External schema generation device and external schema generation method
The external schema generation device and method address the challenge of developers lacking business knowledge by automatically mapping database columns to business goals, creating an intuitive schema that aligns with business data structures.
Patent Information
- Application Number
- JP2023078922
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-11
- Publication Date
- 2024-11-21
AI Technical Summary
Developers without business knowledge face challenges in mapping data used in applications to data in databases, requiring them to learn business knowledge, which is inconvenient and time-consuming.
An external schema generation device and method that identifies corresponding columns in a database based on similarity with target information, generating an external schema that reflects business goals, using metadata extraction, lineage management, and adaptive mapping to create views that align with business data structures.
Facilitates the creation of an external schema that is easy for application developers to use, automatically reflecting business content and reducing the need for developers to acquire business knowledge.
Smart Images

Figure 2024162920000001_ABST
Abstract
Description
[Technical field]
[0001] The present disclosure relates to an external schema generating device and an external schema generating method. [Background technology]
[0002] Databases generated in business are used in a variety of applications.
[0003] Although business personnel have knowledge of the business, they do not have knowledge of databases or programming knowledge for designing applications, so application development is usually outsourced to a developer other than the business personnel.
[0004] However, developers typically lack business knowledge and therefore may have difficulty mapping between the data used in the application and the data contained in the database, and in some cases may need to learn business knowledge.
[0005] In response to this, Patent Document 1 discloses a technology that makes it easy to create applications by providing users with various user interfaces and automatically mapping data used in the application to data in a database in the backend in response to operations such as drag and drop. [Prior art documents] [Patent documents]
[0006] [Patent Document 1] US Patent No. 2017 / 0115968 Summary of the Invention [Problem to be solved by the invention]
[0007] However, the technology described in Patent Document 1 is intended to assist business personnel in creating applications, and requires that the data specified by the user through operations such as drag-and-drop be a perfect match with the data in the database, making it difficult to use for developers who do not have knowledge of the business.
[0008] An object of the present disclosure is to provide an external schema generation device and an external schema generation method that are capable of providing an external schema that is easy to use for application developers. [Means for solving the problem]
[0009] An external schema generation device according to one aspect of the present disclosure is an external schema generation device that generates an external schema of a database generated by a business, and has an identification unit that identifies, based on the similarity between each column of data included in the database and each item of goal information related to an index value indicating a goal of the business, a column corresponding to each item as a corresponding column, and a generation unit that generates the external schema based on the corresponding column. Effect of the Invention
[0010] According to the present invention, it is possible to provide an external schema that is easy for application developers to use. [Brief description of the drawings]
[0011] [Figure 1] FIG. 2 is a diagram illustrating a hardware configuration of a view generating device according to an embodiment of the present disclosure. [Diagram 2] FIG. 1 is a diagram illustrating a functional configuration of a view generating device according to an embodiment of the present disclosure. [Diagram 3] FIG. 11 is a diagram illustrating an example of data. [Figure 4] FIG. 2 is a diagram illustrating an example of metadata. [Diagram 5] FIG. 13 is a diagram illustrating an example of lineage related information. [Figure 6]FIG. 13 is a diagram illustrating another example of the lineage related information. [Figure 7] FIG. 13 is a diagram illustrating an example of a KPI tree. [Figure 8] FIG. 2 is a diagram showing the tree structure of a KPI tree. [Figure 9] FIG. 11 is a diagram illustrating an example of mapping data. [Figure 10] FIG. 13 is a diagram illustrating an example of a view. [Figure 11] 11 is a flowchart illustrating an example of a metadata extraction process. [Figure 12] 13 is a flowchart illustrating an example of a lineage management process. [Figure 13] FIG. 13 is a diagram illustrating an example of a KPI tree management process. [Figure 14] 11 is a flowchart illustrating an example of an overall process. [Figure 15] 11 is a flowchart illustrating an example of an adaptive mapping process. [Figure 16] 11 is a flowchart illustrating an example of a similarity determination process. [Figure 17] 13 is a flowchart illustrating an example of a view generation process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.
[0013] Fig. 1 is a diagram showing a hardware configuration of a view generating device according to an embodiment of the present disclosure. The view generating device 100 shown in Fig. 1 is an external schema generating device that generates and provides a view, which is an external schema of a database. The view generating device 100 includes a storage device 101, a processor 102, a memory 103, a communication unit 104, an input interface 105, and an output interface 106, which are interconnected via a bus 107. Note that the external schema is not limited to a view, and may be a material view or the like.
[0014] The storage device 101 stores a program that prescribes the operation of the processor 102, and various data used and generated by the program. The processor is, for example, a CPU (Central Processing Unit), which reads the program stored in the storage device 101 into a memory 103 and executes processing according to the program using the memory 103. The communication unit 104 is connected to a client device 110, which is an external device, via a network 120 so as to be able to communicate with each other, and transmits and receives various information to and from the client device 110. The input interface 105 is an input unit that receives various information from a user who uses the view generating device 100. The output interface 106 is an output unit that outputs (for example, displays) various information. The client device 110 is used by a user who uses the view generating device 100.
[0015] Fig. 2 is a diagram showing a functional configuration of the view generating device 100. As shown in Fig. 2, the storage device 101 stores data 200, metadata 300, lineage related information 400, and a KPI tree (Key Performance Indicator Tree) 500. The processor 102 also executes a program (not shown) stored in the storage device 101 to realize a metadata extraction unit 601, a lineage management unit 602, a KPI tree management unit 603, and an adaptive mapping unit 604.
[0016] 3 is a diagram showing an example of the data 200. The data 200 is data that constitutes a database generated by a business operation.
[0017] 3, data 200 includes equipment data 201 related to equipment used in the business, procedure data 202 related to the procedure of the business, operation log data 203 related to an operation log (Operation_log) which is a record of operations performed in the business, and plan data 204 related to a plan for the business. Each of the data 201 to 204 has a table structure having one or more columns 210.
[0018] Fig. 4 is a diagram showing an example of metadata 300. Metadata 300 is data related to data 200, and is generated for each type of data 200, for example. In the example of Fig. 4, metadata 300 includes metadata 301 related to facility data 201 and metadata 302 related to procedure data 202. Metadata 300 indicates, for example, table name 310 which is the name of data 200, and column name 311 which is the name of each column of data 200. Note that metadata 300 may include not only table names and column names, but also any information associated therewith.
[0019] 5 and 6 are diagrams showing an example of the lineage related information 400. The lineage related information is information related to the history of the data 200, and in the example of Fig. 5 and Fig. 6, the lineage related information 400 includes a table access log 410, a column access log 420, job information 430, lineage record information 440, and dataset information 450.
[0020] The table access log 410 is information showing the history of access to the data 200 having a table structure, and includes columns 411 to 414. Column 411 stores an ID, which is identification information for identifying an access. Column 412 stores a data ID (DID), which is identification information for identifying the accessed data. Column 413 stores an access time (ACCESS_TIME), which is the time when the data was accessed. Column 414 stores a job ID (JOB_ID) for identifying the job that accessed the data.
[0021] Column access log 420 is information indicating a history of access to columns in data 200, and includes columns 421 to 425. Column 421 stores an ID for identifying an access. Column 422 stores a table name (TABLE_NAME) that is the name of data 200 having an accessed column. Column 423 stores a column name that is the name of the accessed column. Column 424 stores an access time that is the time when the column was accessed. Column 425 stores a job ID that identifies the job that accessed the column.
[0022] Job information 430 is information about a job that accesses data 200, and includes columns 431 to 433. Column 431 stores a job ID. Column 432 stores job generation information indicating the source that generates the job. In this embodiment, a SQL (Structured Query Language) statement is shown as the job generation source, but the job generation source is not limited to an SQL statement and may be a query or a program, etc. Column 433 stores a job name (JOB_NAME), which is the name of the job.
[0023] Lineage record information 440 is lineage data indicating the history of data 200, and includes columns 441 to 443. Column 441 stores a source ID (SOURCE_ID) that identifies source data, which is data 200 from which data was generated. Column 442 stores a target ID (TARGET_ID) that identifies target data, which is data 200 generated from the source data. Column 443 stores a job ID that identifies the job that generated the target data.
[0024] The data set information 450 is a list of data 200, and includes columns 451 to 453. Column 451 stores the data ID of the data 200. Column 452 stores the table name of the data 200. Column 453 stores a view flag (VIEW_FLG) indicating whether the data 200 is a view or not. In the example of FIG. 8, the view flag indicates "1" if the data 200 is a view, and indicates "0" if the data 200 is not a view.
[0025] FIG. 7 is a diagram showing an example of a KPI tree 500. The KPI tree 500 is information describing elements necessary for achieving the major goals of a business, and is data having a tree structure (hierarchical structure) including nodes corresponding to the elements and links indicating the relationships between the nodes. In this embodiment, each layer of the tree structure is called a rank, and the ranks from the top layer may be called rank 1, rank 2, .... Also, a node connected to a node and above the node may be called a parent node, and a node connected to a node and below the node may be called a child node.
[0026] In the example of FIG. 7, the KPI tree 500 includes a node list 510 that collects node information about each node, and link information 502 that indicates the relationships between the nodes.
[0027] Node list 510 includes columns 511 to 515. Column 511 stores node IDs (NODE_ID) that identify nodes. Column 512 stores node categories (CATEGORY). Column 513 stores KPIs that are index values that indicate goals for nodes. Column 514 stores KPI calculation methods that are methods (METHOD) for calculating KPIs. Column 515 stores usage data that is data (DATA) used to calculate KPIs. Each record in node list 510 is node information related to the node of the node ID stored in column 511 of that record. Each field of node information may also be called an item.
[0028] Link information 520 includes columns 521 to 523. Column 521 stores node IDs. Column 522 stores parent node IDs (PARENT_ID) that identify the parent node of the node of the node ID. Column 523 stores child node IDs (CHILD_ID) that identify the child node of the node of the node ID. If there is no child node, column 523 stores "NULL."
[0029] Fig. 8 is a diagram showing the tree structure of the KPI tree 500. The example of Fig. 8 shows a tree structure having nodes 1 to 4. Node 1 belongs to rank 1, node 2 belongs to rank 2, and nodes 3 and 4 belong to rank 2. In this case, for example, when viewed from node 2, node 1 is the parent node, and nodes 3 and 4 are the child nodes.
[0030] Returning to the explanation of Fig. 2, the metadata extraction unit 601 extracts the metadata 300 from the data 200. The lineage management unit 602 manages the lineage related information 400. For example, the lineage management unit 602 generates the lineage record information 440. The KPI tree management unit 603 manages the KPI tree 500. For example, the KPI tree management unit 603 generates and updates the KPI tree in accordance with input from the user.
[0031] The adaptive mapping unit 604 generates a view, which is an external schema of the data 200. In this embodiment, the adaptive mapping unit 604 generates a view for each node of the KPI tree 500.
[0032] Specifically, the adaptive mapping unit 604 includes an overall management unit 610, a similarity determination unit 611, and a mapping management unit 612. The overall management unit 610 is an identification unit that executes adaptive mapping processing to identify a corresponding column of the data 200 corresponding to each item of each node of the KPI tree 500 as a mapping destination. The similarity determination unit 611 is a calculation unit that calculates the similarity between each item of each node of the KPI tree 500 and the column of the data 200 based on the metadata 300. The similarity is used in the adaptive mapping processing of the overall management unit 610. The mapping management unit 612 is a generation unit that generates a view of each node of the KPI tree 500 based on the result of the adaptive mapping processing by the overall management unit 610. The mapping management unit 612 also generates mapping data indicating a list of views of each node.
[0033] 9 is a diagram showing an example of mapping data. The mapping data 700 shown in FIG.
[0034] Mapping list 710 is information indicating the correspondence between the nodes and views of KPI tree 500, and in the example of Fig. 9, includes columns 711 to 713. Column 711 stores a mapping ID (MAP_ID) that identifies the mapping (association). Column 712 stores a node ID (NODE_ID) of the node. Column 713 stores a view ID (VIEW_ID) that identifies the view corresponding to the node.
[0035] View list 720 is a list of views, and in the example of Fig. 9, includes columns 721 to 722. Column 721 stores the view ID of the view. Column 722 stores the view name (VIEW_NAME), which is the name of the view.
[0036] Fig. 10 is a diagram showing an example of a view. A view 800 shown in Fig. 10 is an example of a view corresponding to a certain node, and includes multiple columns 801 that are corresponding columns corresponding to the items of the node among the columns included in the data 200.
[0037] FIG. 11 is a flowchart for explaining an example of metadata extraction processing by the metadata extraction section 601.
[0038] In the metadata extraction process, the metadata extraction unit 601 first receives access information for accessing the data 200 from the client device 110 (step S101). The access information is, for example, a host name, a port number, a user name, and a password.
[0039] Next, the metadata extraction unit 601 accesses the data 200 based on the received access information, and extracts information necessary for the metadata 300 from the data (step S102). The information necessary for the metadata 300 is, for example, the table name of the data, the columns of each column of the data, and information associated therewith. The necessary information can be extracted, for example, by using an existing data catalog tool.
[0040] The metadata extraction unit 601 stores the extracted information as metadata 300 in the storage device 101 (step S103), and ends the process.
[0041] FIG. 12 is a flowchart illustrating an example of the lineage management process performed by the lineage management unit 602.
[0042] In the lineage management process, the lineage management unit 602 first receives access information for accessing the data 200 from the client device 110 (step S201).
[0043] Next, the lineage management unit 602 accesses the data 200 based on the received access information, and starts a loop process (A) in which the processes of steps S203 to S204 are repeated at a predetermined timing (step S202). The predetermined timing is a regular timing, or a timing when the access log (the table access log 410 and the column access log 420) is updated, etc.
[0044] In the loop process (A), the lineage management unit 602 checks the access log and extracts access details related to accesses to the data 200 since the previous check timing (step S203). The lineage management unit 602 updates the lineage record information 440 stored in the storage device 101 based on the extracted access details (step S204).
[0045] Thereafter, when a predetermined condition is satisfied, the lineage management unit 602 exits the loop process (step S205), cuts off access to the data (step S206), and ends the process.
[0046] FIG. 13 is a diagram for explaining an example of the KPI tree management process performed by the KPI tree management unit 603. As shown in FIG.
[0047] The KPI tree management unit 603 receives new node information on a new node, which is a node to be added to the KPI tree, from the client device 110 (step S301). The new node information includes, for example, the contents of the new node (node ID, category, KPI, KPI calculation method, and usage data, etc.) and the parent-child relationship of the new node (parent node ID, child node ID, etc.).
[0048] The KPI tree management unit 603 updates the KPI tree 500 stored in the storage device 101 based on the new node information (step S302), and ends the process.
[0049] FIG. 14 is a flowchart for explaining an example of the overall processing by the adaptive mapping unit 604.
[0050] In the overall process, the overall management unit 610 of the adaptive mapping unit 604 receives target designation information designating a view generation target from the client device 110. The target designation information indicates a target KPI tree, which is a KPI tree for which a view is to be generated, and a target node, which is a top node for which a view is to be generated. The overall management unit 610 acquires the target KPI tree from the storage device 101 based on the target designation information (step S401). The adaptive mapping unit 604 performs an adaptive mapping process (see FIG. 15) based on the target KPI tree and the metadata 300, in which the target node and nodes lower than the target node are set as target nodes for which a view is to be generated, and determines corresponding columns corresponding to the items of these target nodes (step S402). The adaptive mapping unit 604 performs a view generation process (see FIG. 17) for generating and outputting a view based on the processing result of the adaptive mapping process (step S403), and ends the process.
[0051] FIG. 15 is a flowchart illustrating an example of the adaptive mapping process in step S402 of FIG.
[0052] In the adaptive mapping process, the overall management unit 610 of the adaptive mapping unit 604 identifies a target node from the target KPI tree based on the target designation information (step S501). The adaptive mapping unit 604 sets the target node as the initial value of the target node (step S502).
[0053] Overall management unit 610 starts loop processing (B) in which steps S501 to S515 are repeated while changing the target node in a depth-first search procedure until the mapping destination of each item of each node below the target node is determined (step S503). The depth search procedure refers to increasing the rank of the target node from the initial value (target node) in order (going down the hierarchy) until the target node becomes a node for which a view is defined, and when the target node becomes a node for which a view is defined, conversely decreasing the rank of the target node (going up the hierarchy) until a view of the target node is generated.
[0054] In the loop process (B), overall manager 610 determines whether or not the target node is a leaf node (step S504).
[0055] If the target node is a leaf node (step S504: Yes), the overall management unit 610 checks whether a sibling leaf node exists that has the same parent node as the target node and is a leaf node of the same rank as the target node, and further checks whether a view of that sibling leaf node already exists (step S505).
[0056] If a view of a sibling leaf node exists (step S505: Yes), the overall management unit 610 transmits the node information, metadata 300, and lineage association of the target node to the similarity determination unit 611, and performs a mapping process to generate mapping candidate information indicating candidates for mapping destinations for each item of the node information of the target node. In this embodiment, the mapping destination of the usage data of the node information is a column of the data 200, and the mapping destination of the KPI calculation method of the node information is the job generation source. At this time, if there are multiple data 200 having a column that is a candidate for the mapping destination of the usage data, the overall management unit 610 adds a common column that is a column common to the multiple data 200 and the join key of the view of the sibling leaf node to the mapping candidate information as candidates for the join key of the view of the target node (step S506).
[0057] On the other hand, even if the external schema of the sibling leaf node is not defined (step S505: No), the overall management unit 610 performs the same process as step S506. However, if there are multiple data 200 having a column that is a candidate for mapping destination of the usage data, the overall management unit 610 adds a common column that is common to the multiple data 200 to the mapping candidate information as a candidate for the join key of the view of the target node. (Step S507).
[0058] When the processing of step S506 or S507 is completed, the overall management unit 610 checks whether a column of the data 200 is mapped to the usage data of the target node in the mapping candidate information (whether there is a column candidate corresponding to the usage data) (step S508).
[0059] If a column is mapped to the usage data (step S508: Yes), the overall management unit 610 sorts the mapping candidate information by arranging the mapping destination candidates in descending order of similarity for each item of the node information of the target node (usage data and KPI calculation method) (step S509). The overall management unit 610 determines a predetermined number of mapping destination candidates in descending order of similarity for each item of the node information of the target node as mapping destinations, and transmits the mapping information indicating the mapping destinations and the node information of the target node to the mapping management unit 612 (step S510). The predetermined number is, for example, 1 or more.
[0060] If no column is mapped to the usage data (step S508: No), the adaptive mapping unit 604 prompts the user via the client device 110 to modify or add to the usage data (step S511). The overall management unit 610 receives a request for modifying or adding to the usage data from the user via the client device 110, modifies or adds to the usage data of the target node in accordance with the request (step S512), and returns to the processing of step S505.
[0061] Also, if the target node is not a leaf node (step S504: No), the overall management unit 610 determines based on the mapping list 710 whether or not a lower node view (lower schema), which is a view corresponding to a lower node (child node) of the target node, exists (step S513).
[0062] If a lower node view does not exist (step S513: No), the processing for the target node is terminated. On the other hand, if a lower node view exists (step S513: Yes), the overall management unit 610 acquires the definition of the lower node view from the mapping list (step S514). Then, the overall management unit 610 performs the same processing as in step S506. However, based on the definition of the lower node view, the overall management unit 610 also adds the columns of data 200 included in the lower node view as mapping candidates for the data used by the target node, and adds the join key of the lower node view to the mapping candidates as a candidate for the join key of the view of the target node (step S515). Then, the processing of step S508 is executed.
[0063] Then, when mapping information has been generated for all nodes following the target node, overall manager 610 exits loop process (B) (step S516) and ends the process.
[0064] FIG. 16 is a flowchart for explaining an example of the similarity determination process by similarity determination section 611.
[0065] In the similarity determination process, the similarity determination unit 611 identifies two data sets 1 and 2 to be processed (step S601). Data set 1 is, for example, a set whose elements are columns of data 200, and data set 2 is, for example, a set whose elements are items of node information of a target node.
[0066] The similarity determination unit 611 starts a loop process (C) for repeating the processes of steps S603 to S605 for each combination of elements of the data sets 1 and 2 (step S602).
[0067] In the loop process, first, the similarity determination unit 611 calculates the similarity between the elements of the target combination using the metadata 300 (step S603). For example, the similarity determination unit 611 calculates the similarity for each piece of information about each element (name, type, content (data itself), and explanatory information given to the element). The similarity is, for example, the Levenshtein distance, but is not limited to this as long as it can calculate the similarity between character strings.
[0068] The similarity determination unit 611 determines whether the similarity is equal to or greater than a predetermined threshold (step S604). When the similarity determination unit 611 calculates the similarity for each piece of information on an item, etc., it determines whether the statistical value (e.g., average value or total value) of the similarity for each item is equal to or greater than a threshold.
[0069] If the similarity is equal to or greater than the threshold (step S604: Yes), the similarity determination unit 611 determines the elements of the target combination as candidates for corresponding mapping destinations (step S605). On the other hand, if the similarity is less than the threshold (step S604: No), the similarity determination unit 611 skips the process of step S605.
[0070] Then, when the processes of steps S603 to S605 have been executed for all combinations, the similarity determination unit 611 exits the loop process (C) (step S606) and ends the process.
[0071] By the above process, elements with high similarity can be made mapping candidates. In this embodiment, the result is that the usage data of the node information is mapped to the column of the data 200. Similarly, the KPI calculation method of the node information can be mapped to the job generation source.
[0072] FIG. 17 is a flowchart illustrating an example of a view generation process performed by the mapping manager 612.
[0073] In the mapping management process, the mapping management unit 612 receives target node information, which is node information of a target node, and mapping information (step S701). The mapping management unit 612 starts a loop process (D) in which the processes of steps S703 to S711 are repeated for each piece of mapping information (step S702).
[0074] The mapping management unit 612 judges whether or not there is an existing table that is an existing view generated from the same mapping information as the relevant mapping information (step S703).
[0075] If an existing table exists (step S703: Yes), the mapping management unit 612 adds the view ID for identifying the existing table to the view list as the view ID of the view corresponding to the target node (step S704). Note that the view ID can be obtained, for example, by an existing DBMS (Data Base Management System).
[0076] The mapping management unit 612 acquires the node ID of the target node from the node list 501 of the KPI tree 500 based on the target node information (step S705). The mapping management unit 612 associates the acquired node ID of the target node with the view ID of the view corresponding to the target node, and adds them to the mapping list (step S706).
[0077] If there is no existing table (step S703: No), the mapping management unit 612 generates a view definition for the target node based on the mapping information (step S707). For example, the mapping management unit 612 defines a view by combining data having corresponding columns corresponding to each item of the node information of the target node in the mapping information with a combine key included in the mapping information. At this time, the name of the view may be generated randomly, may be input by the user, or may be input in advance by the user. The mapping management unit 612 may also define a view using a job generation source. For example, the view definition is expressed as follows when written using SQL: create view xxx select yyyy from “job origin” … In addition, the source of jobs can also be inferred for KPIs with similar calculation methods.
[0078] The mapping management unit 612 generates a view based on the view definition (step S708). The mapping management unit 612 acquires a view ID of the generated view and adds it to the view list (step S709). The generation of the view and the acquisition of the view ID can be performed, for example, by a DBMS.
[0079] The mapping management unit 612 acquires the node ID of the target node from the node list 501 of the KPI tree 500 based on the target node information (step S710). The mapping management unit 612 associates the acquired node ID of the target node with the view ID of the view corresponding to the target node (the view ID of the generated view), and adds them to the mapping list (step S711).
[0080] Then, when the mapping management unit 612 has performed the processes of steps S703 to S711 for all the mapping information, it exits the loop process (D) (step S712) and ends the process.
[0081] As described above, according to this embodiment, the overall management unit 610 identifies columns of the data 200 that correspond to each item of the KPI tree 500 as corresponding columns, based on the similarity between each column of the data 200 and each item of the KPI tree 500. The mapping management unit 612 generates an external schema of the database made up of the data 200, based on the corresponding columns. This makes it possible to automatically generate an external schema that reflects the contents of the business, making it possible to provide an external schema that is easy to use for application developers who have little knowledge about the business.
[0082] Furthermore, in this embodiment, the similarity determination unit 611 calculates the similarity between each column of the data 200 and each item of the KPI tree 500 based on the metadata 300 related to the data 200. This makes it possible to easily calculate the similarity.
[0083] In this embodiment, the mapping manager 612 generates an external schema by combining data including corresponding columns using a common column as a combine key, which is a column common to the data. This makes it possible to generate an external schema that is more appropriate for the business.
[0084] In this embodiment, the overall management unit 610 identifies columns that correspond to the usage data used to calculate the KPIs as corresponding columns, which makes it possible to generate a more appropriate external schema that reflects the content of the business.
[0085] Furthermore, according to this embodiment, the nodes below the specified target node are selected as target nodes, and a view is generated for each target node. This makes it possible to generate an appropriate external schema according to the content of the business.
[0086] Furthermore, according to this embodiment, the join keys of the views of the sibling leaf nodes and child nodes of the target node are used as the join keys of the view of the target node, making it possible to generate an appropriate external schema according to the content of the business.
[0087] The above-described embodiments of the present disclosure are illustrative examples of the present disclosure, and are not intended to limit the scope of the present disclosure to only these embodiments. A person skilled in the art can implement the present disclosure in various other forms without departing from the scope of the present disclosure. [Explanation of symbols]
[0088] 100: View generating device 101: Storage device 102: Processor 103: Memory 104: Communication unit 105: Input interface 106: Output interface 107: Bus 110: Client device 120: Network 200: Data 300: Metadata 400: Lineage related information 500: KPI tree 601: Metadata extraction unit 602: Lineage management unit 603: KPI tree management unit 604: Adaptive mapping unit 610: Overall management unit 611: Similarity determination unit 612: Mapping management unit
Claims
1. An external schema generation device for generating an external schema of a database generated by a business, comprising: an identification unit that identifies, based on a similarity between each column of data included in the database and each item of goal information related to an index value indicating a goal of the business, the column corresponding to each item of the goal information as a corresponding column; a generating unit configured to generate the external schema based on the corresponding columns.
2. The external schema generating apparatus according to claim 1 , further comprising a calculation unit that calculates the similarity based on metadata related to the data included in the database.
3. 2 . The external schema generating device according to claim 1 , wherein the generating unit generates the external schema by combining the data including the corresponding column with a common column as a combine key.
4. 2. The external schema generator of claim 1, wherein the goal information items include usage data used to calculate the index value.
5. The external schema generating device according to claim 1 , wherein the item of the goal information includes a calculation method for calculating the index value.
6. The plurality of pieces of target information are hierarchically organized, The identification unit identifies, for each target target information, which is the target information for which the external schema is to be generated, the corresponding column corresponding to each item of the target target information; The external schema generating device according to claim 1 , wherein the generating unit generates the external schema for each of the target information.
7. The external schema generating device according to claim 6 , wherein the specifying unit determines the specified goal information and goal information in a lower hierarchy than the specified goal information as the target goal information.
8. The external schema generating device of claim 6, wherein when a subschema, which is the external schema for the target information subordinate to the target information, has already been generated, the generating unit generates the external schema of the target information using a join key in the subschema.
9. 7. The external schema generation device according to claim 6, wherein, when the target target information is leaf information under which no target information exists, when sibling leaf information exists that is lower target information than the target information under which the target target information exists and that is leaf information at the same hierarchical level as the target target information under which the target target information exists, and when a sibling schema exists that is the external schema for the sibling leaf information, the external schema for the sibling schema is generated using a join key in the sibling schema.
10. 1. An external schema generating method for generating an external schema of a database generated by a business process, comprising: identifying, based on a similarity between each column of data included in the database and each item of goal information related to an index value indicating a goal of the business, the column corresponding to each item as a corresponding column; The external schema generating method generates the external schema based on the corresponding columns.
Citation Information
Patent Citations
Application builder with automated data objects creation
US20170115968A1
