JP2010113399A - Pharmaceutical affairs law feature model creating device and pharmaceutical affairs law feature model creating method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the burden of confirmation works by visualizing the features of the Pharmaceutical Affairs Law. <P>SOLUTION: The Pharmaceutical Affairs Law feature model creating device acquires the article text data of the Pharmaceutical Affairs Law and the input from outside, creates conversion rules for converting the article text data into a Pharmaceutical Affairs Law feature model indicating the features of the Pharmaceutical Affairs Law based on the acquired input information, and registers the rules in a database. The device extracts the plurality of features from the article text data based on the registered conversion rules, creates the Pharmaceutical Affairs Law model containing the information regarding the features and the information to associate the features mutually, and registers the model in the database for management. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an apparatus and method for analyzing a legal structure and converting it into a data format that can be handled on a computer.

  A company that manufactures and sells products regulated by the Pharmaceutical Affairs Law (hereinafter referred to as “manufacturer / seller”) must comply with the Pharmaceutical Affairs Law and regulations related to the Pharmaceutical Affairs Law. For this reason, it is necessary to verify that the company's standard business process is documented as SOP (Standard Operation Procedure) and that it is deployed in the relevant business manuals and business systems. On the other hand, in the Pharmaceutical Affairs Law and the Pharmaceutical Affairs Law-related ministerial ordinances, the provisions that must be observed differ depending on the product category. Therefore, manufacturers and distributors are required to strictly manage the Pharmaceutical Affairs Law and the Pharmaceutical Affairs Law related ministerial orders and their SOPs, business manuals, business systems, etc., that is, to ensure traceability of the Pharmaceutical Affairs Law. .

  However, the Pharmaceutical Affairs Law itself is an unstructured document, and in order to perform the above management in cooperation with the SOP, business manual, and business system that are generally managed on a computer, Need to be analyzed and converted to a data format (model) that can be handled by a computer.

As a conventional technique related to conversion of an unstructured document, a character string (hereinafter referred to as “keyword”) expressing a logical structure such as “Chapter 1” or “1.1” is extracted from the unstructured document. A method of generating a structured document by recognizing the logical structure of the entire document using extracted keywords as a clue is known (for example, see Patent Documents 1 and 2).
JP 2004-295500 A JP 2001-312489 A

  However, the above prior art has the following problems.

(1) Since there is no means to analyze the provisions (unstructured documents) of the Pharmaceutical Affairs Law and Pharmaceutical Affairs Law-related ministerial ordinances and reflect the characteristics (at least classification) of the Pharmaceutical Affairs Law and Pharmaceutical Affairs Law-related ministerial ordinances in the structure, it is difficult to handle with computers .

(2) The characteristics of the Pharmaceutical Affairs Law and the Pharmaceutical Affairs Law-related ministerial ordinances cannot be expressed in an easy-to-understand format for users (manufacturers and distributors), and it takes a lot of time to confirm the articles to be observed.

  Therefore, in view of the above-described problems of the prior art, the present invention visualizes the characteristics (common part and variable part) of the Pharmaceutical Affairs Law and the Ministerial Ordinance related to the Pharmaceutical Affairs Law and creates a Pharmaceutical Affairs Law feature model creation apparatus and creation that can reduce the burden of confirmation work It aims to provide a method.

  The Pharmaceutical Affairs Law feature model creation device according to the present invention includes an interface section for acquiring the Pharmaceutical Affairs Law text data and external input information, a Pharmaceutical Affairs Law feature model representing the characteristics of the Pharmaceutical Affairs Law, and the predefined text data. A database unit for storing the conversion rule, and a conversion rule for creating and registering the conversion rule on the basis of the external input information acquired in the interface unit, and managing update and reference of the conversion rule The pharmaceutical law feature model including a management unit and a plurality of features extracted from the sentence text data based on the conversion rules stored in the database unit, and information relating to the features and information associating the features with each other Create a Feach A model conversion unit, and registers the converted Pharmaceutical Affairs Law feature model to the database unit, and having a a feature model information managing unit for managing the update and reference of the Pharmaceutical Affairs Law feature model.

  The Pharmaceutical Affairs Law feature model creation method according to the present invention includes a data acquisition step for acquiring text data and external input information of the Pharmaceutical Affairs Law, and the text data of the text based on the acquired external input information. A conversion rule registration step for creating a conversion rule for converting into a pharmacy law feature model representing the data and registering it in the database, and extracting a plurality of features from the text data based on the registered conversion rule, A feature model conversion step of creating the pharmacy law feature model including information relating to the feature and information associating the features with each other, and a feature model registration step of registering the converted pharmacy law feature model in the database. It is characterized by

  ADVANTAGE OF THE INVENTION According to this invention, the pharmaceutical law feature model creation apparatus and creation method which can visualize the characteristic (common part and variable part) of the Pharmaceutical Affairs Law and related ministerial ordinances, and can reduce the burden of confirmation work are provided.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an example of the overall configuration of a pharmaceutical law feature model creation device 1 according to an embodiment of the present invention. As shown in the figure, the Pharmaceutical Affairs Law feature model creation device 1 includes an interface unit 11, a database unit 12, a conversion rule management unit 13, a feature model conversion unit 14, and a feature model information management unit 15. It is a computer that creates a pharmaceutical law feature model from text data. A “feature” is one of typical units for managing requirements, and includes both functions and non-functions. The “feature model” refers to a description format for capturing features hierarchically and a data group described based on the description format. In this feature model, the result of analyzing the commonality and variability of features can be expressed in a hierarchical and / or tree. In the present embodiment, a feature model specialized in the Pharmaceutical Affairs Law and the Pharmaceutical Affairs Law-related ministerial ordinance (hereinafter also referred to as “Pharmaceutical Affairs Law”) is referred to as a Pharmaceutical Affairs Law feature model. This Pharmaceutical Affairs Law feature model expresses the requirements under the Pharmaceutical Affairs Law that must be followed for each product by a company that manufactures and sells products regulated by the Pharmaceutical Affairs Law (hereinafter referred to as “manufacturers”). . FIG. 2 is a diagram showing a specific example of the Pharmaceutical Affairs Law feature model.

  The interface unit 11 is a program that provides a screen for inputting and outputting various types of information to the user and an interface for data communication. External input information from the user is delivered to a conversion rule management unit 13, a feature model conversion unit 14, and a feature model information management unit 15 connected to the interface unit 11. In the interface unit 11, the text data of the Pharmaceutical Affairs Law is acquired by data communication with the outside.

  FIG. 3 is a block diagram illustrating a configuration example of the database unit 12. As shown in the figure, the database unit 12 includes a feature conversion rule definition table 121, a division conversion rule definition table 122, a semi-conversion conversion rule definition table 123, a replacement conversion rule definition table 124, a feature information storage table 125, and a feature relation. It is a database including an information storage table 126. The database unit 12 is connected to a conversion rule management unit 13, a feature model conversion unit 14, and a feature model information management unit 15, respectively.

  The feature conversion rule definition table 121 stores feature conversion rules. FIG. 4 is a diagram illustrating a specific example of the feature conversion rule definition table 121. Here, the feature conversion rule ID, the feature conversion rule, and the hierarchy are listed as items. The feature conversion rule is described by a regular expression. If there is a matching character string in the text data, the part is extracted as a feature of the specified hierarchy. For example, if the text string of “Chapter 1” is included in the text text data, it conforms to the feature conversion rule of “Chapter {$ 1}”, so “Chapter 1” is the first layer. It is extracted as a feature.

  The classification conversion rule definition table 122 stores classification conversion rules. “Category” means a product category regulated by the Pharmaceutical Affairs Law. Examples of the category include “pharmaceuticals”, “quasi-drugs”, “cosmetics”, and “medical devices”. FIG. 5 is a diagram showing a specific example of the segment conversion rule definition table 122. Here, the category conversion rule ID and the category conversion rule are listed as items, and the category conversion rule is described by a regular expression. If there is a matching character string in the text text data, a predetermined section is added to the feature related to the part.

  The quasi-use conversion rule definition table 123 stores quasi-use conversion rules. FIG. 6 is a diagram showing a specific example of the quasi-conversion rule definition table 123. Here, the quasi-use conversion rule ID and the quasi-use conversion rule are listed as items, and the quasi-use conversion rule is described by a regular expression. If there is a matching character string in the text text data, related information based on the quasi-application information is added to the feature related to the part.

  The replacement conversion rule definition table 124 stores replacement conversion rules. FIG. 7 is a diagram showing a specific example of the replacement conversion rule definition table 124. Here, the replacement conversion rule ID and the replacement conversion rule are listed as items, and the replacement conversion rule is described by a regular expression. When there is a matching character string in the text data, a new feature based on the replacement information is added to the feature related to the part.

  The feature information storage table 125 stores feature information. FIG. 8 is a diagram showing a specific example of the feature information storage table 125. Here, feature ID, feature name, classification (pharmaceutical, quasi-drug, cosmetic, medical device) and text are listed as items.

  The feature related information storage table 126 stores feature related information.

FIG. 9 is a diagram showing a specific example of the feature related information storage table 126. Here, the original feature ID, the destination feature ID, and the related category are listed as items. The original feature and the destination feature are associated with each other, and the original feature indicates a higher-level feature. For example, a feature “F002” (feature name: Article 1) is associated with a lower layer of the feature whose original feature ID is “F001” (feature name: Chapter 1). The related category is a flag for determining whether or not the category to be observed in the original feature is included in the category to be observed in the previous feature. For example, in the case of a feature whose original feature ID is “F013” (feature name: Article 4), the categories to be observed are “pharmaceuticals, quasi-drugs, cosmetics, medical devices”, but the previous feature ID However, in “F015” (feature name: (article 4), item 2), there are two categories of “pharmaceuticals and medical devices”, so the related category is set to “selection”. On the other hand, if the category to be observed in the previous feature matches the category of the original feature, “required” is set in the related category.

  FIG. 10 is a block diagram illustrating a configuration example of the conversion rule management unit 13. As shown in the figure, the conversion rule management unit 13 includes a conversion rule management distribution unit 131, a feature conversion rule management unit 132, a classification conversion rule management unit 133, a quasi-conversion conversion rule management unit 134, and a replacement conversion rule management unit. 135 is a program that has a function of registering various conversion rules delivered from the interface unit 11 in the database unit 12 and updating and referencing the database unit 12. The conversion rule management unit 13 is connected to the interface unit 11, the database unit 12, and the feature model conversion unit 14, respectively.

  The conversion rule management distribution unit 131 distributes various information to each management unit based on the information delivered from the interface unit 11, and delivers the processing result from each management unit to the interface unit 11.

  The feature conversion rule management unit 132 is activated based on the information from the conversion rule management distribution unit 131, registers the feature conversion rule created from the input information acquired from the interface unit 11 in the feature conversion rule definition table 121, and It is a program having a function of updating (modifying / deleting) and referencing.

  The division conversion rule management unit 133 is activated based on the information from the conversion rule management distribution unit 131, registers the division conversion rule created from the input information acquired from the interface unit 11 in the division conversion rule definition table 122, and It is a program having a function of updating (modifying / deleting) and referencing.

  The quasi-conversion rule management unit 134 is activated based on the information from the conversion rule management allocating unit 131, registers the quasi-conversion rule created from the input information acquired from the interface unit 11 in the quasi-conversion conversion rule definition table 123, and It is a program having a function of updating (modifying / deleting) and referencing.

  The replacement conversion rule management unit 135 is activated based on the information from the conversion rule management distribution unit 131, registers the replacement conversion rule created from the input information acquired from the interface unit 11 in the replacement conversion rule definition table 124, and It is a program having a function of updating (modifying / deleting) and referencing.

  FIG. 11 is a block diagram illustrating a configuration example of the feature model conversion unit 14. As shown in the figure, the feature model conversion unit 14 includes a conversion processing distribution unit 141, a feature conversion processing unit 142, a segment conversion processing unit 143, a semi-conversion conversion processing unit 144, and a read conversion processing unit 145, and includes an interface unit. 11 and the conversion rule management unit 13 are programs for converting the text data of the Pharmaceutical Affairs Law into the Pharmaceutical Affairs Law feature model. The feature model conversion unit 14 is connected to the interface unit 11, the database unit 12, and the conversion rule management unit 13. The feature model conversion unit 14 delivers the pharmacy feature model created by the conversion process to the feature model information management unit 15. 12 is registered.

  Based on the information delivered from the interface unit 11, the conversion processing distribution unit 141 distributes various types of information to each processing unit, and converts the text data of the Pharmaceutical Affairs Law into a pharmaceutical law feature model using each processing section. In addition, the processing result from each processing unit is delivered to the interface unit 11.

  The feature conversion processing unit 142 is a program that is activated based on information from the conversion processing distribution unit 141 and applies a feature conversion rule to text data of the Pharmaceutical Affairs Law to convert it into a feature.

  The classification conversion processing unit 143 is activated based on the information from the conversion processing distribution unit 141, determines the classification that each feature is a restriction target, and sets the classification.

  The quasi-use conversion processing unit 144 is activated based on the information from the conversion process allocating unit 141, applies the quasi-use conversion rule to the text data of the Pharmaceutical Affairs Law, acquires the quasi-use information, and the feature related information based on the quasi-use information Set the related category (required / selected).

  The replacement conversion processing unit 145 is activated based on information from the conversion processing distribution unit 141, applies the replacement conversion rule to the text data of the Pharmaceutical Affairs Law, and adds the replacement part as a feature.

  FIG. 12 is a block diagram illustrating a configuration example of the feature model information management unit 15. As shown in the figure, the feature model information management unit 15, the feature model information management unit 15, the information management distribution unit 151, the feature information management unit 152, the feature related information management unit 153, and the display model creation unit 154 The feature model information created by the input information from the interface unit 11 or the conversion process in the feature model conversion unit 14 is registered in the database unit 12, and the database unit 12 stores the feature model information based on the information delivered from the interface unit 11. It is a program having a function of performing update and reference. The feature model information management unit 15 is connected to the interface unit 11, the database unit 12, and the feature model conversion unit 14, respectively.

  Based on the information delivered from the interface unit 11, the information management distribution unit 151 activates the feature information management unit 152, the feature related information management unit 153, or the display model creation unit 154 for various information, and distributes the processing. In addition, the information management distribution unit 151 delivers the processing result in each processing unit to the interface unit 11.

  The feature information management unit 152 is activated based on the information from the information management distribution unit 151, and the feature information created by the input information from the interface unit 11 or the conversion processing in the feature model conversion unit 14 is stored in the feature information storage table 125. Is a program that has the function of performing registration (update / deletion) and reference.

  The feature related information management unit 153 is activated based on the information from the information management distribution unit 151, and the feature related information generated by the input information from the interface unit 11 or the conversion processing in the feature model conversion unit 14 is used. It is a program that has a function of registering in the storage table 126 and performing update (correction / deletion) and reference.

  The display model creation unit 154 acquires feature information from the feature information storage table 125 and feature related information from the feature related information storage table 126 based on input information from the interface unit 11, and based on the feature information and feature related information. This is a program that creates display model data that visualizes the Pharmaceutical Affairs Law feature model. This display model is displayed on a display device (not shown) by being delivered to the interface unit 11, and the user can easily grasp the characteristics of the Pharmaceutical Affairs Law.

  Hereinafter, the operation of the pharmaceutical law feature model creation device 1 according to an embodiment of the present invention will be described with reference to the drawings. FIG. 13 is a flowchart showing a specific example of processing of the pharmaceutical law feature model creation device 1 according to an embodiment of the present invention.

  In step S <b> 1301, the interface unit 11 acquires various input data input by a user from a business screen (not illustrated).

  In step S1302, the interface unit 11 performs processing distribution according to the type of input data. If the input data is determined to be a conversion rule registration / update request, the process advances to step S1303. If it is determined that the input data is a conversion request to the feature model, the process proceeds to S1304. If it is determined that the input data is a registration / update / display request for feature model information, the process advances to step S1306. If it is determined that the input data is other data, the process proceeds to S1309.

  In step S1303, the interface unit 11 activates the conversion rule management unit 13, and causes the conversion rule management unit 13 to perform conversion rule registration / update processing based on input data from the user.

  In step S <b> 1304, the interface unit 11 activates the feature model conversion unit 14 and causes the feature model conversion unit 14 to convert the text data of the Pharmaceutical Affairs Law into a pharmaceutical law feature model.

  In S 1305, the feature model conversion unit 14 outputs the feature model information obtained by various conversion processes to the feature model information management unit 15, and outputs it to the database unit 12 (the feature information storage table 125 and the feature related information storage table 126). To register or update.

  In step S1306, the interface unit 11 activates the feature model information management unit 15, outputs the feature model information created based on the input data from the user to the feature model information management unit 15, and registers / updates the feature model information.・ Perform display processing.

  In step S <b> 1307, the conversion rule management unit 13, the feature model conversion unit 14, or the feature model information management unit 15 determines whether there is output data for the interface unit 11. If it is determined that there is output data, the process advances to step S1308. On the other hand, if it is determined that there is no output data, the process advances to step S1309.

  In S <b> 1308, the conversion rule management unit 13, the feature model conversion unit 14, or the feature model information management unit 15 outputs data to the interface unit 11.

  In step S1309, the interface unit 11 determines whether there is data input from the user. If it is determined that there is an input from the user, the process returns to S1301, and the processes of S1301 to S1309 are repeated. On the other hand, when it is determined that there is no input from the user, the process is terminated.

<Conversion rule registration / update process>
FIG. 14 is a flowchart showing a specific example of processing of the conversion rule management unit 13, and corresponds to S1303 of FIG.

  In step S <b> 1401, the conversion rule management distribution unit 131 acquires data related to a conversion rule registration / update request from the interface unit 11.

  In S1402, the conversion rule management distribution unit 131 distributes processing according to the acquired data. That is, the process is distributed according to the type of conversion rule for performing the registration / update process.

  In step S1403, the conversion rule management distribution unit 131 activates the feature conversion rule management unit 132, causes the feature conversion rule management unit 132 to create a feature conversion rule based on the input data acquired from the interface unit 11, and converts the feature. The rule definition table 121 is made to perform feature conversion rule registration processing or update processing.

  In S1404, the conversion rule management distribution unit 131 activates the division conversion rule management unit 133, causes the division conversion rule management unit 133 to create a division conversion rule based on the input information acquired from the interface unit 11, and performs the division conversion. The rule definition table 122 is made to perform the classification conversion rule registration process or update process.

  In step S1405, the conversion rule management distribution unit 131 activates the quasi-conversion rule management unit 134, causes the quasi-conversion rule management unit 134 to create a quasi-conversion rule based on input data acquired from the interface unit 11, and converts the quasi-conversion The rule definition table 123 is made to perform registration processing or update processing of the quasi-conversion rules.

  In S1406, the conversion rule management distribution unit 131 activates the replacement conversion rule management unit 135, causes the replacement conversion rule management unit 135 to create a replacement conversion rule based on the input data acquired from the interface unit 11, and replaces the conversion. The rule definition table 124 is caused to perform a replacement conversion rule registration process or update process.

  In step S <b> 1407, the feature conversion rule management unit 132, the division conversion rule management unit 133, the semi-conversion conversion rule management unit 134, or the replacement conversion rule management unit 135 determines whether there is output data for the conversion rule management distribution unit 131. If it is determined that there is output data, the process advances to step S1408. On the other hand, if it is determined that there is no output data, the process advances to step S1409.

  In step S1408, the data is output to the conversion rule management distribution unit 131.

  In step S1409, the conversion rule management distribution unit 131 determines whether there is unprocessed acquired data. If it is determined that there is unprocessed data, the process returns to S1402, and the processes of S1402 to S1409 are repeated until all the processes are completed. On the other hand, if it is determined that there is no unprocessed data, the process is terminated.

<Feature model conversion process>
FIG. 15 is a flowchart showing a specific example of the process of the feature model conversion unit 14, and corresponds to S1304 in FIG.

  In step S <b> 1501, the conversion processing distribution unit 141 acquires data related to the feature model conversion request from the interface unit 11.

  In step S1502, the conversion processing distribution unit 141 determines a program to be activated based on the processing procedure described in the program.

  In step S <b> 1503, the feature conversion processing unit 142 acquires a feature conversion rule from the feature conversion rule definition table 121.

  In step S1504, the feature conversion processing unit 142 performs feature conversion of the text text data based on the acquired feature conversion rule. FIG. 16 is a diagram for specifically explaining feature conversion (first layer) in the feature conversion processing unit 142. Here, it is shown that a chapter number is extracted as a feature of the first layer based on the feature conversion rule (“# {$ 1} chapter)” shown in FIG. FIG. 17 is a diagram illustrating a result of feature conversion (first layer).

  FIG. 18 is a diagram for specifically explaining feature conversion (second level) in the feature conversion processing unit 142. Here, it is shown that the item number is extracted as the feature of the second layer based on the feature conversion rule (“Item {$ 1}”) shown in FIG. FIG. 19 is a diagram illustrating the result of the feature conversion (second layer) in FIG. The feature related to the section number is arranged as related to the feature related to the chapter number. For example, when the original feature is “Chapter 1”, it is shown that feature-related information having “Article 1” and “Article 2” as destination features is created.

  FIG. 20 is a diagram specifically explaining feature conversion (third layer) in the feature conversion processing unit 142. Here, it is shown that the item number is extracted as the feature of the third layer based on the feature conversion rule (“the {$ 1} item”) shown in FIG. FIG. 21 is a diagram showing the result of the feature conversion (third layer) in FIG. The feature related to the item number is arranged as related to the feature related to the item number. For example, when the original feature is “Article 2”, it is shown that feature-related information having “1st term” to “5th term” as the previous feature is created.

  In step S <b> 1505, the partition conversion processing unit 143 acquires the partition conversion rule from the partition conversion rule definition table 122.

  In step S1506, the section conversion processing unit 143 performs section conversion of the text text data based on the acquired section conversion rule. FIG. 22 is a diagram for specifically explaining the segment conversion in the segment conversion processing unit 143. Here, if the text of “Chapter 2 Pharmaceuticals” is included in the text data, it will conform to the section conversion rules of “Chapter {. *} Chapter {. *} Pharmaceuticals {. *}”. Therefore, a medicine is added as a feature classification of “Chapter 2”. FIG. 23 is a diagram illustrating a result of the segment conversion in FIG.

  In S <b> 1507, the quasi-conversion processing unit 144 acquires a quasi-conversion rule from the quasi-conversion rule definition table 123.

  In step S1508, the quasi-use conversion processing unit 144 performs quasi-use conversion of the text text data based on the acquired quasi-use conversion rule. FIG. 24 is a diagram specifically describing the quasi-use conversion in the quasi-use conversion processing unit 144. Here, the text text data includes the character string “Applying Article 19, Article 3, ...”, and the “Item {$ 1} for Application {$ 2}” shown in FIG. * Since it conforms to the quasi-conversion rule of “*”, the related information is created as the “Article 19” feature and the “Article 3” feature are related. FIG. 25 is a diagram showing the result of the semi-conversion of FIG.

  In step S <b> 1509, the replacement conversion processing unit 145 acquires a replacement conversion rule from the replacement conversion rule definition table 124.

  In S1510, the replacement conversion processing unit 145 performs replacement conversion of the text text data based on the acquired replacement conversion rule. FIG. 26 is a diagram specifically explaining the read conversion in the read conversion processing unit 145. The text of the text contains the character string “Quality Assurance Manager” in “Article 3, Paragraph 1,“ Quality Assurance Manager as stipulated in Paragraph 3 of the next article ””. Since “{$ 2} in section {$ 1} in section {$ 1} in section 7 is compatible with the replacement conversion rule of {$ 3}” shown in FIG. A feature related to the replacement information is added to. FIG. 27 is a diagram illustrating a result of the read conversion in FIG.

  In step S1511, the presence / absence of output data for the conversion processing distribution unit 141 is determined.

  In step S <b> 1512, data is output to the conversion processing distribution unit 141.

  In step S1513, the conversion processing distribution unit 141 determines whether there is unprocessed acquired data. If it is determined that there is unprocessed data, the process returns to S1502, and the processes of S1502 to S1513 are repeated until all the processes are completed. On the other hand, if it is determined that there is no unprocessed data, the process proceeds to S1514.

  In step S1514, the pharmacy feature model is output to the feature model information management unit 15, and the process ends.

<Registration / update / reference processing of the Pharmaceutical Affairs Law feature model>
FIG. 28 is a flowchart showing a specific example of the process of the feature model information management unit 15, and corresponds to S1305 and S1306 in FIG.

  In step S2801, the information management distribution unit 151 acquires data from the interface unit 11 or the feature model conversion unit 14.

  In step S2802, the information management distribution unit 151 performs processing according to the type of acquired data. If the acquired data is determined as feature information, the process advances to S2803. If it is determined that the acquired data is feature-related information, the process proceeds to S2804. If it is determined that the acquired data is a request for displaying the Pharmaceutical Affairs Law feature model, the process proceeds to S2805. If it is determined that the acquired data is other data, the process advances to step S2808.

  In step S2803, the information management distribution unit 151 activates the feature information management unit 152, and performs registration / update processing to the feature information storage table 125 based on the feature information acquired from the interface unit 11 or the feature model conversion unit 14. Do.

  In S 2804, the information management distribution unit 151 activates the feature related information management unit 153, and registers / updates the feature related information storage table 126 based on the feature information acquired from the interface unit 11 or the feature model conversion unit 14. Process.

  In step S2805, the information management distribution unit 151 activates the display model creation unit 154 and constructs display data of the pharmaceutical law feature model based on the information in the feature information storage table 125 and the feature related information storage table 126.

  In step S <b> 2806, the feature information management unit 152, the feature related information management unit 153, or the display model creation unit 154 determines whether there is output data for the information management distribution unit 151. If it is determined that there is output data for the information management distribution unit 151, the process advances to step S2807. On the other hand, if it is determined that there is no output data, the process advances to step S2808.

  In step S <b> 2807, the feature information management unit 152, the feature related information management unit 153, or the display model creation unit 154 outputs data to the information management distribution unit 151. Examples of output data include display data of a pharmaceutical law feature model delivered to the interface unit 11, notification of completion of registration / update processing of feature information and feature related information, and the like.

  In step S2808, the information management distribution unit 151 determines whether there is unprocessed acquired data. If it is determined that there is unprocessed data, the process returns to S2802, and the processes in S2802 to S2808 are repeated until all the processes are completed. On the other hand, if it is determined that there is no unprocessed data, the process is terminated.

  In this way, by converting the text data of the Pharmaceutical Affairs Law and regulations related to the Pharmaceutical Affairs Law into a model that can be handled by a computer, the feature model can be registered, modified, updated, and deleted, and the feature model can be managed centrally. In addition, depending on the product category, it is possible to narrow down and refer to the Pharmaceutical Affairs Law clauses that must be observed, so the workload can be greatly reduced. Furthermore, it can be used as input information in a traceability system that associates the Pharmaceutical Affairs Law, Pharmaceutical Affairs Law-related ministerial ordinances, SOPs, business manuals, and business systems.

  Note that 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. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment.

1 is a block diagram showing an example of the overall configuration of a pharmaceutical law feature model creation device 1 according to an embodiment of the present invention. The figure which shows the specific example of a pharmacy law feature model. FIG. 3 is a block diagram illustrating a configuration example of a database unit 12. The figure which shows the specific example of the feature conversion rule definition table 121. FIG. The figure which shows the specific example of the division | segmentation conversion rule definition table 122. FIG. The figure which shows the specific example of the conversion rule definition table for semi-application. The figure which shows the specific example of the replacement conversion rule definition table. The figure which shows the specific example of the feature information storage table 125. FIG. The figure which shows the specific example of the feature related information storage table 126. FIG. The block diagram which shows the structural example of the conversion rule management part 13. FIG. FIG. 3 is a block diagram illustrating a configuration example of a feature model conversion unit 14. FIG. 3 is a block diagram illustrating a configuration example of a feature model information management unit 15. The flowchart which shows the specific example of the process of the pharmacy law feature model production apparatus 1 which concerns on one Embodiment of this invention. 10 is a flowchart illustrating a specific example of processing of a conversion rule management unit 13. 5 is a flowchart showing a specific example of processing of a feature model conversion unit 14. The figure which demonstrates the feature conversion (1st hierarchy) in the feature conversion process part 142 concretely. The figure which shows the result of the feature conversion (1st hierarchy) of FIG. The figure which demonstrates the feature conversion (2nd hierarchy) in the feature conversion process part 142 concretely. The figure which shows the result of the feature conversion (2nd hierarchy) of FIG. The figure which demonstrates the feature conversion (3rd hierarchy) in the feature conversion process part 142 concretely. The figure which shows the result of the feature conversion (3rd hierarchy) of FIG. The figure explaining concretely the division conversion in the division conversion process part 143. FIG. The figure which shows the result of the division conversion of FIG. FIG. 10 is a diagram for specifically explaining the quasi-use conversion in the quasi-use conversion processing unit 144; The figure which shows the result of semi-conversion of FIG. FIG. 10 is a diagram for specifically explaining read conversion in a read conversion processing unit 145. The figure which shows the result of the replacement conversion of FIG. 10 is a flowchart showing a specific example of processing of the feature model information management unit 15.

Explanation of symbols

1… Pharmacy Law feature model creation device,
11 ... Interface part,
12 ... Database section,
13 ... Conversion rule management unit,
14 ... feature model conversion unit,
15 ... Feature model information management department,
121 ... Feature conversion rule definition table,
122 ... Classification conversion rule definition table,
123 ... Conversion rule definition table for quasi-use,
124 ... replacement conversion rule definition table,
125 ... Feature information storage table,
126 ... Feature related information storage table,
131 ... Conversion rule management distribution unit,
132. Feature conversion rule management unit,
133... Section conversion rule management unit,
134 ... Conversion rule management unit for quasi-use,
135 ... replacement conversion rule management unit,
141 ... Conversion processing distribution unit,
142 ... feature conversion processing unit,
143 ... division conversion processing unit,
144... Conversion processing unit for quasi-use,
145 ... replacement conversion processing unit,
151. Information management distribution unit,
152 ... feature information management unit,
153. Feature related information management department,
154 ... Display model creation unit.

Claims (5)

An interface part for obtaining the text data and external input information of the Pharmaceutical Affairs Law,
A database part that stores a conversion rule relating to a pharmacy law feature model representing the characteristics of the pharmacy law and the text data;
A conversion rule management unit that creates the conversion rule based on the external input information acquired in the interface unit and registers the conversion rule in the database unit, and manages update and reference of the conversion rule;
A feature that extracts a plurality of features from the text data based on the conversion rule stored in the database unit and creates the pharmaceutical law feature model including information on the feature and information for associating the features with each other A model converter,
A feature model information management unit that registers the converted pharmacy law feature model in the database unit and manages update and reference of the pharmacy law feature model;
A pharmaceutical law feature model creation device characterized by comprising: The feature model conversion unit
A feature conversion processing unit that extracts a plurality of features from the text data based on the feature conversion rules stored in the database unit, and creates the pharmaceutical law feature model from the information related to the features;
A section conversion processing section for extracting a section related to the feature from the text data based on the section conversion rule stored in the database section, and adding the section to the pharmaceutical law feature model;
Applying quasi-use information related to the feature from the text data based on the quasi-use conversion rule stored in the database unit, and adding information relating the features to the pharmacy feature model based on the quasi-use information A conversion processing unit;
A replacement conversion processing unit that extracts replacement information related to the feature from the text data based on the replacement conversion rule stored in the database unit, and adds a new feature to the pharmacy feature model based on the replacement information When,
The pharmaceutical law feature model creation device according to claim 1, wherein:   The pharmaceutical feature model creation apparatus according to claim 1 or 2, wherein the feature conversion processing unit extracts at least a chapter number, a section number, and a section number from the section text data as the feature. A data acquisition step for acquiring text data and external input information of the Pharmaceutical Affairs Law,
A conversion rule registration step of creating a conversion rule for converting the text data of the article into a pharmacy law feature model representing the characteristics of the pharmacy law based on the acquired external input information and registering it in a database;
A feature model conversion step of extracting a plurality of features from the sentence text data based on the registered conversion rule and creating the pharmacy feature model including information relating to the feature and information for associating the features; ,
A feature model registration step of registering the converted pharmaceutical law feature model in the database;
A pharmaceutical law feature model creation method characterized by comprising: The feature model conversion step includes:
A feature conversion step of extracting a plurality of features from the article text data based on the feature conversion rule registered in the conversion rule registration step, and creating the pharmaceutical law feature model from the information relating to the feature;
A category conversion step of extracting a category related to the feature from the text data based on the category conversion rule registered in the conversion rule registration step, and adding the category to the pharmaceutical law feature model;
Based on the quasi-use conversion rule registered in the conversion rule registration step, quasi-use information related to the feature is extracted from the text data and information for associating the features with each other is added to the pharmacy law feature model. A correspondingly applied conversion step;
A replacement conversion that extracts replacement information related to the feature from the text data based on the replacement conversion rule registered in the conversion rule registration step, and adds a new feature to the pharmacy law feature model based on the replacement information. Steps,
The method of creating a pharmaceutical law feature model according to claim 4, comprising: