JP2015094988A - Data structure, data generation device, data generation method, and data generation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data structure and the like capable of ensuring robust data management and efficient operation without the need of complicated table designing and management even for data of no relation and managed by a plurality of systems.SOLUTION: A data structure includes: a universal table 43 for managing data collected from a plurality of ordinary systems 41 using different definitions by item IDs without defining item names; a mask table 45 storing therein mask information for each row unit for necessary item IDs that are minimum necessary item IDs for identifying the data from among the item IDs, the mask information being for enabling only the data stored in the necessary item IDs to be extracted; and a master table 46 for managing a plurality of extracted identical data as one data if the data is extracted from all the data in the universal table 43 according to the mask information stored for every row unit in the mask table 45 and the identical extracted data is extracted from a plurality of rows among the extracted data.

Description

  The present invention relates to a data structure for efficiently managing a large amount of complex data with a simple table design.

  Commonly known database technologies include RDB (Relational Data Base) and KVS (Key-Value Store). RDB is translated as a relational database and is currently most frequently used. Data is managed in tabular tables and defines relationships between tables. Data is acquired, changed, deleted, etc. from multiple tables. On the other hand, KVS is commercialized as a distributed file system or database, and stores a value (Value) as a set for a unique KEY. This is a technology corresponding to recent big data, and data can be acquired by specifying KEY.

  Since RDB has a table format, there is an advantage that the engineer intuitively designs the data in the table, but the table design is different for each system, and the linkage information between the tables must be designed each time. Theoretical design is required and complexity cannot be denied. In addition, because of these characteristics, maintenance, expandability, and flexibility are weak, and in the relationship, the classification of items by the master is often managed in one category, and if the intention of the designer is different from the sensitivity of the user Difficulty in taking out

  On the other hand, the KVS technology solves the flexibility of table design for RDB by having KEY (field information) in the data itself. Also, extensibility is improved because it is not tied to the table structure. However, a plurality of different information is stored in a single field, and the data design in the table cannot be intuitively performed, and it is difficult to grasp from a bird's-eye view.

  As a technique of using such RDB and KVS in combination, for example, a technique disclosed in Patent Document 1 is disclosed. The technique shown in Patent Document 1 is accessed by a data storage destination determination unit that determines to arrange access target data in the first database and the second database when an access pattern exists, and a data storage destination determination unit. When the target data is arranged in the first database and the second database, the period in which the write access is a predetermined number of times or more is determined, and the access target data in the period corresponding to the determined period is arranged in the first database. In this way, the migration schedule is determined, the period in which the read access is a predetermined number of times or more is determined, and the migration schedule is determined so that the access target data in the period corresponding to the determined period is arranged in the second database. And a determination unit.

JP 2012-103847 A

  The technique disclosed in Patent Document 1 uses RDB and KVS separately according to the time required for updating, thereby distributing the load and improving operational efficiency. However, processing such as data migration occurs. As a result, the management becomes complicated and the maintainability becomes poor.

  The present invention can perform robust operation and efficient operation without requiring complicated table design and management even for irrelevant data managed by a plurality of systems. Provide data structure etc.

  The data structure according to the present invention is a universal table for managing data collected from a plurality of tables having different definitions by item IDs without defining item names, and for identifying data from the item IDs. From the mask table in which mask information for enabling extraction of only the data stored in the necessary item ID is stored for each necessary item ID, and all data in the universal table. When data is extracted according to the mask information stored in units of rows of the mask table, and the same extracted data is extracted from the extracted data in a plurality of rows, the extracted plurality of the same data is And a master table managed as one piece of data.

  As described above, in the data structure according to the present invention, variously defined data are managed by a single universal table in which item names are not defined. A mask table in which mask information (for example, mask information of “0” or “1”) for enabling extraction of only the data stored in the item is stored for each row is prepared. Since all the data is provided with a master table that extracts data according to the mask information stored in units of each row of the mask table and manages the extracted same data as one data, the master table is all managed with a universal table Only the information that becomes the master of the data is centralized and managed, and the processing such as search can be performed very efficiently. To.

  In addition, it is possible to search the universal table in detail using the data in the master table as a search condition as necessary, and the required data can be efficiently extracted even with extremely large amounts of big data. There is an effect.

  The data structure according to the present invention includes a pattern ID assigned to the data of the master table, a local ID assigned to the data of the universal table, and an ID for managing an integrated ID obtained by integrating the pattern ID and the local ID. A management table is provided, and data having a common pattern ID is managed as a common integrated ID, while data having a common local ID is managed as a common integrated ID.

  As described above, in the data structure according to the present invention, the data having the common local ID is managed as the common integrated ID while the data having the common pattern ID is managed as the common integrated ID. Thus, there is an effect that the name identification process can be efficiently performed by managing only the ID.

  The data structure according to the present invention includes an item definition table for managing item names of item IDs corresponding to the mask information of the mask table.

  As described above, in the data structure according to the present invention, since the item definition table for managing the item name of the item ID corresponding to the mask information of the mask table is provided, all data is managed only by the item ID as in the universal table. Even in such a case, an administrator or a user can refer to the item name of the item definition table and recognize what kind of data it is.

  In the data structure according to the present invention, the universal table has information related to the implementation of the data contents stored in the plurality of tables having the different definitions as a fixed item for each data.

  As described above, in the data structure according to the present invention, the universal table is information related to the implementation of data contents stored in a plurality of tables having different definitions (for example, one or more of 5W1H information). Information) as a fixed item for each data, there is an effect that it can be used as important information in terms of operation.

  In the data structure according to the present invention, the master table is information that the universal table does not have, and additionally stores information related to the extracted data.

  As described above, in the data structure according to the present invention, the master table is information that the universal table does not have, and additionally stores information related to the extracted data. There is an effect that information can be provided with additional elements.

  A data generation device according to the present invention is a data generation device that generates the data structure, inputs data from a plurality of tables having different definitions, and expands the input data that has been input into the universal table Universal table generation means, extraction means for extracting necessary data from all data of the universal table according to the mask information of the mask table, and a plurality of lines of the same extracted data among the extracted extracted data are extracted And a master table generating means for storing the extracted plurality of the same data as one data in the master table.

  As described above, in the data generation device according to the present invention, the universal table is generated, the necessary data is extracted according to the mask information of the mask table, and a plurality of identical data among the extracted data is mastered as one data. Since it is stored in the table, only the information that becomes the master of all data managed by the universal table can be collected and managed in the master table, and the processing such as search can be performed very efficiently Play.

  In addition, it is possible to search the universal table in detail using the data in the master table as a search condition as necessary, and the required data can be efficiently extracted even with extremely large amounts of big data. There is an effect.

  The data generation device according to the present invention manages a pattern ID assigned to the data of the master table, a local ID assigned to the data of the universal table, and an integrated ID obtained by integrating the pattern ID and the local ID. ID management table generating means for generating an ID management table, and adding the common integrated ID to the data having the common pattern ID, and adding the common integrated ID to the data having the common local ID Integrated ID management means for managing IDs.

  As described above, in the data generation device according to the present invention, the data having the common local ID is managed as the common integrated ID while the data having the common pattern ID is managed as the common integrated ID. Thus, there is an effect that the name identification process can be efficiently performed by managing only the ID.

  The data generation device according to the present invention outputs an output corresponding item name of data extracted from the universal table based on an item definition table that manages an item name of an item ID corresponding to the mask information of the mask table Means are provided.

  As described above, in the data generation device according to the present invention, based on the item definition table that manages the item name of the item ID corresponding to the mask information of the mask table, the corresponding item name of the data extracted from the universal table is obtained. In order to output, even if all data is managed and stored only by item ID as in the universal table, how the administrator or user refers to the item name in the item definition table It is possible to recognize whether the data is correct.

  In the data generation device according to the present invention, the universal table generation means adds information related to the implementation of the data contents stored in the plurality of tables having the different definitions as a fixed item for each data. Is.

  As described above, in the data generation device according to the present invention, the universal table is information related to the implementation of data contents stored in a plurality of tables having different definitions (for example, one or more of 5W1H information) By having each piece of data as a fixed item for each data, there is an effect that it can be used as important information in terms of operation.

  The data generation device according to the present invention comprises output means for outputting additional information related to the extracted data that is stored in the master table and is not included in the universal table. is there.

  As described above, in the data generation device according to the present invention, the master table is information that the universal table does not have, and additionally stores information related to the extracted data. There is an effect that the side can provide information with additional elements.

  The data search device according to the present invention is a data search device using the data structure, based on input means for inputting data stored in the master table as search condition data, and the input search condition data And extracting means for extracting data that matches the search condition data from the universal table, and output means for outputting the extracted data.

  As described above, in the data search device according to the present invention, the data stored in the master table is input as search condition data, and based on the input search condition data, the data that matches the search condition data is Since it is extracted from the universal table and output, even if a huge amount of big data is found, there is an effect that a search condition that always hits is set and an efficient search can be performed.

It is a hardware block diagram of the data generation apparatus which concerns on 1st Embodiment. It is a functional block diagram which shows the structure of the data generation apparatus which concerns on 1st Embodiment. It is a figure which shows the data structure of the universal table which concerns on 1st Embodiment, a mask table, and a master table. It is a figure which shows the data structure of the pattern map and ID management table which concern on 1st Embodiment. It is a figure which shows an example of a universal table and a mask table. It is a figure which shows an example of the master table produced | generated using the data of FIG. It is a figure which shows the search of the data using master data. It is a figure which shows the 1st process in name identification. It is a figure which shows the 2nd process in name identification. It is a figure which shows the 3rd process in name identification. It is a figure which shows the 4th process in name identification. It is a figure which shows the table structure which concerns on 1st Embodiment. It is a flowchart which shows operation | movement of the data generation apparatus which concerns on 1st Embodiment. It is a functional block diagram which shows the structure of the data search device which concerns on 2nd Embodiment. It is a figure which shows the table structure of the data used for the data search device which concerns on 2nd Embodiment. It is a functional block diagram which shows the structure of the data generation apparatus which concerns on 3rd Embodiment. It is a figure which shows the table structure which concerns on 3rd Embodiment. It is a figure which shows the process in the case of combining an item definition table and a master table, and displaying an item name. It is a figure which shows the utilization method of an item definition table. It is a functional block diagram which shows the structure of the data generation apparatus which concerns on 4th Embodiment. It is a figure which shows the table structure which concerns on 4th Embodiment. It is a figure which shows an example of the information of the universal table containing a fixed item. It is a functional block diagram which shows the structure of the data generation apparatus which concerns on 5th Embodiment. It is a figure which shows the table structure which concerns on 5th Embodiment. It is a figure which shows the specific example of additional information.

  Embodiments of the present invention will be described below. Throughout the present embodiment, the same elements are denoted by the same reference numerals.

(First embodiment of the present invention)
The data structure and data generation apparatus according to this embodiment will be described with reference to FIGS. The data structure according to the present embodiment is a large amount of big data by centrally managing a plurality of data individually defined in different systems in one table and controlling access to each data by an index. Can manage data efficiently with a small capacity.

  FIG. 1 is a hardware configuration diagram of a data generation apparatus according to the present embodiment. In FIG. 1, the data generation device 1 includes a CPU 11, a RAM 12, a ROM 13, a hard disk (HD) 14, a communication I / F 15, and an input / output I / F 16. The ROM 13 and the HD 14 store an operating system and various programs, which are read into the RAM 12 as necessary, and each program is executed by the CPU 11. The generated data and its original data are stored in the HD 14 and can be accessed by a program as necessary.

  The communication I / F 15 is an interface for performing communication between apparatuses. The input / output I / F 16 is an interface for receiving input from an input device such as a touch panel, a keyboard, and a mouse, and outputting data to a printer, a screen, and the like. The input / output I / F 16 can be connected to a drive corresponding to a removable disk such as a magneto-optical disk, a floppy disk, a CD-R, a DVD-R, or the like, if necessary. Each processing unit is connected via a bus and exchanges information. The hardware configuration is merely an example, and can be changed as necessary.

  FIG. 2 is a functional block diagram showing the configuration of the data generation apparatus according to this embodiment. The data generation device 1 includes a data input unit 21 that inputs data managed by a plurality of different systems (10a to 10d). The systems 10a to 10d are, for example, medical management systems, and the system 10a is used in different and independent systems such as the system data of the medical institution A and the system 10b is the system data of the medical institution B. System data that has been defined.

  In addition, the data generation device 1 includes a universal table generation unit 22 that stores data input by the data input unit 21 in the universal table storage unit 23 and data stored in the universal table storage unit 23 in advance by an administrator. The data extraction unit 25 that extracts based on the mask information of the created mask table storage unit 24 and the same data among the extracted data are collected as one data to generate master data, and the master table storage unit 27 And a master table generating unit 26 for storing the data. Here, the universal table, the mask table, and the master table will be described.

  FIG. 3 is a diagram showing a data structure of the universal table, the mask table, and the master table according to the present embodiment. 3A shows the structure of the universal table, FIG. 3B shows the structure of the mask table, and FIG. 3C shows the structure of the master table. Usually, the field name defines a name indicating the type of data (for example, name, age, date of birth, etc.) to be entered in the field. In the case of the universal table in FIG. The field name is managed by the SLOT number so that it can be entered. That is, for example, in the case of data used in the system 10a, “local ID” is stored in SLOT1, “sex” is stored in SLOT2, “name” is stored in SLOT3, and “address” is stored in SLOT4. In the case of the stored data, the universal table is formed in such a manner that “local ID” is stored in SLOT1, “name” is stored in SLOT2, “telephone number” is stored in SLOT3, and “zip code” is stored in SLOT4.

  The mask table of FIG. 3B has only information “0” or “1” as data, and “1” is set in a field necessary for identifying the data. By multiplying with each record, only necessary information can be extracted. For example, in the second record (PP-ID = 2) in FIG. 3B, “1” is set in SLOT1, 5 and 6, and “0” is set as mask data in the other records. That is, by multiplying the data of the universal table by the mask data of PP-ID = 2 in FIG. 3B, only the information of SLOT1, 5, and 6 can be extracted from the universal table. However, at this time, since all data is randomly stored in the universal table, the data extracted by multiplying PP-ID = 2 in FIG. 3B is not all meaningful data. That is, although data necessary for identifying the data is always included by the mask information of PP-ID = 2, data irrelevant (data not required by the user) is also included.

  The master table in FIG. 3C stores information as master data by combining the same data into one data among the data extracted by the mask information of the mask table. For example, when a plurality of data such as SLOT1 = 11, SLOT5 = patient Taro, and SLOT6 = 1976/10/10 are extracted by the mask information of PP-ID = 2, those data are regarded as one data. One ID (hereinafter referred to as PTN-ID) is assigned. More specific data structures of the universal table, mask table, and master table will be described later.

  In this way, the master data of the universal table is generated by combining the discriminating information extracted from the mask information of the mask table into one data among the data stored in the universal table. In other words, in the search when handling a huge amount of data such as tens of millions, setting of search conditions is very important, but by creating master data as in this embodiment, Master data can be provided as a search condition that always makes a hit, and work efficiency can be greatly improved.

  Returning to FIG. 2, when the master table is generated by the master table generation unit 26, LOCAL_ID (hereinafter referred to as L-ID) which is a unique identification code used in the original systems 10a to 10d. Is extracted from the universal table, the PTN_ID for each master data given at the time of generating the master table is extracted from the master table, and a pattern map generated by adding them together is stored in the pattern map storage unit 29. And an integrated ID management unit 30 that creates an ID management table for managing IDs from the pattern map and stores the ID management table in the ID management table storage unit 31.

  FIG. 4 is a diagram showing the data structure of the pattern map and ID management table according to the present embodiment. 4A shows a pattern map, and FIG. 4B shows an ID management table. L-ID shown in FIG. 4 (A) is LOCAL_ID, which is an ID having uniqueness used in the original system. For example, in the case of a medical management system, it is a unique ID that connects a medical institution code (code for each medical institution) and an examination ticket code (code for each patient). The PTN-ID is a unique ID assigned when the master table is generated. In the pattern map, a combination of L-ID and PTN-ID is stored.

  Here, the L-ID has uniqueness on the system, but, for example, in a medical management system, the relationship between the L-ID and the patient is not always one-to-one. That is, when the same patient is examined at different medical institutions, the patient has two L-IDs. On the other hand, PTN-ID also has uniqueness in the system. For example, in a medical management system, one master data (for example, name, zip code, gender, date of birth) is one patient. It does not always indicate. That is, a person whose sex has changed after marriage or a person whose postal code has changed due to moving or the like has a plurality of different master data even if the person is the same person.

  In order to integrate such data into the same person (hereinafter referred to as name identification), the ID management table in FIG. 4B is generated. In the ID management table of FIG. 4B, a common integrated ID is assigned to data having the same pattern ID, and a common integrated ID is assigned to data having the same PTN-ID. That is, there is always a one-to-one relationship between the integrated ID and the person (thing depending on the type of system) identified by the integrated ID. Therefore, all the data extracted with the same integrated ID can be extracted as information related to the same person (or thing). Details of the specific process for assigning the integrated ID will be described later.

  Next, a process until the master table is generated and a method for using the process will be described with a specific example. FIG. 5 is a diagram illustrating an example of the universal table and the mask table. FIG. 5A shows an example of a universal table, and FIG. 5B shows an example of a mask table. In the universal table, patient information is stored in lines 1 to 3, drug information is stored in lines 4 to 6, and examination information is stored in lines 7 to 9. In PP-ID = 1 of the mask table, “1” is set for SLOT 2 to 6. This is set in advance by the administrator in order to create master data of patient information, and includes information on SLOT 2 to 6 (specifically, account, name, date of birth, gender, zip code). Therefore, master data of identifiable patient information can be created. Similarly, “1” is set to SLOT2 to 5 in PP-ID = 2, and “1” is set to SLOT2, 3, and 5-7 in PP-ID = 3.

  FIG. 6 is a diagram illustrating an example of a master table generated using the data of FIG. By multiplying the universal table of FIG. 5A and the mask table of FIG. 5B, a plurality of identical data is handled as one data (one PTN-ID). In FIG. 6, data with PP-ID = 1 is data created with the mask information with PP-ID = 1 in FIG. Here, “patient Hanako” in the third row from the top completely matches “patient Hanako” in the second row from the top, and thus is not included in the master data. In such a case, the PTN-ID may not be assigned as shown in FIG. 6, or the data itself may be deleted. Similarly, “patient Hanako” on the 12th line, “eye drops B” on the 15th line, “patient Hanako” on the 21st line, and “urinalysis (protein)” on the 27th line have the same data. No PTN-ID is assigned and there is no data. Thus, the master table is master data in which the minimum data extracted in the minimum necessary fields is collected.

  By creating the master data as described above, it is possible to efficiently search for data. FIG. 7 is a diagram showing data search using master data. Here, for example, it is assumed that the user wants to retrieve drug information data from the universal table. Since the mask information of the data related to the drug information is PP-ID = 2 (details will be described later in another embodiment, by using an item definition table in which item names are defined in association with the mask table, the drug information is used. The user can grasp that the mask information for acquiring the information is PP-ID = 2), and the master data of FIG. 7A extracted with the mask information of PP-ID = 2 is searched. Subject to choice. The user designates a search condition from these options as shown in FIG. 7B, and hits the line in FIG. 7C. Necessary data can be obtained from the universal table by specifying the integrated ID from the PTN-ID of the hit line and searching the data of the integrated ID from the universal table. The item name in parentheses for each SLOT in FIG. 7B is an item name defined in the item definition table that defines the item name in association with the mask table.

  Next, specific processing for name identification will be described with four examples. FIG. 8 is a diagram illustrating a first process in name identification. Here, it is assumed that new data is added to the third item in the universal table (FIG. 8A). Since the added data C is different from other patient information, a new PTN-ID (= 300) is assigned (FIG. 8B). Data C is newly added to the pattern map in order to manage the L-ID (here, L-ID is a connection between HOSP_CD and LOCAL_CD) and PTN-ID (FIG. 8C). Along with the addition to the pattern map, a new integrated ID is assigned and data C is added to the ID management table (FIG. 8D). Since the data A, B, and C in the ID management table are all different PTN-IDs and different L-IDs, they are managed as different person information.

  FIG. 9 is a diagram illustrating a second process in name identification. Here, it is assumed that new data is added to the fourth case in the universal table (FIG. 9A). Since the added data D completely coincides with the patient information of data A, a new PTN-ID is not assigned (FIG. 9B). In order to manage the L-ID and the PTN-ID, data C is newly added to the pattern map (FIG. 9C). Since data A and data D have the same patient information, they are added to the pattern map as the same PTN-ID = 100, and are separately managed. Along with the addition to the pattern map, a new integrated ID is assigned and data D is added to the ID management table (FIG. 9D). At this time, since data A having the same PTN-ID (= 100) already exists, the integrated ID of data D is re-assigned to the same ID as data A (FIG. 9E). Although the data A and D in the ID management table are assigned different L-IDs but have the same integrated ID, they are managed as information on the same person.

  FIG. 10 is a diagram illustrating a third process in name identification. Here, it is assumed that new data is added to the fifth case of the universal table (FIG. 10A). Since the added data E is different from other patient information, a new PTN-ID (= 400) is assigned (FIG. 10B). In order to manage the L-ID and the PTN-ID, data E is newly added to the pattern map (FIG. 10C). Along with the addition to the pattern map, a new integrated ID is assigned and data E is added to the ID management table (FIG. 10D). At this time, since the data B having the same P-ID (= 4199-999010) already exists, the integrated ID of the data E is changed to the same ID as the data B (FIG. 10E). Although the data B and E in the ID management table are assigned different PTN-IDs but have the same integrated ID, they are managed as information on the same person.

  FIG. 11 is a diagram illustrating a fourth process in name identification. Here, it is assumed that new data is added to the third item in the universal table (FIG. 11A). Since the added data C completely matches the patient information of the data B, a new PTN-ID is not assigned (FIG. 11B). In order to manage the L-ID and the PTN-ID, data C is newly added to the pattern map (FIG. 11C). Since data B and data C have the same patient information, the same PTN-ID = 400. Since data L and data C have the same L-ID, they are added to the pattern map as the same L-ID. Is done. Along with the addition to the pattern map, a new integrated ID is assigned and data C is added to the ID management table (FIG. 11D). At this time, since data B having the same PTN-ID (= 400) already exists, the integrated ID of data C is re-assigned to the same ID as data B (FIG. 11E). Next, when the L-ID matches are found, since the data A and the data C are the same L-ID (FIG. 11F), the integrated ID of the data C is changed to the integrated ID of the data A (FIG. 11). 11 (G)). Furthermore, when the PTN-ID matches are found, the integrated IDs are different even though the PTN-IDs of the data B and the data C match (FIG. 11 (H)). The integrated ID is reacquired according to the integrated ID of data C (FIG. 11 (I)).

  Here, when the integrated ID is re-acquired, the young code is preferentially performed. In other words, the reason why the integrated ID of data B is re-acquired in accordance with the integrated ID of data C in FIG. However, it is not always necessary to give priority to the young code, and any one code based on a predetermined rule may be given priority.

Next, the data structure according to this embodiment will be described. FIG. 12 is a diagram showing a table configuration according to the present embodiment. In this embodiment, a plurality of general systems 41 are converted into a CSV 42 format and developed in each SLOT of the universal table 43. At this time, the expansion process is performed based on the 1 ^ST MAP 44 indicating which field in the general system 41 corresponds to which SLOT. Based on the universal table 43, a master table 46 is automatically generated based on a mask table 45 created in advance by an administrator. A pattern map 47 is generated from the L-ID of the universal table 43 and the PTN-ID of the master table 46. Based on this pattern map 47, an ID management table 48 for managing an integrated ID when name identification processing is performed. Generated.

Next, the operation of the data generation apparatus according to this embodiment will be described. FIG. 13 is a flowchart showing the operation of the data generation apparatus according to this embodiment. First, the data input unit 21 inputs system data from a plurality of systems 10a to 10d (S1), and the universal table generation unit 22 converts the input system data into CSV, and the universal table based on 1 ^ST MAP 44 43 is generated and stored in the universal table storage unit 23 (S2). The data extraction unit 25 reads the mask table 45 created in advance from the mask table storage unit 24, and extracts information from the universal table 43 according to the mask information (S3). The master table generation unit 26 generates a master table 46 so that the same data among the extracted information is managed by the same PTN-CD, and stores the master table 46 in the master table storage unit 27 (S4).

  The pattern map generating unit 28 combines the L-ID of the universal table 43 stored in the universal table storage unit 24 with the PTN-ID of the master table 46 stored in the master table storage unit 27 in association with each other. Is generated and stored in the pattern map storage unit 29 (S5). The integrated ID management unit 30 performs the name identification process described above based on the pattern map 47, creates an ID management table 48 to which an integrated ID is assigned, stores the ID management table 48 in the ID management table storage unit 31 (S6), and performs processing Exit.

  As described above, in the data generation device and the data structure according to the present embodiment, variously defined data are managed by a single universal table in which item names are not defined, and are at least necessary for identifying the data. For each item, prepare a mask table that stores the mask information for each row so that only the data stored in the item can be extracted. Since the master table includes a master table that extracts data according to the stored mask information and manages the extracted same data as one data, the master table has only information that becomes a master of all data managed by the universal table. It is managed in a centralized manner, and processing such as search can be performed very efficiently.

  Moreover, it becomes possible to search the universal table in detail using the data in the master table as a search condition as required, and the required data can be efficiently extracted even with extremely large amounts of big data.

  Further, by managing data having a common PTN-ID as a common integrated ID and managing data having a common L-ID as a common integrated ID, name identification processing can be efficiently performed by managing only the ID. be able to.

(Second embodiment of the present invention)
A data search apparatus according to this embodiment will be described with reference to FIGS. 14 and 15. The data search device according to the present embodiment searches for predetermined data from the data structure and generated data in the first embodiment.
In addition, in this embodiment, the description which overlaps with the said 1st Embodiment is abbreviate | omitted.

  FIG. 14 is a functional block diagram showing the configuration of the data search apparatus according to this embodiment. The data search device 50 according to the present embodiment includes a condition information input unit 52 that inputs search condition information 51 based on master information stored in the master table 46 of the data generation device 1 according to the first embodiment, and an input The integrated ID of the data that matches the retrieved search condition is extracted from the ID management table 48, the retrieved unit 53 that extracts the extracted integrated ID data from the universal table 43, and the result extracted by the retrieved unit 53 is displayed on a display, for example. And an output unit 54 for outputting.

  The master table 46 stores master information (minimum necessary information) of the universal table 43 to be searched. By using this master information as search condition information, necessary information can be obtained from a large amount of data. It becomes easy to narrow down. Then, by determining the integrated ID from the search condition based on the master information and extracting the data corresponding to the integrated ID from the universal table 43, more detailed data can be efficiently searched from the narrowed down data.

  FIG. 15 is a diagram illustrating a data table configuration used in the data search apparatus according to the present embodiment. Based on the set search condition information, the integrated ID is extracted from the ID management table 48, and the data of the universal table 43 combined by the extracted integrated ID is output as the search result information.

  As described above, in the data search device according to the present embodiment, data stored in the master table is input as search condition data, and data matching the search condition data is input based on the input search condition data. Since the universal table is extracted and output, even if a large amount of big data is found, a search condition that always hits is set and an efficient search can be performed.

(Third embodiment of the present invention)
The data structure and data generation apparatus according to this embodiment will be described with reference to FIGS. The data generation apparatus according to the present embodiment includes an item definition table that manages the SLOT name of the SLOT corresponding to the mask information of the mask table 45.
In addition, in this embodiment, the description which overlaps with each said embodiment is abbreviate | omitted.

  FIG. 16 is a functional block diagram illustrating a configuration of the data generation apparatus according to the present embodiment, and FIG. 17 is a diagram illustrating a table configuration according to the present embodiment. 16 is different from FIG. 14 in the second embodiment in that an item definition table storage unit 32 that stores an item definition table associated with the mask table storage unit 24 is provided. 17 is different from FIG. 15 in the second embodiment in that an item definition table 49 corresponding to the mask table 45 is provided.

  The item definition table 49 is a table for associating meaningful information with the SLOT number of the universal table 43, and is associated with the mask information of the mask table 45. The item name management method using the item definition table 49 will be specifically described below. FIG. 18 is a diagram illustrating a process when the item definition table and the master table are combined to display the item name. FIG. 18A shows an example of an item definition table, and FIG. 18B shows an example of a general-purpose screen when extracting master information. As shown in FIG. 18A, item names are defined for each SLOT in accordance with the PP-ID of the mask table 45. Here, PP-ID = 1 indicates patient information, PP-ID = 2 indicates drug information, and PP-ID = 3 indicates examination information. When searching for patient information, FIG. 18B shows. Thus, by displaying the information of PP-ID = 1 in the column name, the user can recognize the meaning of the SLOT contents of the patient information in the master table, and the usability is improved.

  FIG. 19 is a diagram illustrating a method of using the item definition table. FIG. 19A shows a general-purpose screen on which item names in the item definition table 49 are displayed as column names, and FIG. 19B shows setting of search conditions and search results. As shown in FIG. 19, the user can extract necessary information by setting a search condition in consideration of the item names stored in the item definition table. That is, since the user can be aware of the item name, it is possible to efficiently set an appropriate extraction condition for extracting necessary information from a unified table in which various pieces of information are mixed.

  As described above, in the data generation device according to the embodiment of the present invention, since the item definition table for managing the item name of the item ID corresponding to the mask information of the mask table is provided, all data is stored in the item ID as in the universal table. Even if it is managed and stored only by the administrator, users and users can refer to the item name in the item definition table and recognize what kind of data it is, Operability and work efficiency can be improved.

(Fourth embodiment of the present invention)
The data structure and data generation apparatus according to this embodiment will be described with reference to FIGS. In the data structure according to the present embodiment, the universal table has, for example, 5W1H information for each data as a fixed item.
In addition, in this embodiment, the description which overlaps with each said embodiment is abbreviate | omitted.

  FIG. 20 is a functional block diagram showing the configuration of the data generation apparatus according to this embodiment, and FIG. 21 is a diagram showing the table configuration according to this embodiment. In FIG. 20, the data input unit 21 is different from FIG. 16 in the third embodiment in that the 5W1H information (when, where, who ( When one or a plurality of information is input to (who), what, why, how) and a universal table is generated, the universal table generator 22 generates one or more of 5W1H information for each data. It is to add information as a fixed item. 21 is different from FIG. 17 in the third embodiment in that, when the CSV 42 is generated from the general system 41, 5W1H information for each data is added to the CSV 42 as a fixed item. .

  FIG. 22 is a diagram illustrating an example of information in the universal table including fixed items. As shown in FIG. 22A, the information of “date and time” to “pattern” stored in the general system 40 is extracted as 5W1H information, and as shown in FIG. Append to Such a fixed item is information that is always fixedly added to the universal table regardless of each SLOT item. By holding such information, the range of search processing can be expanded, and what If a problem occurs, it can be used to investigate the cause.

  More specifically, for example, “implementation date / time”, “place (place without an exclusive element)”, “target person (person who carries out the contents of information stored in the universal table)”, “implementer” (Person who executes the contents of the information stored in the universal table) ”and“ Manager (person who instructed the execution of the contents of the information stored in the universal table) ”are added as fixed items. The contents of the information are information developed in each SLOT, which can be called action information. That is, if the fixed item such as 5W1H information is the same (for example, when the same implementation action is repeated for several days), the actual action (for example, a specific date) is designated. It is possible to search for overwriting or deleting data.

  Thus, in the data structure according to the present embodiment, the universal table has one or a plurality of pieces of information out of 5W1H for each piece of data as a fixed item, so that it can be used as important information in terms of operation.

(Fifth embodiment of the present invention)
The data structure and data generation apparatus according to this embodiment will be described with reference to FIGS. The data structure according to the present embodiment is obtained by adding additional information to the master information of the master table as necessary.
In addition, in this embodiment, the description which overlaps with each said embodiment is abbreviate | omitted.

  FIG. 23 is a functional block diagram showing a configuration of the data generation apparatus according to the present embodiment, and FIG. 24 is a diagram showing a table configuration according to the present embodiment. 23 is different from FIG. 2 in the first embodiment in that additional information other than the master information generated by the master table generating unit 26 is added to the master table 46 stored in the master table storing unit 27. That is. 24 is different from FIG. 12 in the first embodiment in that additional information is added to the master table 46 and the information can be used as search information.

  FIG. 25 is a diagram illustrating a specific example of additional information. FIG. 25A shows data provided from the general system 41, and shows an example of information stored in the universal table 43. FIG. 25B shows the data provided in FIG. An example of a table storing information is shown, and FIG. 25C shows information items finally provided. The table storing known information in FIG. 25B corresponds to additional information, and the contents are information unique to one piece of master information. For example, there are information such as “medicinal effect”, “inhibition”, and “sales source” as information that is already generally known and can be easily acquired with respect to the stomach medicine A, and these information are managed in the general system 41. Even if it is not, the administrator can add the additional information separately so that the provision to the user can be further enhanced.

  The final management items in each table when additional information is added are as shown in FIG. 25C, and the information stored in the universal table 43 is “type” as shown in FIG. “Medicine name”, “medicine effect 1”, “medicine effect 2”, “prescription amount 1”, “prescription amount 2”, “prescription amount 3”. Since the information stored in the master table 46 is a minimum pattern indicating uniqueness, there are four items of “type”, “drug name”, “medicine effect 1”, and “medicine effect 2”. The table finally provided as search information is obtained by adding additional information items to the universal table 43, and has all items from “type” to “sales” in FIG. It will be what you are doing. At this time, the item name of each item may be defined by the item definition table 49.

  As described above, in the data structure according to the present embodiment, the master table is information that the universal table does not have and additionally stores information related to the extracted data. Information can be provided with specific elements.

  In each of the above embodiments, the process of creating the master table has been described on the assumption that the data stored in the universal table 43 is completely prepared. However, in the original general system 41, the necessary data is There are many things that are lacking. In that case, the pattern map 47 indicating the relationship between the L-ID and the PTN-ID may be manually generated. By doing so, an appropriate master table 46 can be generated.

  Moreover, each component in each said embodiment can be combined arbitrarily, and the system construction according to a use environment is possible.

1 data generation device 10a to 10d system 11 CPU
12 RAM
13 ROM
14 HD
15 Communication I / F
16 Input / output I / F
21 Data Input Unit 22 Universal Table Generation Unit 23 Universal Table Storage Unit 24 Mask Table Storage Unit 25 Data Extraction Unit 26 Master Table Generation Unit 27 Master Table Storage Unit 28 Pattern Map Generation Unit 29 Pattern Map Storage Unit 30 Integrated ID Management Unit 31 ID Management table storage unit 32 Item definition table storage unit 41 General system 42 CSV
43 Universal table 44 1 ^ST MAP
45 Mask Table 46 Master Table 47 Pattern Map 48 ID Management Table 49 Item Definition Table 50 Data Retrieval Device 51 Search Condition Information 52 Condition Information Input Unit 53 Search Unit 54 Output Unit

Claims (13)

A universal table that manages data collected from multiple tables with different definitions by item ID without defining item names;
For each necessary item ID that is at least necessary for identifying data from among the item IDs, mask information for enabling extraction of only data stored in the necessary item ID is stored for each row. Mask table,
When data is extracted from all data of the universal table according to mask information stored in units of rows of the mask table, and when the same extracted data is extracted from the extracted data, a plurality of rows are extracted. And a master table for managing a plurality of identical data as one data. The data structure of claim 1, wherein
An ID management table for managing a pattern ID given to the data of the master table, a local ID given to the data of the universal table, and an integrated ID obtained by integrating the pattern ID and the local ID;
A data structure characterized by managing data having a common local ID as a common integrated ID while managing data having a common pattern ID as the common integrated ID. The data structure according to claim 1 or 2,
A data structure comprising an item definition table for managing item names of item IDs corresponding to mask information of the mask table. The data structure according to any one of claims 1 to 3,
A data structure, wherein the universal table has information related to the implementation of data contents stored in the plurality of tables having the different definitions as fixed items for each data. The data structure according to any one of claims 1 to 4,
A data structure, wherein the master table additionally stores information related to the extracted data, which is information that the universal table does not have. A data generation device for generating the data structure according to claim 1,
Universal table generation means for inputting data from a plurality of tables having different definitions and expanding the input data into the universal table;
Extracting means for extracting necessary data from all data of the universal table according to the mask information of the mask table;
Master table generating means for storing the extracted plurality of the same data as one data in the master table when the same extracted data is extracted from a plurality of lines of the extracted data. A data generation device. The data generation device according to claim 6, wherein a pattern ID assigned to the data of the master table, a local ID assigned to the data of the universal table, and an integrated ID obtained by integrating the pattern ID and the local ID are managed. ID management table generation means for generating an ID management table to be performed;
An integrated ID management unit that assigns the common integrated ID to the data having the common pattern ID, and assigns the common integrated ID to the data having the common local ID to manage the integrated ID; A data generation device characterized by the above. In the data generation device according to claim 6 or 7,
Data comprising output means for outputting a corresponding item name of data extracted from the universal table based on an item definition table for managing an item name of an item ID corresponding to the mask information of the mask table Generator. The data generation device according to any one of claims 6 to 8,
The universal table generation means adds information related to the implementation of data contents stored in a plurality of tables having different definitions as a fixed item for each data. The data generation device according to any one of claims 6 to 9,
A data generation apparatus comprising: output means for outputting additional information related to the extracted data that is stored in the master table and is not included in the universal table. A data search device using the data structure according to claim 1,
Input means for inputting the data stored in the master table as search condition data;
Based on the input search condition data, extraction means for extracting data that matches the search condition data from the universal table;
A data search apparatus comprising: output means for outputting the extracted data. A data generation method for generating the data structure according to claim 1,
Computer
A universal table generation step of inputting data from a plurality of tables having different definitions and expanding the input data into the universal table;
An extraction step of extracting necessary data from all data of the universal table according to the mask information of the mask table;
A master table generating step of storing the extracted plurality of the same data as one data in the master table when the same extracted data is extracted from a plurality of lines of the extracted data. Characteristic data generation method. A data generation program for causing a computer to function to generate the data structure according to claim 1,
A universal table generating means for inputting data from a plurality of tables having different definitions and expanding the input data into the universal table;
Extraction means for extracting necessary data from all data of the universal table according to the mask information of the mask table;
When a plurality of the same extracted data is extracted from the extracted data, the computer is caused to function as a master table generating means for storing the extracted plurality of the same data as one data in the master table. A data generation program characterized by

Images