items in the specification sheet related to the related construction work, the equipment that is the subject of the construction work, and the descriptions in the instructions from at least the design documents by referring to a keyword file, A form creation support method, wherein the first selection unit, the second selection unit, the extraction unit, the allocation unit, and the output unit are realized by an integrated processor.

5. On the computer, When a user specifies a piece of equipment for which a report is to be created through an information input interface, selecting one or more of the documents related to the equipment by linking the documents to the report in advance or by searching a database for the name of the equipment specified through the user's operation; form data for the form, in which entry fields are set in a standard format, is created in advance and saved, and one form is selected from among the plurality of form data in response to a user operation via the information input interface, the form matching the form that is required to be submitted to a licensing authority to confirm the conformity of the construction work for the facility; The selected related books are sorted into setting categories corresponding to the respective entry fields, and related information relating to each of the entry fields is extracted from the selected related books based on an extraction rule; a step of allocating sentences generated by changing the sentence expressions of the related information to each of the entry fields in accordance with the allocated expression pattern models; outputting the form in which the text is allocated to the entry columns in an electronic or paper medium; The extraction rules are created by at least one of specifying know-how accumulated by a user, learning from past examples, and based on the characteristics of sentences used in the related information, and are stored as data; The related documents correspond to at least one of a rule book describing the rules applicable to the equipment, a design document prepared for constructing the equipment, and other existing documents for which the equipment was created; The extraction rules include a function to refer to a technical term dictionary file and extract the related information from at least the existing form, a function to refer to a classification model having a plurality of classification items and extract the related information corresponding to the classification items from at least the rule book, and a function to refer to a keyword file and extract the related information from at least the design documents by searching and narrowing down the items in the specification sheet related to the related construction work, the equipment that is the subject of the construction work, and the contents of the instructions.

Images