Creation of Ontological Knowledge Bases in the Semantic Web …
211
By the formation of ontological knowledge bases is meant the implementation of
a complex of works on converting the contents of knowledge sources into an objectoriented or property-centric representation of ODR objects and their instances—
components of ontological knowledge bases.
An analysis of articles [19–21] shows that many approaches to the formation of
OKB are based on structural analysis methods, some of which are based on the use
of top-level ontologies.
So, an approach to the formation of OKB from existing data warehouses is
described by analyzing the ER-diagrams of these data warehouses and constructing a
mapping of the relational schema to an object based on top-level ontologies and workflow technologies. A feature of this approach is the need to prepare a user description
of the organization scheme for the content of complexly structured sources, which
complicates the practical implementation of OKB generation tools built on the basis
of this approach.
The study of the types of SKTS received by municipal services from legal entities and the methods for presenting these SKTS in modern information-analytical
systems made it possible to develop a unified SKTS layout for the presentation of
reporting documents and their storage in the information storage. The structures of
database tables, data being analyzed, and ways of interconnecting with system metadata are also defined. The necessity of creating a base of used terms and structures
for organizing the analysis of SKTS unified layouts is noted.
An overview of the main problems that arise when organizing the extraction of
data from spreadsheet documents in MS Excel format for the intellectual analysis
and formation of ODR objects is given in [19]. The main problem, according to the
authors of this article, is that the structure of user files in Excel format is difficult to
formalize, which leads to the need for reconciliation, cleaning and quality control of
the extracted data with the help of a human specialist. It is also noted that the resource
consumption of the processes of forming ODR objects and their subsequent storage
in the data warehouse is directly proportional to the complexity of the structure
of the SSK (structured source of knowledge) and exponentially depends on their
number. As a solution to the problem, we propose fixing the used source structures
in a centralized meta-database of control information.
Problems arising in the structuring of data extracted from SSK based on the
analysis of ODR ontological models for organizing further software processing were
considered in [20]. In addition, the problem of establishing the boundaries of the
terminological representation of ODR objects during SSK analysis is noted, which
impedes the creation of tools for automatic formation of OKB.
An approach to the semantic coordination of SSK data in knowledge management
systems, which significantly determine the quality of the conducted mining using
Data Mining methods, is given in [5]. It is noted that it is impossible to coordinate the
components of the structure and meta-information of the source with the ontological
models of subject areas used in knowledge management systems without human
intervention (automatically).
211
By the formation of ontological knowledge bases is meant the implementation of
a complex of works on converting the contents of knowledge sources into an objectoriented or property-centric representation of ODR objects and their instances—
components of ontological knowledge bases.
An analysis of articles [19–21] shows that many approaches to the formation of
OKB are based on structural analysis methods, some of which are based on the use
of top-level ontologies.
So, an approach to the formation of OKB from existing data warehouses is
described by analyzing the ER-diagrams of these data warehouses and constructing a
mapping of the relational schema to an object based on top-level ontologies and workflow technologies. A feature of this approach is the need to prepare a user description
of the organization scheme for the content of complexly structured sources, which
complicates the practical implementation of OKB generation tools built on the basis
of this approach.
The study of the types of SKTS received by municipal services from legal entities and the methods for presenting these SKTS in modern information-analytical
systems made it possible to develop a unified SKTS layout for the presentation of
reporting documents and their storage in the information storage. The structures of
database tables, data being analyzed, and ways of interconnecting with system metadata are also defined. The necessity of creating a base of used terms and structures
for organizing the analysis of SKTS unified layouts is noted.
An overview of the main problems that arise when organizing the extraction of
data from spreadsheet documents in MS Excel format for the intellectual analysis
and formation of ODR objects is given in [19]. The main problem, according to the
authors of this article, is that the structure of user files in Excel format is difficult to
formalize, which leads to the need for reconciliation, cleaning and quality control of
the extracted data with the help of a human specialist. It is also noted that the resource
consumption of the processes of forming ODR objects and their subsequent storage
in the data warehouse is directly proportional to the complexity of the structure
of the SSK (structured source of knowledge) and exponentially depends on their
number. As a solution to the problem, we propose fixing the used source structures
in a centralized meta-database of control information.
Problems arising in the structuring of data extracted from SSK based on the
analysis of ODR ontological models for organizing further software processing were
considered in [20]. In addition, the problem of establishing the boundaries of the
terminological representation of ODR objects during SSK analysis is noted, which
impedes the creation of tools for automatic formation of OKB.
An approach to the semantic coordination of SSK data in knowledge management
systems, which significantly determine the quality of the conducted mining using
Data Mining methods, is given in [5]. It is noted that it is impossible to coordinate the
components of the structure and meta-information of the source with the ontological
models of subject areas used in knowledge management systems without human
intervention (automatically).