CN111813798B

CN111813798B - Mapping method, device, equipment and storage medium based on R2RML standard

Info

Publication number: CN111813798B
Application number: CN202010661422.9A
Authority: CN
Inventors: 吴思竹; 修晓蕾; 钱庆; 邬金鸣; 何晓琳; 孙小康
Original assignee: Institute of Medical Information CAMS
Current assignee: Institute of Medical Information CAMS
Priority date: 2020-07-10
Filing date: 2020-07-10
Publication date: 2023-06-23
Anticipated expiration: 2040-07-10
Also published as: CN111813798A

Abstract

The invention provides a mapping method, a device, equipment and a storage medium based on R2RML standard, which are used for determining a mapping task from a relational database based on the R2RML standard created by a user to a resource description framework and connecting the relational database indicated by the mapping task; performing explicit structural features, data features and latent semantic analysis on the relational database to obtain an association relation table of the relational database; determining a target mapping mode selected by a user from at least one preset mapping mode; and carrying out RDF triple definition by utilizing the target mapping mode and the association relation table, automatically generating an R2RML mapping document based on the R2RML mapping rule, and further generating an RDF triple file according to the R2RML mapping document. The invention can automatically generate the R2RML mapping document and realize the automatic generation of the RDF triplet file.

Description

Mapping method, device, equipment and storage medium based on R2RML standard

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a mapping method, apparatus, device, and storage medium based on an R2RML standard.

Background

With the continuous development of science and technology, more and more mapping conversion tools for mapping relational databases to resource description frameworks are available, such as DB2Triple, morph-RDB, R2RML Parser, virtuoso Universal Server, XSPARQL, ontop, sparqlMap, sparqlify, geoTriples, R RML-kit. These mapping transformation tools are mainly performed by parsing and mapping focused on R2RML mapping rules, mapping relational databases to resource description frameworks.

However, in the use process of the actual mapping conversion tool, the existing mapping conversion tool lacks functions of relational database pattern analysis and latent semantic discovery; the existing mapping tool is generally based on a command line or interface, does not support RDF vocabulary recommendation or importing of a domain model, generally directly edits rules for applying domain knowledge into a mapping document by a user, lacks a domain semantic modeling auxiliary function, and has higher requirements on the technology and professional ability of the user. Moreover, the existing mapping conversion tool only provides the input and mapping execution functions of the R2RML mapping document, and the R2RML mapping document cannot be automatically generated; and further, automatic generation of the RDF triplet file cannot be realized. If the R2RML mapping document is obtained, manual editing by a user is needed, and the requirements on the technology and professional ability of the user are high. The existing mapping conversion tool also lacks a quality control mechanism for the mapping process, and cannot check and comprehensively report the mapping result RDF triplet file. The user cannot learn the quality of the generated RDF triplet file.

Disclosure of Invention

In view of the above, the present invention provides a mapping method, apparatus, and device storage medium based on R2RML standard, so as to automatically generate an R2RML mapping document, thereby implementing automatic generation of an RDF triplet file. The technical scheme is as follows:

The first aspect of the invention discloses a mapping method based on R2RML standard, comprising the following steps:

determining a mapping task from a relational database based on R2RML standard created by a user to a resource description framework, and connecting the relational database indicated by the mapping task;

performing explicit structural features, data features and invisible semantic analysis on the relational database to obtain an association relation table of the relational database;

determining a target mapping mode selected by a user from at least one preset mapping mode, wherein the at least one mapping mode comprises a direct mapping mode, a custom mapping mode and a mapping mode by means of a domain model;

performing RDF triple definition by utilizing the target mapping mode and the association relation table, and automatically generating an R2RML mapping document based on an R2RML mapping rule;

and generating an RDF triplet file according to the R2RML mapping document.

Optionally, if the target mapping mode is a mapping mode by means of a domain model, the method further includes:

determining a domain ontology of the mapping task, wherein the domain ontology is a domain ontology recommended to the user based on task description and data source description carried by the mapping task, or the domain ontology is a domain ontology imported by the user, or the domain ontology is a domain ontology built by the user on line;

Determining a mapping relation between fields in a data table of the relational database and domain ontology classes of the domain ontology;

the RDF triplet definition is performed by using the target mapping mode and the association relation table, and the R2RML mapping document is automatically generated based on the R2RML mapping rule, which comprises the following steps: and carrying out RDF triple analysis by utilizing the target mapping mode, the association relation table and the mapping relation, and automatically generating an R2RML mapping document based on an R2RML mapping rule.

Optionally, if the target mapping mode is a custom mapping mode or a mapping mode by means of a domain model, the performing RDF triplet definition by using the target mapping mode and the association table, and automatically generating an R2RML mapping document based on an R2RML mapping rule, includes:

determining RDF vocabulary, wherein the RDF vocabulary comprises existing RDF vocabulary, custom RDF vocabulary and RDF vocabulary recommended for users;

and generating an R2RML mapping document based on the RDF vocabulary and the R2RML mapping rule by using the target mapping mode, the association relation table and the defined RDF triple structure.

Optionally, the method further comprises:

detecting grammar errors of the R2RML mapping document to generate grammar check results of the R2RML mapping document, wherein the grammar check results comprise mapping rule spelling errors, grammar errors, logic errors, and wrong problem descriptions and problem line numbers;

Performing null value verification on the relational database to obtain a null value verification result of the relational database;

pre-computing a result of R2RML mapping performed according to the R2RML mapping document to generate a mapping result and predicting a mapping result generation time, wherein the mapping result comprises a triplet number, a redundant triplet number, a unique subject number, a unique predicate number and a unique object number; the number of blank nodes comprises a subject blank node number and an object blank node number;

and displaying the processing mechanism of the blank node and the processing mechanism of the redundant node.

Optionally, the generating an RDF triplet file according to the R2RML mapped document includes:

and generating an RDF triplet file according to the R2RML mapping document and each processing mechanism by combining the setting operation of the user on whether to add the database field annotation and the constraint to the triplet and the selection operation of the user on the processing mechanism of the blank node and the processing mechanism of the redundant node.

Optionally, the method further comprises:

performing data evaluation on the RDF triplet file to obtain a result evaluation report of the RDF triplet file, wherein the result evaluation report comprises the following steps: statistics of the number of resulting data, validity of a specific data format, and RDF data quality assessment indicators.

Optionally, the method further comprises:

receiving an SPARQL retrieval request sent by the user, and acquiring a query result of the SPARQL retrieval request and returning the query result to the user; the SPARQL retrieval request indicates a virtual retrieval or RDF document retrieval;

if the SPARQL search request indicates virtual search, the obtaining the query result of the SPARQL search request, which is returned to the user, includes: and converting the SPARQL query statement corresponding to the SPARQL retrieval request into an SQL query statement, directly accessing the relational database according to the SQL query statement to obtain a query result, converting the query result into an RDF triplet, and returning the RDF triplet to the user.

The second aspect of the present invention discloses a mapping apparatus based on R2RML standard, including:

the first determining unit is used for determining a mapping task from the R2RML relational database created by a user to the resource description framework and connecting the relational database indicated by the mapping task;

the analysis unit is used for carrying out explicit structural characteristics, data characteristics and invisible semantic analysis on the relational database to obtain an association relation table of the relational database;

the first receiving unit is used for determining a target mapping mode selected by a user from at least one preset mapping mode, wherein the at least one mapping mode comprises a direct mapping mode, a custom mapping mode and a mapping mode by means of a domain model;

The first generation unit is used for defining RDF triples by utilizing the target mapping mode and the association relation table and automatically generating R2RML mapping documents based on R2RML mapping rules;

and the second generation unit is used for generating an RDF triplet file according to the R2RML mapping document.

In a third aspect of the invention, an apparatus is disclosed comprising: the device comprises a processor and a memory, wherein the processor and the memory are connected through a communication bus; the processor is used for calling and executing the program stored in the memory; the memory is used for storing a program, and the program is used for realizing the mapping method based on the R2RML standard.

A fourth aspect of the present invention discloses a computer-readable storage medium having stored therein computer-executable instructions for performing the mapping method based on the R2RML standard as disclosed in any one of the above first aspects of the present invention.

The invention provides a mapping method, a device, equipment and a storage medium based on R2RML standard, which are characterized in that a mapping task from an R2RML relational database created by a user to a resource description framework is determined, and the relational database indicated by the mapping task is connected; performing explicit structural features, data features and invisible semantic analysis on the relational database to obtain an association relation table of the relational database; determining a target mapping mode selected by a user from at least one preset mapping mode, wherein the at least one mapping mode comprises a direct mapping mode, a custom mapping mode and a mapping mode by means of a domain model; and carrying out RDF triple definition by utilizing the target mapping mode and the association relation table, automatically generating an R2RML mapping document based on an R2RML mapping rule, and further automatically generating an RDF triple file according to the R2RML mapping document, thereby solving the problem that the automatic generation of the RDF triple file cannot be realized due to the fact that the R2RML mapping document cannot be automatically generated in the prior art.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic structural diagram of an implementation of a RDB2RDF mapping conversion tool function based on the R2RML standard according to an embodiment of the present invention;

FIG. 2 is a main data interaction flow chart of an RDB2RDF mapping conversion tool based on the R2RML standard according to an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of an RDB2RDF mapping conversion tool based on the R2RML standard according to an embodiment of the present invention;

fig. 4 is an exemplary diagram of parameter configuration according to an embodiment of the present invention;

FIG. 5 is an exemplary diagram of mapping language conversion according to an embodiment of the present invention;

FIG. 6 is a schematic flow chart of a mapping from a relational database to a resource description framework implemented by an RDB2RDF mapping transformation tool based on the R2RML standard according to an embodiment of the present invention;

FIG. 7 is a schematic diagram of a relational database data structure and semantic analysis according to an embodiment of the present invention;

FIG. 8 is a schematic structural diagram of a design concept for providing 3 mapping modes by an RDB2RDF mapping conversion tool based on R2RML standard according to an embodiment of the present invention;

FIG. 9 is a schematic diagram of a mapping definition view and a mapping matching recommendation according to an embodiment of the present invention;

FIG. 10 is a schematic diagram of a structure of multiple data mapping document inspection and result data evaluation according to an embodiment of the present invention;

FIG. 11 is a schematic diagram of a mechanism for performing data mapping and optimizing data query performance according to an embodiment of the present invention;

FIG. 12 is a schematic diagram of a structure of a data mapping document and a data collaboration sharing platform according to an embodiment of the present invention;

FIG. 13 is a schematic structural diagram of a mapping device based on R2RML standard according to an embodiment of the present invention;

fig. 14 is a hardware block diagram of a server according to an embodiment of the present invention;

fig. 15 is a block diagram of a hardware structure of a terminal used in a mapping method based on an R2RML standard according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

In this application, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

As is apparent from the background above, a relational database can be mapped to a resource description framework using a mapping transformation tool.

The applicant finds through research that the existing mapping conversion tool cannot automatically generate an R2RML mapping document, and further cannot realize automatic generation of an RDF triplet file, and has the following problems:

(1) The existing mapping conversion tool only provides connection to the data source, obtains browsing and displaying of the data source, and does not provide full analysis of the original data structure. Before mapping the relational database to the resource description framework, if comprehensive analysis and comprehensive knowledge of multiple aspects such as structure, grammar, semantics and the like of original data are lacking, mapping data quality and mapping efficiency can be affected when mapping rules are written and generated. Or the mapping problem is found after mapping, the editing mapping rule needs to be modified again, and then mapping and re-execution are carried out, so that the workload of a user is increased.

(2) In the existing mapping process, when the custom mapping mode is completed based on the R2RML standard or the mapping mode is completed by means of the domain model, a user is required to know own relational database data, have IT technology and write mapping rules. In addition, the user is familiar with domain knowledge, can find a proper RDF vocabulary or construct a proper domain model, and has high requirements on the technical and domain knowledge of the user. In addition, the conventional mapping conversion tool is usually based on a command line or an interface, does not support RDF vocabulary recommendation or domain model import, is directly edited in an R2RML mapping document by a user by applying domain knowledge, and is difficult for the user without knowing the domain model to complete the task. Although the existing mapping conversion tool can integrate the domain ontology modeling tool prot into the mapping function, further support domain ontology editing and integration application, support the import and export of R2RML mapping documents, but does not provide the editing function of R2RML mapping rules, only support editing of self-created language, and the prot has limited support for Chinese ontologies.

(3) In existing mapping transformation tools, part of the mapping transformation tool provides a mapping validity ware, performs syntax and program checking of mapping rules, such as field name errors, mapping rule errors, etc., but such detection is performed in actual operation. If the data volume is large, the program can detect errors after running for a period of time, the program reports the errors and does not continue to run to generate RDF triples, and the user needs to modify the mapping rules and restart the mapping execution. The program will stop running whenever a problem is encountered, and restarting may also stop at the next problem, causing excessive time consumption. In addition, existing mapping transformation tools only have the capability to perform mapping, lack quality control over the data mapping process, and do not provide for inspection and comprehensive reporting of RDF triples. The user cannot learn the quality of generating RDF triples, which may have a large number of redundant triples or nonsensical blank nodes, severely mapping data quality.

(4) A large number of users develop different mapping practices from Relational DataBase data (RDB) to RDF based on R2RML standards, and some users share open source tools through a Github and other platforms or issue converted open shared data through an open data platform, but less involves the mapping task and sharing of R2RML mapping documents. From the practical mapping, the mapping task and the mapping document of other people can be known and referred to more quickly to obtain the writing experience of the mapping rule and the realization experience of the mapping flow. Currently, there is no shared medium or platform for mapping task resources (including mapping tasks, R2RML mapping documents, domain models, etc.).

Therefore, the invention provides a mapping conversion tool from a relational database based on R2RML standards to a resource description framework, which can automatically generate R2RML mapping documents on the basis of improving the quality and the mapping efficiency of mapping data, further automatically realize the generation of RDF triple files according to the automatically generated R2RML mapping documents, evaluate the generated RDF triple files and generate an evaluation report. After the R2RML mapping document is automatically generated, the generated R2RML mapping document can be shared, so that other users can know and reference, and further, the writing experience of the mapping rule and the realization experience of the mapping flow can be obtained more quickly. The RDB2RDF mapping conversion tool based on the R2RML standard can also perform mapping, RDF instantiation, SPARQL query supporting data and visual browsing according to the requirements of users.

The application realizes the mapping of the relational database to the resource description framework by constructing an RDB2RDF mapping conversion tool and a data sharing platform based on the R2RML standard. The application mainly describes the development and implementation of an RDB2RDF mapping conversion tool based on the R2RML standard. First, a brief discussion of development tools, tool operating environments, platform operating environments, etc. is provided; then, introducing the function development and main data interaction flow of an RDB2RDF mapping conversion tool based on the R2RML standard; finally, the implementation effect of the RDB2RDF mapping conversion tool based on the R2RML standard is introduced in a graphic combination mode by combining experimental examples.

The applicant discovers and analyzes the existing RDB2RDF mapping conversion tool that Ontop has better effect on virtual query, and DB2Triple has best comprehensive performance on instantiation mapping. In order to quickly construct an R2RML mapping conversion tool with universality, the application adopts an Otop and DB2Triple as R2RML mapping conversion bottom technical support based on JAVA language, and adopts a mainstream SpringBoot framework to develop an IMI R2RML mapping conversion tool.

In the embodiment of the application, the running environment of the RDB2RDF mapping conversion tool based on the R2RML standard comprises a server configuration, a server environment and a client environment. The Server configuration comprises an operating system (Windows Server 2008R 2 enterprise edition), a system type (64-bit Chinese operating system), a processor (Intel (R) Xeon (R) CPU E7-4820@2.00GHz), a memory (64 GB) and a bandwidth (5 Mbps); server environments include databases (MySQL 5.6.17), JAVA environments (jdk 1.8), python environments (Python 3.7), tomcat containers (Tomcat 8.0), full text index services (solr 6.5); the client environment includes an operating system (Windows 7, windows10, linux, etc.), a memory (above 16G memory), a browser (google browser (recommendation), ie10+, very fast browser, 360 browser).

The platform running environment comprises a server configuration, a server environment and a client environment. The system comprises an operating system (Windows Server 2008R 2 enterprise edition), a system type (64-bit Chinese operating system), a processor (Intel (R) Xeon (R) CPU E7-4820@2.00GHz), a memory (16 GB) and a bandwidth (10 Mbps); server environments include databases (MySQL 5.6.17), JAVA environments (jdk 1.8), python environments (Python 3.7), tomcat containers (Tomcat 8.0), full text index services) (solr 6.5); the client environment includes an operating system (Windows 7, windows10, linux, etc.), a memory (above 16G memory), a browser (google browser (recommendation), ie10+, very fast browser, 360 browser).

In the embodiment of the application, the integrated open source of the RDB2RDF mapping transformation tool based on the R2RML standard comprises an associated open vocabulary (Linked OpenVocabularies, LOV), and LOV aims to help the associated data publisher and the user better acquire, share and reuse the vocabulary in the associated data. The LOV vocabulary contains a series of definitions of classes and attributes that describe specific types or specific domains or specific uses of things, as well as links to various associated data. The existing LOV functional network provides an open API interface of Https, and interface calling can be performed in a mode of Httppost or HttpGet, and the application adopts the latter mode to call. The BioPortal vocabulary retrieval API consists of a set of resources (ontologies, classes, etc.) and associated endpoints (annotators, recommenders, etc.), which are linked together by links. When the method and the device are used for calling, interface calling is carried out in an Httppost or HttpGet mode, and access can be carried out through a registration key in the calling process. AnotherRDF Parser, anotherRDF Parser is functional to provide an RDF triplet file verification service, currently in use version 2-alpha-1, that currently supports the last work draft specification published by the RDF core workgroup (Last Call Working Draft specifications) and no longer supports obsolete elements and attributes in the RDF model and grammar specifications. When the application is called, the JAVA open source code is downloaded in the Github official network, and then the modified source code is placed in the project, so that the verification service function is integrated into the RDB2RDF mapping conversion tool based on the R2RML standard. D2rq_r2rml, which provides the functionality to mutually translate the two mapping languages D2RQ and R2 RML. When the application is called, the code is written in the Python language, so that the encapsulated Python interface is requested to be called through the Http interface. DB2Triple belongs to a lightweight tool, is easy to call and integrate into other applications, has good effects in following R2RML grammar rules and instantiation mapping performance tests, downloads source codes from Github when calling, packages the source codes into jar packages through Maven, and refers to the jar packages in a program. The application calls the tool to complete generating the RDF Triple file based on the R2RML mapping document, and after the program obtains the R2RML mapping document and JDBC related configuration, the RDF Triple file is output by calling a built-in method of DB2 Triple. Onttop provides powerful support for on-demand mapping, which improves query performance through the latest generation of query rewrite techniques as well as query optimization techniques. When the application is called, the OnTop source code is downloaded from the Github and then packaged into a jar package through Maven, and then the jar package is referenced in a program. The virtual query of the RDF triple file is realized by calling the tool, namely, the SPARQL language for querying the RDF triple file is converted into the SQL language for querying the relational data, and the relational data is converted into the RDF triple file for outputting the result after being acquired.

The RDB2RDF mapping conversion tool based on the R2RML standard can independently complete browsing, inquiring and visually displaying and sharing from pattern analysis of a data source to mapping result, and specific processes comprise creating and managing a mapping task, importing a data source connection and a domain model, analyzing the pattern, defining the mapping, editing and detecting the mapping, executing the mapping, evaluating the quality of RDF triplet files, visually browsing and inquiring, sharing the mapping task and the like. The specific implementation of the RDB2RDF mapping conversion tool function based on the R2RML standard and the main data interaction flow are shown in fig. 1 and 2.

Aiming at the key problems of high R2RML mapping rule writing difficulty, insufficient support strength in field knowledge modeling and semantic enrichment, lack of evaluation and quality control mechanisms for R2RML mapping results and the like in the existing R2RML standard mapping mode and mapping conversion tool mapping process, the invention provides an interactive, easy-to-understand and operable RDB2RDF mapping conversion tool based on the R2RML standard. The RDB2RDF mapping conversion tool based on the R2RML standard assists a user in carrying out relational database pattern analysis by providing visual relational database pattern analysis, implicit semantic prompt and an understandable and easy-to-operate mapping definition view; the method comprises the steps of supporting 3 different mapping modes including a direct mapping mode, a custom mapping mode and a mapping mode by means of a domain model, providing RDF vocabulary recommendation and query, R2RML mapping document detection according to the characteristics, requirements and purposes of the different mapping modes, providing automatic mapping pair recommendation through various semantic similarity calculations, domain model recommendation, creation and editing, data evaluation and quality control (including detection and processing of empty white nodes and redundancy triples) of the R2RML mapping document, and visualizing results. The tool can reduce the mapping operation difficulty of a user to a certain extent, improve the convenience and the understandability of the user operation, and promote the learning, popularization and application of the R2RML mapping standard.

Referring to fig. 3, a schematic structural diagram of an RDB2RDF mapping conversion tool based on R2RML standard according to an embodiment of the present invention is shown. The RDB2RDF mapping conversion tool based on the R2RML standard specifically comprises a base layer, provides tool bottom layer support and comprises an operating system, a hardware server, base software and network equipment.

And the storage layer is mainly used for storing the generated data, including user data, log data, basic data in the mapping process and generated RDF file data.

The support layer provides application support and comprises functions of user management, task management, relational data management, mapping base management, data mapping management, field model editing, data mapping result detection and quality control, data retrieval and log management.

The user management is mainly to perform user registration, user information browsing, user roles and authority management. And the task management mainly comprises the steps of mapping task creation, task browsing, task editing, task deletion and task sharing. The relationship data management mainly supports the user to create data source connection, and performs the creation, editing, modification and deletion of the relationship database connection, and the browsing and the query of the data. The mapping base library management is the base library management for supporting data mapping, and comprises data source management, RDF vocabulary management, domain model management and mapping optimization parameter setting management, wherein parameter configuration is carried out through the mapping optimization parameter setting management as shown in figure 4. And the data mapping management is used for carrying out data mapping step management, including namespace configuration, relational database data source mode analysis and mapping definition (including direct mapping mode, custom mapping mode and mapping mode by means of a domain model, and providing a visual mapping definition view, a mapping definition audit view and a mapping document detection editing view), and a user can realize data mapping definition and audit of different modes through different views. R2RML mapping document detection, grammar detection of R2RML mapping documents generated according to mapping definition, and pre-running result processing of mapping results. R2RML data mapping is executed, and R2RML mapping document execution is performed after detection. The mapping semantic enrichment auxiliary function is that in the mapping process, an RDB2RDF mapping conversion tool based on the R2RML standard provides auxiliary functions in a data mapping flow, wherein the auxiliary functions comprise similar data mapping task recommendation, namely recommending shared mapping tasks similar to the mapping tasks to be used for reference by a user; data mapping matching recommendation based on various similarity algorithms (general data similarity and domain data similarity algorithms) is performed in an auxiliary manner by means of data mapping matching of a domain model (comprising matching of data columns and domain ontology classes or attributes); a relational database recessive semantic relation prompts, so that semantic deletion in conversion is reduced; the data mapping mode of the domain model is used for supporting domain ontology, and the tool recommends the existing domain knowledge ontology according to the mapping task description and the source data to expand domain semantics; tools provide RDF vocabulary recommendation functionality from multiple sources to support data description semantic expansion. The field model management is oriented to the field semantic driven data mapping requirement and supports the functions of importing, editing, creating, modifying, browsing and storing the field knowledge model. And detecting and controlling the quality of the data mapping result, providing statistical analysis and evaluation for generating the RDF triplet file, evaluating and controlling the quality of the RDF triplet file through evaluation indexes, and providing multi-dimensional mapping result data evaluation. And performing intervention and processing according to the conditions of influencing the data quality, such as redundant triples, blank nodes and the like, in the evaluation result, and improving the data quality of the RDF triples. Visual browsing and query, and provides SPARQL to SQL query and optimization mechanism for on-demand conversion. While for instantiation transformations, a Jena-based SPARQL query is provided. And the log management provides log editing, log browsing, log inquiring and log deleting functions.

The application layer provides mapping services through a user interface, including data mapping, data browsing and querying, data detection and evaluation, data mapping document sharing and other services.

The data mapping provides data mapping service, and by creating mapping tasks, mapping modes in different modes such as relational data mode analysis, direct mapping mode, custom mapping mode, mapping mode by means of domain model and the like are supported. Data browsing and query, supporting different mapping modes, converting and instantiating converted data query modes according to needs, and supporting data query result visualization, RDF triple file downloading and domain knowledge model editing and browsing. And (3) detecting and evaluating mapping data, and supporting R2RML mapping document detection, data evaluation of an RDF triple file, quality control processing of an evaluation result of the RDF triple file and statistic analysis of the RDF triple file. Data and RDF mapping document sharing, supporting mapping task full-flow file sharing, RDF triplet file sharing, R2RML mapping document sharing and R2RML mapping standard browsing. Other services, in addition to the service functions described above, provide for conversion between other tool mapping languages, such as D2RQ and R2RML, where conversion between D2RQ and R2RML is shown in FIG. 5.

Referring to fig. 6, a schematic flow chart of a mapping from a relational database to a resource description framework is shown, which is provided by an embodiment of the present invention, by using an RDB2RDF mapping conversion tool based on R2RML standard, and includes 8 main steps of mapping task creation, data connection and domain model import, pattern analysis, mapping definition, mapping editing and detection, mapping implementation, result evaluation, and mapping task management and sharing.

Mapping task creation: and creating a mapping task, and carrying out mapping task description and related resource allocation in the mapping task. If the RDB2RDF mapping conversion tool based on the R2RML standard is connected with the platform after logging in, the RDB2RDF mapping conversion tool based on the R2RML standard is connected with the platform, and the RDB2RDF mapping conversion tool based on the R2RML standard can provide recommendation of related data tasks according to the mapping task description for learning and reference of a public user.

Data connection and domain model importation: after determining the mapping task from the R2RML relational database created by the user to the resource description framework, establishing data connection with the relational database indicated by the mapping task, such as MySQL, H2, ms SQL and the like. The schema and data of the relational database may be accessed through a connection to the relational database. If the pattern is mapped by means of a domain model, the domain ontology needs to be imported. The system supports self-built domain ontology import (namely, a user imports a domain ontology by himself), also supports network ontology import (namely, a user builds a domain ontology online), can edit and browse the ontology, and can provide domain ontology recommendation related to data mapping (namely, recommending the domain ontology to the user based on task descriptions and data source descriptions carried by created mapping tasks) by an RDB2RDF mapping conversion tool based on an R2RML standard.

Pattern analysis: after connection is established with the relational database indicated by the created mapping task, the explicit structural features and data characteristics of the relational database need to be fully analyzed, and especially the implicit semantics and the explicit mapping targets need to be analyzed, and the external semantics need to be properly expanded according to the needs. If the domain ontology is required to be introduced into the mapping, the appropriate domain ontology is required to be constructed, and the mode structure and semantic characteristics of the domain ontology are sufficiently understood, analyzed and researched, so that the subsequent mapping can be better defined. Further, the problems of lack of data source RDB mode analysis and latent semantic discovery in the prior art are solved.

Referring to FIG. 7, relational data schema analysis is primarily connected to a relational database through a data source connection providing mapping to one or more tables in the source relational database for browsing. Fully exhibiting and presenting the dominant structure and the semantics of the structure, the data and the association relation in the relational database, giving a prompt for the auxiliary discovery of the hidden structure and the semantics, and enabling the user to select and represent the latent structure and the semantics in a dominant way. The explicit structure includes a data table, a data column (also referred to as a field, including information of column name, data type, comments, etc.), a value (including data value), and a constraint (including primary key constraint, foreign key constraint, non-null constraint, and unique constraint), and the supported operations include viewing of the data table, browsing, viewing of the constraint, editing of the comments, etc.

In addition to explicit structures, the relational database also includes implicit structures and semantics that are often easily ignored by the user in the mapping or that need to be expanded, supplemented, and refined in the mapping objective. The invisible results include column name potential associations, column name comments, data statistics, and inspection constraints. RDB2RDF mapping transformation tool based on R2RML standard perfects the semantics of transformed data according to easily ignored hidden structure and semantics through pattern analysis.

Wherein, the column names are potentially associated: before data mapping, the data to be mapped is split as clearly as possible, and is split into a plurality of entity tables, and a clear relation between the entity tables is established. However, in actual mapping, the user does not split the data too finely, even does not create an association relationship between the primary key and the foreign key for the data, but the relationship between the hidden primary key and the hidden foreign key exists in the data, and at this time, the relationship should be automatically identified and prompted to the user so that the user can adopt an appropriate mapping strategy when performing data mapping. Column name annotation: the tool supports viewing column name annotations created by source data and also provides labeling and editing functions for column name meanings that are difficult for humans and machines to understand in mapping relational database table applications or conversion to generate RDF triples files, as column naming of a data table is not necessarily a full description of the column representation meaning, most likely an abbreviation or code, such as version_name is abbreviated as PN. Therefore, the RDB2RDF mapping conversion tool based on the R2RML standard provides a function of adding comments to the existing column names in pattern analysis, can enhance the semantics of the converted RDF triplet file, and is convenient for a machine and a human to understand and process data. In addition, the addition of column name annotation can also be applied to assist in achieving semantic similarity calculation in the subsequent mapping pair matching process, and is used for recommending the mapping pair. Checking constraints: the checking constraint is to define an identifier in the database relational table that checks a column of new input data of the relational table against the set logic for limiting the range of values in the column. The checking constraint may be a numerical range of values or an enumerated type.

Mapping definition: RDB2RDF map transformation tools based on the R2RML standard provide 3 mapping modes for definition, including direct mapping modes, custom mapping modes, and mapping modes by means of a domain model. When it is detected that the user-selected map is a direct-mapped pattern, the user-selected map pattern, i.e., the direct-mapped pattern, is determined as the target map pattern for ease of understanding.

In the embodiment of the application, the three mapping modes are not isolated, the direct mapping mode and the custom mapping mode are the basis of the mapping mode by means of the domain model, and the mapping by means of the domain model mode can be performed on the basis of the direct mapping mode and the custom mapping mode.

In a custom mapping mode and a mapping mode by means of a domain model mapping mode, an RDB2RDF mapping conversion tool based on an R2RML standard mainly provides support of domain semantics from three aspects, namely, recommendation and import of a domain knowledge model (ontology); 2. providing a custom domain ontology modeling tool; 3. the multi-source RDF vocabulary recommendation is provided to solve the problem that the current tool is lack of domain semantic auxiliary functions, and the specific structure is shown in FIG. 8.

Wherein domain knowledge model (ontology) recommendations and importation are provided. The mapping mode of the domain model is supported, corresponding domain ontology is recommended from a self-defined ontology library and a BioPortal interface integrated by a tool through various semantic similarity algorithms according to the created description of the mapping task and the source data description, and the self-built domain ontology is also supported to be directly imported, or the network open domain ontology is imported through adding-URI to serve as a domain knowledge model. The self-defined domain ontology modeling tool is provided, and on the basis of source data pattern analysis, the RDB2RDF mapping conversion tool design based on the R2RML standard provides a domain semantic modeling editing function, so that simple Chinese and English domain ontology creation, editing, modification, storage and export can be performed. And mapping is performed according to the mapping task and the relationship data description recommendation proper domain ontology filled in when the user creates the mapping task, so that more choices are provided for the user. Providing multi-source RDF vocabulary recommendation, wherein in the mapping process of a custom mapping mode by means of a domain model mapping mode, predicates can be originally derived from column names of relational database data, the predicates can not fully and clearly express semantics, and the predicates can also multiplex the vocabulary in a universal RDF vocabulary. However, the existing mapping conversion tool does not provide a vocabulary in a suitable general or field RDF vocabulary, and the RDB2RDF mapping conversion tool based on the R2RML standard integrates a multi-source associated data open vocabulary Linked Open Vocabularies (LOV) and ontology resources of biomedical field BioPortal, and supports building of a custom RDF vocabulary, and in the mapping process, RDF vocabulary recommendation and RDF vocabulary search defined by predicates and data types are provided through various semantic similarity algorithms, so that semantic expression of mapping data is improved, and convenience is brought to users.

In this embodiment of the present application, 3 definition views and an R2RML mapping rule library may be designed in the mapping definition stage according to different mapping modes, where the definition views and the R2RML mapping rules are combined, and the mapping rules are called according to the mapping definition of the user, so as to help the user complete the data mapping definition as shown in fig. 9. The definition view comprises a visual mapping definition view, a mapping definition audit view and an R2RML editing detection view. In the mapping process, in a mapping mode by means of a domain model, an RDB2RDF mapping conversion tool based on an R2RML standard also provides mapping semantic matching of a relational database and a domain ontology through various general and domain semantic similarity algorithms, performs mapping pair recommendation, and supports mapping matching of Chinese and English data.

The visual mapping definition view displays the structure of the relational database through a visual interface, if the structure is in a mapping mode by means of a domain model, the domain ontology structure is displayed, and mapping between columns or mapping between columns and classes or attributes is performed in a dragging mode. If the mapping mode is performed, the table without the primary key needs to specify a subject, and then the mapping definition is directly completed without dragging. If the custom mapping mode is performed, the relationship between tables can be established. The mapping mode of the domain model can be used for mapping the relational data table, the ontology class and the attribute based on direct mapping and a custom mapping basis, and the RDB2RDF mapping conversion tool based on the R2RML standard provides mapping matching based on various similarity algorithms. 3 mappings trigger pairs of mappings by drag, error relationships allow correction and editing. And establishing an R2RML bottom mapping rule base through objectification programming, and mainly generating rules according to different mapping modes. Once the mapping pairs in the pattern are generated, R2RML mapping rule generation is triggered.

In the visual mapping definition view, an RDB2RDF mapping conversion tool based on the R2RML standard provides semantic mapping matching algorithms of Chinese and English data columns and domain ontology to assist a user to complete matching of the names of the relational data columns and domain ontology classes or attributes. This matching is based mainly on column names and class names, attribute names, and column name annotations are provided in the relational data pattern analysis section of the tool, because column name information is limited and the matching degree is low, but semantic matching of column name annotations and ontology annotations is more conducive to mapping matching. The mapping matching similarity algorithm comprises WordNet-based similarity, UMLS-based similarity, ontology-combined similarity, Q-gram character similarity, paragraph vector method and supervised semantic similarity algorithm, and supports Chinese and English matching. The RDB2RDF mapping conversion tool based on the R2RML standard provides ontology class or attribute with high mapping matching degree according to the relation data column name, and gives out different scores of semantic similarity calculation for the user to screen and determine.

The map defines an audit view and the data map pairs are determined by visualizing the map definition view. And then, mapping and defining an audit view to perform data table structure and data browsing, checking the data table and data, and performing semantic enrichment and standardization of RDF triples such as main, predicate and object types and predicates. The mapping definition audit view is formed by performing definition audit of a main object, a predicate object and an object on each triplet generated by each table, modifying and editing contents, and can set predicates and data types, including multi-source RDF vocabulary recommendation and search.

The R2RML edit detection view provides for inspection and custom editing of the generated R2RML mapped document according to the mapping rules. The R2RML editing and detecting view respectively displays different column names of the relational database by using different colors, and if the domain ontology is used, the class or attribute from the ontology is also identified by using different colors so as to remind the user of writing correctness. If misspelled, it cannot be displayed normally. The R2RML editing detection view also divides the R2RML mapping rule into small tripleMaps, each tripleMap is split into source data which can be queried by SQL sentences, a RDB2RDF mapping conversion tool based on the R2RML standard provides a small number of sample queries to test the correctness of data source selection and connection, and the aim is to type or attribute on the domain ontology, and whether the spelling is correct or not through color labeling. In addition, the user grasping the R2RML mapping syntax can make modification and perfection of the mapping rule through the view.

Map editing and detection: according to the R2RML mapping file generation method and device, the R2RML mapping file can be automatically generated according to the mapping definition, the use threshold of a tool can be effectively reduced, and the application range of the tool is enlarged. In addition, the R2RML standard-based RDB2RDF map transformation tool provides an R2RML editor, supporting browsing, editing and modification of R2RML, wherein the R2RML standard-based RDB2RDF map transformation tool supports a multiple data map document inspection structure as shown in FIG. 10. When the R2RML mapping rule is completed, the RDB2RDF mapping conversion tool based on the R2RML standard provides detection of the mapping rule, and ensures smooth operation of the mapping execution process. The mapping rule part is mainly used for quickly running a small amount of data through a mapping effective device of the tool, checking errors in terms of spelling, grammar, logic and the like of the mapping rule, and if a problem occurs, giving out description of the problem and line numbers of the problem, so that a user can conveniently locate, search and modify the problem. Meanwhile, an RDB2RDF mapping conversion tool based on an R2RML standard supports mapping pre-execution, namely pre-calculating the result of R2RML mapping to be executed, wherein the pre-calculation result of the mapping result comprises the number of triples, the number of redundant triples, the number of unique subjects, the number of unique predicates and the number of unique objects; the number of blank nodes comprises the number of subject blank nodes and the number of object blank nodes; the mapping result generation time and the generation time of the RDF triplet file are predicted. RDB2RDF mapping transformation tool based on R2RML standard provides processing mechanism for blank node, including generating no triplet containing blank node, generating no triplet containing subject blank node, generating no triplet containing object blank node, and replacing different blank node with URI according to certain rule for effective distinction. The RDB2RDF map transformation tool based on the R2RML standard also provides a recommendation processing mechanism for the expected redundancy case, and the user can choose to choose the choice of "do not repeatedly generate RDF triples". The pre-calculation of the mapping result enables the user to comprehensively understand the condition of generating the RDF triplet file based on the R2RML standard, and to perform certain quality control, data deletion and trimming and R2RML mapping rule correction according to the statistics and evaluation result.

Mapping implementation: for large data volume data processing, the RDB2RDF mapping transformation tool based on the R2RML standard provided by the application provides two implementation modes of instantiation and on-demand in terms of data mapping transformation, as shown in FIG. 11. The data mapping performance optimizing strategy is used for supporting the generation performance of data in a multithreading mode aiming at large data volume processing when mapping data are generated; before the mapping rule is executed, the RDB2RDF mapping conversion tool based on the R2RML standard judges the pre-generated data volume condition, and whether to call multithreading and set the starting thread number, so that the formation optimization is automatically performed. Mapping result query optimization strategies. In terms of data query, an RDB2RDF mapping conversion tool based on the R2RML standard provides two mapping conversion modes of on-demand mapping and instantiation mapping, and different strategies are adopted to provide data query respectively. The on-demand mapping does not generate the actual RDF triplet file, but rather performs data query by converting SPARQL to SQL. If the amount of the query data is large as required, performance problems can be generated during query statement conversion, and the data query performance is improved mainly by optimizing the written SQL statements. When the instantiated mapping result is accessed, jena is utilized to perform data query, SPARQL is directly used to perform RDF data query, and a method for improving query performance is explored

Evaluation of results: the result evaluation of the RDB2RDF mapping conversion tool based on the R2RML standard mainly refers to the validity check of the W3C on the RDF grammar and the RDF data evaluation of the LUZZ, and combines with some effective applications, thereby formulating the data evaluation index of the RDFRDF triplet file and assisting the user in carrying out the data evaluation and quality control of the RDF triplet file. Statistical evaluation of RDF triples by RDB2RDF mapping transformation tools based on the R2RML standard is largely divided into the following three aspects, and an evaluation report is generated. One aspect is the statistics of the number of result data, including the statistics of the number of triples, such as the total number of triples, the number of unique triples, the number of redundant triples, the number of object URI triples, the number of object value triples, the number of object blank node triples; the three-tuple component count, such as unique subject number, unique predicate number, unique object number, unique URI object number, unique value object number, unique blank node object number; and counting blank nodes, such as the number of blank nodes, the number of subject blank nodes, the number of object blank nodes and the like. Another aspect is the availability of a particular data format. The validity check of the RDF triplet file mainly applies the validity check of W3C to carry out grammar check on the RDF/XML format data, thereby ensuring that the format data can be correctly analyzed. Yet another aspect is a data quality assessment index for an 8-large RDF triplet file, including human-understandable tags, machine-readable licensing agreements, default URIs, extended conciseness, compatible data types, external connectivity, internal connectivity, and lexical richness, for index specific information, see Table 1. And (3) a quality control mechanism of result data: for blank nodes and redundant triples, the RDF triples can be simplified by removing some blank node triples containing redundant information and mapping some blank nodes to specific URI identifiers, such as deleting triples containing blank nodes, deleting triples containing subject blank nodes, deleting triples containing object blank nodes, replacing different blank nodes with URIs according to a certain rule, deleting redundant triples and the like.

Table 1:

/>

mapping task management and sharing: after the mapping is completed, the RDB2RDF mapping conversion tool based on the R2RML standard supports data mapping task management, including R2RML mapping documents, mapping result data and the like, and can support the release and sharing of related resources of the data mapping task to a semantic data mapping and conversion platform.

At present, a plurality of platforms for open sharing of data are available, but no platform and tools for sharing data mapping tasks and mapping documents exist. Although the purposes of data mapping conversion are different for users, the converted source data and the used domain model are different, the experience of the data mapping conversion documents and conversion tasks can provide experience reference for more primary researchers or users with similar data conversion requirements to a certain extent, and the mapping documents can provide mapping learning basis for other people. Therefore, in the invention, it is proposed to construct an RDB2RDF mapping conversion tool and a semantic data mapping conversion and sharing platform based on the R2RML standard, as shown in fig. 12, on one hand, the RDB2RDF mapping conversion tool based on the R2RML standard solves the key problem in the mapping process, assists the user to complete the mapping work based on the R2RML standard, on the other hand, the RDB2RDF mapping conversion tool and the platform based on the R2RML standard interact, provide a common function in the data mapping task through the semantic data mapping conversion and sharing platform, enable the user to upload data, use the relational data pattern analysis, the domain model editing and browsing, the R2RML mapping document editing, the R2RML mapping document detection, the RDF triplet file validity assessment, the RDF triplet file analysis and assessment, the RDF vocabulary recommendation, the RDF triplet file query and the visualization, and other services provided by the platform, and also can enable the mapping task, the mapping document, the domain model and the converted data to be shared to the platform to the maximum extent after completing the mapping task by using the tool, and the related multiplexing result are promoted.

The RDB2RDF mapping conversion tool based on the R2RML standard can automatically analyze the data source mode: the data structure and the semantic relation of the data source are analyzed by the auxiliary user, so that the problem of semantic deletion is reduced, and the semantic association between data is enhanced; the automatic recommendation of predicate functions can be realized, and the method has the functions of modeling and editing field knowledge and multisource RDF vocabulary recommendation service and assists a user in semantic expansion; r2RML mapping documents with corresponding formats can be automatically generated according to different mapping modes, and RDF triple files are automatically generated according to the generated R2RML mapping documents; the method can also realize data open sharing service, promote multiplexing of mapping documents and promote generation of semantic data and data sharing.

Moreover, the application can also automatically generate the R2RML mapping file and the mapping definition view facing different mapping modes. Multiple mapping schema definitions of RDB2RDF based on the R2RML standard may be implemented using a single view or may be combined with different views. Providing multiple semantic similarity algorithms in mapping patterns by means of a domain model provides a mapping of RDBs and domain ontology patterns. The programmatically encoded R2RML mapping rule library is used to store a plurality of mapping rules created to follow the R2RML syntax, incorporating different schema features. After the data mapping relation is defined through the view, the R2RML standard-based RDB2RDF mapping conversion tool can dynamically call the corresponding R2RML mapping rule in the programming rule base, so that the automatic generation of the R2RML mapping document is realized. And establishing manual R2RML mapping rule definition, and improving the understandability and operability of a user on mapping conversion.

The invention also provides various semantic modeling and semantic enriching functions in the mapping conversion process. The method provides the functions of relational data pattern analysis and potential semantic relation discovery, helps users to comprehensively understand own source data, and achieves a mapping target better. The relational data pattern analysis provides full presentation and presentation of explicit structures and semantics of structures, data and association relations in a relational database, prompts are given to implicit structure and semantic auxiliary discovery, and the implicit structure and the semantic are explicitly represented for selection by a user. The method provides functions of recommending, importing, editing, modifying and the like for a domain knowledge model (ontology) mapped by means of a domain model, supports the importing of domain knowledge, and carries out domain knowledge modeling. And providing multi-source RDF vocabulary inquiry and recommendation functions for supporting a custom mapping mode and enriching the domain semantics of the mapping mode by means of a domain model.

The invention also provides a set of multiple data mapping document detection and result data evaluation and control mechanism. By providing a mechanism for grammar and rule detection of the mapping document, the user is helped to check R2RML mapping rule errors. And (3) providing a mapping conversion pre-calculation mechanism and a pre-calculation index, pre-calculating the number of data triples, the number of main, predicate and guest triples, the number of blank nodes and the like by using a small amount of data, pre-judging the result, and providing a processing mechanism for the blank nodes, wherein the processing mechanism does not generate triples containing blank nodes, triples containing subject blank nodes, triples containing object blank nodes, converts the blank nodes into nodes with basic URIs according to rules, and does not repeatedly generate RDF triples files and the like. And evaluating the generated RDF triplet file, wherein the evaluation indexes comprise W3C validity verification, result quantity statistics (including basic data statistics, redundant data statistics and blank nodes) and multi-evaluation index reports (including human-understandable labels, machine-readable permission protocols, default URIs, multi-quality evaluation indexes such as extended conciseness, compatible data types, external connectivity, internal connectivity, vocabulary richness and the like). For redundant data and blank nodes, the tool provides a data processing mechanism. Thus, the quality of the RDF triplet file is ensured to a certain extent.

Referring to fig. 13, an embodiment of the present invention provides a schematic structural diagram of a mapping apparatus based on R2RML standard, where the mapping apparatus from a relational database to a resource description framework includes:

a first determining unit 131, configured to determine a mapping task from the R2RML relational database created by the user to the resource description framework, and connect the relational database indicated by the mapping task;

the analysis unit 132 is configured to perform explicit structural features, data features and invisible semantic analysis on the relational database to obtain an association relationship table of the relational database;

a first receiving unit 133, configured to determine a selection operation of a target mapping mode selected by a user from at least one mapping mode preset, where the at least one mapping mode includes a direct mapping mode, a custom mapping mode, and a data mapping mode by means of a domain model;

a first generating unit 134, configured to perform RDF triplet definition using the target mapping mode and the association table, and automatically generate an R2RML mapping document based on the R2RML mapping rule;

a second generating unit 135 for generating an RDF triplet file from the R2RML mapped document.

The invention provides a mapping device based on R2RML standard, which is characterized in that a mapping task from an R2RML relational database created by a user to a resource description framework is determined, and the relational database indicated by the mapping task is connected; performing explicit structural features, data features and invisible semantic analysis on the relational database to obtain an association relation table of the relational database; determining a target mapping mode selected by a user from at least one preset mapping mode, wherein the at least one mapping mode comprises a direct mapping mode, a custom mapping mode and a mapping mode by means of a domain model; and carrying out RDF triple definition by utilizing the target mapping mode and the association relation table, automatically generating an R2RML mapping document based on an R2RML mapping rule, and further automatically generating an RDF triple file according to the R2RML mapping document, thereby solving the problem that the automatic generation of the RDF triple file cannot be realized due to the fact that the R2RML mapping document cannot be automatically generated in the prior art.

Further, if the target mapping mode is a mapping mode by means of a domain model, the mapping device based on the R2RML standard provided in the embodiment of the present application further includes;

the second determining unit is used for determining a domain ontology of the mapping task, wherein the domain ontology is a domain ontology recommended to a user based on task description and data source description carried by the mapping task, or the domain ontology is a domain ontology imported by the user, or the domain ontology is a domain ontology built by the user on line;

a third determining unit, configured to determine a mapping relationship between a field in a data table of the relational database and a domain ontology class of the domain ontology;

the first generation unit is further used for carrying out RDF triple analysis by utilizing the target mapping mode, the association relation table and the mapping relation, and automatically generating an R2RML mapping document based on the R2RML mapping rule.

In the embodiment of the present application, if the target mapping mode is a custom mapping mode or a mapping mode with the help of a domain, preferably, the first generating unit includes:

the fourth determination unit is used for determining RDF vocabulary, wherein the RDF vocabulary comprises existing RDF vocabulary, custom RDF vocabulary and RDF vocabulary recommended for users;

and the third generation unit is used for generating an R2RML mapping document based on the RDF vocabulary and the R2RML mapping rule by utilizing the target mapping mode, the association relation table and the defined RDF triple structure.

Further, the mapping device based on the R2RML standard provided in the embodiment of the present application further includes;

the detection unit is used for detecting grammar errors of the R2RML mapping document to generate grammar check results of the R2RML mapping document, wherein the grammar check results comprise mapping rule spelling errors, grammar errors, logic errors, and wrong problem descriptions and wrong problem line numbers;

the verification unit is used for performing null value verification on the relational database to obtain a null value verification result of the relational database;

a pre-calculation unit for pre-calculating a result of R2RML mapping to be performed according to the R2RML mapping document to generate a mapping result including a triplet number, a redundant triplet number, a unique subject number, a unique predicate number, and a unique object number, and predicting a mapping result generation time; the number of blank nodes comprises a subject blank node number and an object blank node number;

and the display unit is used for displaying the processing mechanism of the blank node and the processing mechanism of the redundant node.

In the embodiment of the present application, preferably, the second generating unit includes:

and a fourth generating unit, configured to generate an RDF triplet file according to the R2RML mapping document in combination with a setting operation of whether to add the database field annotation and the constraint to the triplet by the user and a selection operation of a processing mechanism of the blank node and a processing mechanism of the redundant node by the user.

the evaluation unit is used for carrying out data evaluation on the RDF triplet file to obtain a result evaluation report of the RDF triplet file, and the result evaluation report comprises: statistics of the number of resulting data, validity of a specific data format, and RDF data quality assessment indicators.

the second receiving unit is used for receiving the SPARQL retrieval request sent by the user, acquiring a query result of the SPARQL retrieval request and returning the query result to the user; the SPARQL retrieval request indicates a virtual retrieval or RDF document retrieval;

if the SPARQL search request indicates virtual search, the second receiving unit is further configured to convert a SPARQL query statement corresponding to the SPARQL search request into an SQL query statement, directly access the relational database according to the SQL query statement to obtain a query result, and convert the query result into an RDF triplet to be returned to the user.

Based on the commonality, the embodiment of the application further provides an apparatus, which includes: the processor and the memory are connected through a communication bus; the processor is used for calling and executing the program stored in the memory; the memory is used for storing a program, and the program is used for realizing the mapping method based on the R2RML standard.

The device provided in the embodiment of the present application may be a terminal or a server, and the mapping method based on the R2RML standard provided in the embodiment of the present application will now be described in detail from the perspective of the server and the terminal, respectively.

For ease of understanding, a mapping method based on the R2RML standard provided in the embodiments of the present application will be described in detail from the perspective of the server. The server may be a service device for providing services for users on the network side, and may be a server cluster formed by a plurality of servers, or may be a single server.

Fig. 14 is a hardware structure block diagram of a server according to an embodiment of the present application. Referring to fig. 14, the hardware structure of the server may include: a processor 141, a communication interface 142, a memory 143 and a communication bus 144;

in the embodiment of the present invention, the number of the processor 141, the communication interface 142, the memory 143, and the communication bus 144 may be at least one, and the processor 141, the communication interface 142, and the memory 143 complete communication with each other through the communication bus 144;

processor 141 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention, etc.;

Memory 143 may comprise high-speed RAM memory, may also comprise non-volatile memory (non-volatile memory) or the like, such as at least one disk memory;

wherein the memory stores a program, and the processor is operable to invoke the program stored in the memory, the program being operable to:

receiving a target mapping mode selection operation of a user in at least one preset mapping mode, wherein the at least one mapping mode comprises a direct mapping mode, a custom mapping mode and a data mapping mode by means of a domain model;

RDF triples are defined by utilizing the target mapping mode and the association relation table, and R2RML mapping documents are automatically generated based on R2RML mapping rules;

RDF triplet files are generated from the R2RML mapped document.

Alternatively, the refinement function and the extension function of the program may be described with reference to the above.

Fig. 15 is a block diagram of a hardware structure of a terminal to which a mapping method based on an R2RML standard is applicable, provided in an embodiment of the present application.

The terminal may include, as shown in fig. 15: processor 151, memory 152, communication interface 153, input unit 154, and display 155 and communication bus 156.

The memory 152 is used to store one or more programs, and the programs may include program code that includes computer operation instructions, and in embodiments of the present invention, at least the programs for implementing the following functions are stored in the memory:

RDF triplet files are generated from the R2RML mapped document.

Alternatively, the refinement function and the extension function of the program can be described with reference to the following.

The processing module 151, the memory 152, the communication interface 153, the input unit 154, the display 155, all complete communication with each other via the communication bus 156.

In an embodiment of the present invention, the processor 151 may be a central processing unit (Central Processing Unit, CPU), an application-specific integrated circuit (ASIC), a Digital Signal Processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic device, etc.

The processor may call and execute programs stored in memory 152.

The communication interface 153 may be an interface of a communication module, such as an interface of a GSM module.

The present invention may also include an input unit 154, which may include a touch sensing unit that senses touch events on the touch display panel, a keyboard, and the like.

The display 155 includes a display panel such as a touch display panel or the like. In one possible case, the display panel may be configured in the form of a liquid crystal display (Liquid Crystal Display, LCD), an Organic Light-Emitting Diode (OLED), or the like.

Of course, the terminal structure shown in fig. 15 is not limited to the terminal in the embodiment of the present invention, and the terminal may include more or less components than those shown in fig. 15 or may combine some components in practical applications.

Further, the embodiment of the application also provides a computer readable storage medium, in which computer executable instructions are stored, and the computer executable instructions are used for executing the mapping method based on the R2RML standard.

For details of the computer executable instructions, reference is made to the above detailed description of a mapping method based on the R2RML standard provided in the embodiments of the present application, which is not repeated here.

The invention provides a mapping method, a device, equipment and a storage medium based on R2RML standard, wherein a mapping task from a relational database based on the R2RML standard created by a user to a resource description framework is determined, and the relational database indicated by the mapping task is connected, if mapping is needed by a domain model, a domain ontology can be imported; performing intelligent pattern analysis on the relational database, wherein the intelligent pattern analysis comprises explicit structural features, data characteristics and implicit semantic analysis to obtain an association relation table of the relational database; determining a target mapping mode selected by a user from at least one preset mapping mode, wherein the at least one mapping mode comprises a direct mapping mode, a custom mapping mode and a data mapping mode by means of a domain model. Under the mapping mode of custom mapping and domain model, the user can multiplex the existing RDF vocabulary and custom RDF vocabulary or the existing RDF vocabulary which is intelligently recommended by the user and matched with the user to help the user to expand and describe semantic relations; RDF triples are defined by utilizing the target mapping mode and the association relation table, and R2RML mapping documents are automatically generated based on R2RML mapping rules; the invention provides an R2RML editor provided by an RDB2RDF mapping conversion tool based on an R2RML standard, which supports browsing, editing and modifying of R2RML mapping documents, and simultaneously supports checking of R2RML grammar, performs mapping pre-execution and counts pre-generated data results; and realizing data mapping by an R2RML processor to generate an RDF triple file, evaluating the generated RDF triple file, and generating an evaluation report.

According to the technical means provided by the invention, by establishing mapping definition views oriented to different mapping modes, various semantic modeling and semantic enrichment functions in the mapping conversion process are provided, the R2RML mapping document is automatically generated, a set of multiple data mapping document detection and result data evaluation and control mechanism is provided, the mapping operation difficulty of a user is reduced to a certain extent, and the operation convenience and the understandability of the user are improved.

The foregoing describes in detail a mapping method, apparatus, device and storage medium based on R2RML standard, and specific examples are applied to illustrate the principles and embodiments of the present invention, and the description of the foregoing examples is only for helping to understand the method and core idea of the present invention; meanwhile, as those skilled in the art will vary in the specific embodiments and application scope according to the idea of the present invention, the present disclosure should not be construed as limiting the present invention in summary.

It should be noted that, in the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described as different from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.

It is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include, or is intended to include, elements inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A mapping method based on R2RML standard, comprising:

generating an RDF triplet file according to the R2RML mapping document;

displaying the processing mechanism of the blank node and the processing mechanism of the redundant node;

if the target mapping mode is a mapping mode by means of a domain model, determining a domain ontology of the mapping task, wherein the domain ontology is a domain ontology recommended to the user based on task description and data source description carried by the mapping task, or the domain ontology is a domain ontology imported by the user, or the domain ontology is a domain ontology built by the user on line;

correspondingly, the RDF triplet definition is performed by using the target mapping mode and the association relation table, and the R2RML mapping document is automatically generated based on the R2RML mapping rule, which comprises the following steps: and carrying out RDF triple analysis by utilizing the target mapping mode, the association relation table and the mapping relation, and automatically generating an R2RML mapping document based on an R2RML mapping rule.

2. The method of claim 1, wherein if the target mapping pattern is a custom mapping pattern or a mapping pattern by means of a domain model, the performing RDF triplet definition using the target mapping pattern and the association table, and automatically generating an R2RML mapping document based on an R2RML mapping rule, comprises:

3. The method of claim 1, wherein the generating an RDF triplet file from the R2 RML-mapped document includes:

4. The method as recited in claim 1, further comprising:

5. The method as recited in claim 1, further comprising:

6. A mapping apparatus based on R2RML standard, comprising:

the first determining unit is used for determining a mapping task from a relational database based on R2RML standard created by a user to a resource description framework and connecting the relational database indicated by the mapping task;

the first receiving unit is used for determining a target mapping mode selected by a user from at least one preset mapping mode, wherein the at least one mapping mode comprises a direct mapping mode, a custom mapping mode and a data mapping mode by means of a domain model;

the second generating unit is used for generating an RDF triplet file according to the R2RML mapping document;

a detection unit, configured to detect a syntax error of the R2RML mapped document to generate a syntax check result of the R2RML mapped document, where the syntax check result includes a mapping rule spelling error, a syntax error, a logic error, and a wrong question description and a wrong question line number;

a pre-calculation unit configured to pre-calculate a result of R2RML mapping performed according to the R2RML mapping document to generate a mapping result and predict a mapping result generation time, the mapping result including a triplet number, a redundant triplet number, a unique subject number, a unique predicate number, and a unique object number; the number of blank nodes comprises a subject blank node number and an object blank node number;

The display unit is used for displaying the processing mechanism of the blank node and the processing mechanism of the redundant node;

the second determining unit is configured to determine, if the target mapping mode is a mapping mode by means of a domain model, a domain ontology of the mapping task, where the domain ontology is a domain ontology recommended to the user based on a task description and a data source description carried by the mapping task, or the domain ontology is a domain ontology imported by the user, or the domain ontology is a domain ontology built by the user online;

correspondingly, the first generating unit is further configured to perform RDF triplet analysis by using the target mapping mode, the association relation table and the mapping relation, and automatically generate an R2RML mapping document based on an R2RML mapping rule.

7. An apparatus, comprising: the device comprises a processor and a memory, wherein the processor and the memory are connected through a communication bus; the processor is used for calling and executing the program stored in the memory; the memory for storing a program for implementing the R2RML standard-based mapping method as claimed in any one of claims 1-5.

8. A computer-readable storage medium having stored therein computer-executable instructions for performing the R2RML standard-based mapping method of any of claims 1-5.