Methods and techniques for management of ontolodgy-based knowledge representation models in the context of BIG data
Ontology-based knowledge representation models in the context of big data are one way to reduce complexity for data processing across methods of semantic description. This research paper aims at providing an overview of the methods and techniques for efficient management of the ontology-based models...
Gespeichert in:
| Datum: | 2022 |
|---|---|
| 1. Verfasser: | |
| Format: | Artikel |
| Sprache: | English |
| Veröffentlicht: |
Інститут програмних систем НАН України
2022
|
| Schlagworte: | |
| Online Zugang: | https://pp.isofts.kiev.ua/index.php/ojs1/article/view/472 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Назва журналу: | Problems in programming |
Institution
Problems in programming| id |
pp_isofts_kiev_ua-article-472 |
|---|---|
| record_format |
ojs |
| resource_txt_mv |
ppisoftskievua/97/ccb5ed34ce3f81b3572ac4cf826de897.pdf |
| spelling |
pp_isofts_kiev_ua-article-4722023-01-19T07:12:43Z Methods and techniques for management of ontolodgy-based knowledge representation models in the context of BIG data Методи та технології управляння моделями представлення знань, базованих на онтологіях в контексті велики х даних Novitsky, A.V. ontology-based model; ontologies; big data; reasoners; representation system; ontologies; shapes constraint language; information validation; ontology-based knowledge representation models UDC 004.6 модель на основі онтології; онтології; великі дані; вивід; система представлення; онтології; мова обмежень форм; перевірка інформації; моделі представлення знань на основі онтологій УДК 004.6 Ontology-based knowledge representation models in the context of big data are one way to reduce complexity for data processing across methods of semantic description. This research paper aims at providing an overview of the methods and techniques for efficient management of the ontology-based models that improve big data systems. For this case, the shapes constraint language (SHACL) for information validation was reviewed as the key method. The knowledge representation systems and reasoners are studied and reviewed in the paper as well. It describes approaches based on ontologies in the context of big data. The proper management of ontology-based knowledge representation models through offered methods and techniques brings improved data integration, big data quality, and business process integration.Prombles in programming 2021; 4: 19-25 Онтологічні моделі представлення знань у контексті великих даних є одним із способів зменшити складність обробки даних за допомогою семантичних методів. У статті розглянуто методи і засоби ефективного управління моделями на основі онтологій, які покращують системи великих даних. Для цього випадку мова вираження обмежень форм (SHACL) для перевірки інформації була розглянута як ключовий метод. У статті також досліджуються та розглядаються представлення знань та методи виводу. Належне управління моделями представлення знань на основі онтології за допомогою запропонованих методів і засобів забезпечує покращену інтеграцію даних, якість великих даних та інтеграцію бізнес-процесів.Prombles in programming 2021; 4: 19-25 Інститут програмних систем НАН України 2022-02-07 Article Article application/pdf https://pp.isofts.kiev.ua/index.php/ojs1/article/view/472 10.15407/pp2021.04.019 PROBLEMS IN PROGRAMMING; No 4 (2021); 19-25 ПРОБЛЕМЫ ПРОГРАММИРОВАНИЯ; No 4 (2021); 19-25 ПРОБЛЕМИ ПРОГРАМУВАННЯ; No 4 (2021); 19-25 1727-4907 10.15407/pp2021.04 en https://pp.isofts.kiev.ua/index.php/ojs1/article/view/472/476 Copyright (c) 2022 PROBLEMS IN PROGRAMMING |
| institution |
Problems in programming |
| baseUrl_str |
https://pp.isofts.kiev.ua/index.php/ojs1/oai |
| datestamp_date |
2023-01-19T07:12:43Z |
| collection |
OJS |
| language |
English |
| topic |
ontology-based model ontologies big data reasoners representation system ontologies shapes constraint language information validation ontology-based knowledge representation models UDC 004.6 |
| spellingShingle |
ontology-based model ontologies big data reasoners representation system ontologies shapes constraint language information validation ontology-based knowledge representation models UDC 004.6 Novitsky, A.V. Methods and techniques for management of ontolodgy-based knowledge representation models in the context of BIG data |
| topic_facet |
ontology-based model ontologies big data reasoners representation system ontologies shapes constraint language information validation ontology-based knowledge representation models UDC 004.6 модель на основі онтології онтології великі дані вивід система представлення онтології мова обмежень форм перевірка інформації моделі представлення знань на основі онтологій УДК 004.6 |
| format |
Article |
| author |
Novitsky, A.V. |
| author_facet |
Novitsky, A.V. |
| author_sort |
Novitsky, A.V. |
| title |
Methods and techniques for management of ontolodgy-based knowledge representation models in the context of BIG data |
| title_short |
Methods and techniques for management of ontolodgy-based knowledge representation models in the context of BIG data |
| title_full |
Methods and techniques for management of ontolodgy-based knowledge representation models in the context of BIG data |
| title_fullStr |
Methods and techniques for management of ontolodgy-based knowledge representation models in the context of BIG data |
| title_full_unstemmed |
Methods and techniques for management of ontolodgy-based knowledge representation models in the context of BIG data |
| title_sort |
methods and techniques for management of ontolodgy-based knowledge representation models in the context of big data |
| title_alt |
Методи та технології управляння моделями представлення знань, базованих на онтологіях в контексті велики х даних |
| description |
Ontology-based knowledge representation models in the context of big data are one way to reduce complexity for data processing across methods of semantic description. This research paper aims at providing an overview of the methods and techniques for efficient management of the ontology-based models that improve big data systems. For this case, the shapes constraint language (SHACL) for information validation was reviewed as the key method. The knowledge representation systems and reasoners are studied and reviewed in the paper as well. It describes approaches based on ontologies in the context of big data. The proper management of ontology-based knowledge representation models through offered methods and techniques brings improved data integration, big data quality, and business process integration.Prombles in programming 2021; 4: 19-25 |
| publisher |
Інститут програмних систем НАН України |
| publishDate |
2022 |
| url |
https://pp.isofts.kiev.ua/index.php/ojs1/article/view/472 |
| work_keys_str_mv |
AT novitskyav methodsandtechniquesformanagementofontolodgybasedknowledgerepresentationmodelsinthecontextofbigdata AT novitskyav metoditatehnologííupravlânnâmodelâmipredstavlennâznanʹbazovanihnaontologíâhvkontekstívelikihdanih |
| first_indexed |
2025-07-17T09:35:35Z |
| last_indexed |
2025-07-17T09:35:35Z |
| _version_ |
1838499826879168512 |
| fulltext |
19
Моделі та засоби систем баз даних і знань
Introduction
Big data means complex data sets that
are unable to process adequately via tradi-
tional data applications. Big data manage-
ment is handled by special-purpose resource
planning systems called enterprise informa-
tion systems. These systems represent busi-
ness processes adequately and force the over-
all cost-eff ectiveness [1]. Modern enterprises
are focused on the enterprise-wide centralized
information system to validate and integrate
large amounts of complex data. To capture
and represent complex and big data, ontology-
based knowledge representation models are
used. One of the factors which impact big data
processing is the complexity of understanding
the data. Semantic technology is allowed to
automatically recognize data. This article is to
explain the approach of using ontologies for
big data to ensure a common understanding of
information. The ontology-based knowledge
representation models make explicit domain
assumptions [2].
Querying information in the context of
big data becomes accessible for large enter-
prises. Ontologies bring detailed and meaning-
ful distinctions between relationships, classes,
and properties. The paper is devoted to on-
tology-based modeling and its management
in semantic graph databases. Big data quality
is improved with the help of ontology-based
knowledge representation models and the rea-
soners that enable consistency and satisfi abil-
ity checks [3].
The research paper also reviews an
alternative using ontologies to model data.
SHACL (shapes constraint language) is over-
viewed to demonstrate the benefi ts of this
method for information validation in the
triplestore and for validating RDF graphs
against a set of constraints. The overview of
the OWL reasoners and RDF graph capture
systems is used as the guide for big data play-
ers (large enterprises, structure that is in the
stage of developing a large-scale centralized
database) on how to manage ontology-based
models and improve the data quality with the
help of automated reasoning of the informa-
tion in the semantic graph database.
OWL Reasoners for ontologies.
The research paper reviews the main
two reasoners with the wide range of op-
timizations that benefit big data improve-
ments. They contain updated algorithms
and tableaux algorithms that are native to
the ontology-based knowledge representa-
tion models.
FaCT++ is one of the newest reason-
ers that is designed to implement tableaux al-
gorithms and updated heuristic optimization
techniques. The table of characteristics of the
FaCT++ reasoner is given below.
УДК004.6 http://doi.org/10.15407/pp2021.04.019
O. Novytskyi
METHODS AND TECHNIQUES FOR MANAGEMENT
OF ONTOLOGY-BASED KNOWLEDGE REPRESENTATION
MODELS IN THE CONTEXT OF BIG DATA
Ontology-based knowledge representation models in the context of big data are one way to reduce
complexity for data processing across methods of semantic description. This research paper aims to
provide an overview of the methods and techniques for efficient management of the ontology-based models
that improve big data systems. For this case, the shapes constraint language (SHACL) for information
validation was reviewed as the key method. The knowledge representation systems and reasoners are
studied and reviewed in the paper as well. The author describes approaches based on ontologies in
the context of big data. The proper management of ontology-based knowledge representation models
through offered methods and techniques brings improved data integration, big data quality, and business
process integration.
Key words: ontology-based model, ontologies, big data, reasoners, representation system, ontologies,
shapes constraint language, information validation, ontology-based knowledge representation models.
© O. Novytskyi, 2021
ISSN 1727-4907. Проблеми програмування. 2021. №4
20
Моделі та засоби систем баз даних і знань
Description
A new highly-optimized reasoner
with tableaux-based SROIQ
algorithms
License LGPL v2
Semantics
OWL DL Classifi cation
OWL EL Classifi cation
OWL DL Consistency
OWL EL Consistency
OWL DL Realization
OWL EL Realization
Table 1. FaCT++ Characteristics.
Source: ORE
The FaCT++ reasoner implementation
starts with the preprocessing stage. It is ap-
plied to the knowledgebase and can be trans-
formed according to the internal representation
requirements. FaCT++ performs classifi cation.
With the help of applied optimizations, the
FaCT++ reasoner is used to reduce the quantity
of subsumption tests to be performed [4].
The main application of the FaCT++ op-
timizations is to transform concepts into SNF.
The simplifi ed normal form lets users imple-
ment negation, conjunction, universal restric-
tion, at-most restrictions. The main FaCT++
features for big data optimization are:
(a) Absorption – is suitable for rewriting
optimization. There are concept and role ab-
sorption techniques to take into consideration.
The concept absorption is responsible for GCIs
elimination via concept defi nition axioms. The
role absorption eliminates GCIs in the concept-
free mode.
(b) TCE (Told Cycle Elimination) – the
technique for text optimization. This cycle is
often eliminated together with defi nitional
cycles. The user can undertake TCE and defi -
nitional cycles with the help of axiom transfor-
mations.
(c) Synonym Replacement – this
FaCT++ technique aims at extending simpli-
fi cation properties. Synonym Replacement im-
proves clash detection in the early stage. The
knowledgebase is transformed in the context of
synonym elimination with the help of axioms.
The FaCT++ reasoner is used for satis-
fi ability checking optimizations. New ordering
heuristics are available for the implementa-
tion of new optimization methods. There is a
special-purpose To-Do list. The user can force
entry assortment with the help of the FaCT++
To-Do algorithm. It is worth noting that the rea-
soner provides the Backjumping optimization.
The tree label matters when the dependency set
of information items is formed. Boolean opti-
mization that is available with the help of the
FaCT++ reasoner allows users to implement
constant propagation (BCP) [5].
HermiT is the reasoner for ontology-
based knowledge representative models. It is
used for the identifi cation of subsumption rela-
tionships (between classes and other specifi ca-
tions). This reasoner is public and available for
the users without restrictions. The notable fea-
ture of the HermiT reasoner is its new versions
with the updated reasoning algorithms [5], [7].
The main HermiT characteristics are
given in the table below [8]
Description
The conformant reasoner for
the ontology-based knowledge
representation models.
The HermiT uses direct semantics.
It is based on the hyper-tableaux
algorithms.
License LGPL 3.0
Semantics
OWL DL Classifi cation, OWL
EL Classifi cation
OWL DL Consistency
OWL EL Consistency
OWL DL Realization
OWL EL Realization
Table 2. HermiT Characteristics
The HermiT reasoner allows users to
classify the ontology-based knowledge repre-
sentation models faster. The manual classifi ca-
tion often takes hours. The reasoner makes it
possible to classify even big data knowledge-
base and complex information for minutes.
The HermiT reasoner uses direct se-
mantics for optimization processes and hyper-
tableaux algorithm implementation. The last
version of the reasoner is called HermiT 1.3.8.
Besides the main function of DL Safe rule han-
dling, the new version of reasoner allows big
data players to add new rules directly to the
ontology-based models [6].
The number of optimization techniques
of the HermiT reasoner is similar to the FaCT++
one described in the research paper above. The
21
Моделі та засоби систем баз даних і знань
signifi cant feature of the HermiT reasoner is
the high-level DL Safety rules compliance. DL
Safety rules will be considered incomplete if:
a) the knowledgebase contains property
chains in the rule bodies;
b) the KB includes transitivity axioms
in the rule bodies of the knowledgebase;
c) the complex properties are used in
the rule bodies of the ontology-based model.
The HermiT reasoner is one of the new-
est reasoners that is recommended for ontol-
ogy-based knowledge representation model
management in the context of big data. The
direct semantics use and hyper-tableaux algo-
rithm approach improve the quality of data and
simplify business processes related to ontolo-
gies and semantics.
RacerPRO is the improved version
of the former Racer knowledge representa-
tion system. As above-described reasoners and
other programs suitable for ontology-based
knowledge representation model management,
RacerPRO is used for optimized tableaux algo-
rithm implementation. The description logic of
is used for this knowledge repre-
sentation system.
Description
Racer is a knowledge
representation system that
implements a highly optimized
tableau calculus for a very
expressive description logic. It
provides the reasoning for T-boxes
and A-boxes as well.
License -
Semantics
OWL DL Classifi cation, OWL
EL Classifi cation
OWL DL Consistency
OWL EL Consistency
OWL DL Realization
OWL EL Realization
Table 3. HermiT Characteristics
The RacerPRO license is BSD 3-clause.
This system is required for big data projects
because it is the separate-standing knowledge
representation system for solving main reason-
ing problems [9].
The reasoning procedure takes place
in the streaming model that is suitable for
complex data proceedings. Both T-boxes and
A-boxes often include issues to solve when it
comes to knowledge representation. RacerPRO
solves these reasoning problems with the help
of standard tableaux algorithms and unique
interference services (e.g. logical abduction).
The architecture of the latest version of the
RacerPRO system is presented in Figure 1.
RacerPRO
PLUGINS
RDF
RDF-S
OWL-Lite
OWL-DL
Racer Exten on
nRQL
SWRL
SPARQL
AllegroGr
aph
OWL
API
OWL
Link
JRacer/
LRacer
TCP/IP
Figure 1. RacerPRO Architecture.
The additional benefi t of the RacerPRO
reasoning and knowledge representation sys-
tem is its query language called nRQL. Using
the new Racer Query Language means supple-
mentary assistance when it comes to ontology-
based model management:
attribute values of diff erent individuals;
improved propertied for string attri-
butes;
negation-as-failure support.
Reasoning over ontology as a rule is a
complex task. In real tasks for reasoning, we
have to store facts in RDF. Therefore, in the
next part, we make a short review semantic
reasoner with RDF storage.
Snorocket. This is the special-purpose
algorithm based on the healthcare terminology
classifi ers. Snorocket will be suitable for big
data projects related to the clinical, medical,
healthcare, science directions [10]. Snorocket
is not a multifunctional solution for ontology-
based knowledge representation model man-
agement. This is suitable for working with on-
tology related to medical data only.
Snorocket is available for users in
the extension format. The classifi ers of the
algorithm allow healthcare representatives
to manage semantic data related to medi-
22
Моделі та засоби систем баз даних і знань
cal terminology. Big data projects based
on healthcare or medical content, imagery,
and other information can benefi t from us-
ing Snorocket. Nevertheless, this extension
with the implementation of the unique Dres-
den algorithm is not suitable for any other
knowledgebase. The limited ways of the ap-
plication make Snorocket the last RDF store
system in the list of top ones overviewed in
the research paper.
Methods for RDF graphs validation.
Shapes Constraint Language (SHACL)
RDF is a main part of the Semantic
Web. Its simple data model provides power-
ful expressiveness which can be applied to
represent information in any scope. Practi-
cal Semantic Web applications require some
technology to describe and validate the RDF
data [11]. One of such technology for RDF
is SHACL [12], [13], which has developed to
model some restrictions in the form of con-
straints on data.
The shapes constraint language (SHA-
CL) is considered as the alternative to tradi-
tional ontologies that are used for data mod-
eling. SHACL is used for RDF graphs vali-
dation. There is a set of constraints that are
applicable to the validation process. SHACL
includes shapes that specify metadata accord-
ing to its resource. The big data knowledge-
base is compliable with the shapes constraint
language. The special-purpose shape specifi es
the resource in the context of big data as well.
This resource can be the principle of data use,
the reason for data use, and the frequency of
data use. The SHACL data validation process
is applicable for both unavailable and available
data in the triplestore. The shape constraint lan-
guage conditions are called shapes expressed
in the RDF graph format. The main purpose of
the SHACL data validation is to check infor-
mation according to the range of conditions.
Those pieces of data that meet the shape con-
straints can be viewed as a description of data
graphs. It is worth noting that SHACL-gener-
ated descriptions based on the shape constraint
language validation of graphs can be used out
of the validation process [12].
It makes SHACL the key method for
ontology-based knowledge representation
model management. The ready-done descrip-
tions with the help of shape constraint language
validation algorithms can be implemented in
the context of big data:
● for code generation;
● for data generation.
These descriptions are suitable for code
building that is one more technique out of the
validation process. The separate-standing aspect
to take into account is the relationship between
SHACL and RDFs inferencing. The shape con-
straint language includes the property entail-
ment to identify the interference specifi cations.
To protect the knowledgebase items and bring
a smooth validation process, it is recommended
to use only verifi ed RDF resources to proceed in
SHACL RDF-based technologies [13].
The SHACL validation is recom-
mended for big data because, in comparison
with the standard ontologies and semantic
techniques, this is the effi cient way to avoid
ontology limitations (limited set of property
constructs). The RDF resource validation is
suitable for ontology-based knowledge repre-
sentation models in the context of big data for
its shape-generated failure determination and
data improvement properties.
Reasoners with built-in RDF
store features
The range of special-purpose databases
for graphs that store triples is called the RDF
database. It is worth noting that triples or RDF
databases are considered as data points. These
points are represented in the SPO relationship
(subject-predicative-object relationship). All
the data items are stored in the same format – a
triple format. The database receives and uses
information and stores it in triple form. The
RDF database is suitable for ontology-based
knowledge representation management in the
context of big data because all the complex
information is well-organized with the help of
triple sets.
One more reason to use the RDF data-
base as the ontology-based model management
when it comes to big data is the convenience
to display information in graphs provided by
this type of database. To carve the graphs from
the triple database, any query language is used.
The functionality and fl exibility of the RDF
database benefi t enterprise-centric knowledge-
base and big data projects.
23
Моделі та засоби систем баз даних і знань
Not all the databases can be included
in the category of triple ones. There is a range
of requirements for the digital product to be
named as the RDF database. The main features
of the triple database to potential-to-inclusion
products are:
a) suffi cient data storage is provided;
b) data as recorded as triples;
c) users are allowed to retrieve the data
with the help of query language.
These are features of average RDF
store. For the purpose to determine the most
effi cient triple databases for ontology-based
knowledge representation model management.
HyLAR is the special-purpose rea-
soner for ontology-based knowledge repre-
sentation models that contains RDF-based
libraries. These libraries obtain a wide range
of functionality for ontology-based model
management. The HyLAR reasoner can be
considered as the supplementary reasoning
engine for big data. Its rdfstore.js, SPAR-
QLs, and RDF-ext libraries are used as the
triple databases [14].
The HyLAR reasoner is available in
three versions to implement for the knowl-
edgebase:
a) NPM module
b) A server-based solution
c) Browser version.
The HyLAR reasoner with its RFD li-
braries supports business database rules. This
is one more reason to use the HyLAR-based
database for big data projects. The database
processing generated by the HyLAR reasoner
and its RDF-based libraries is presented in the
infographics below [14].
HyLAR
Reasoning Engine
GreedyIncremental
Rules
OWL Parser SPARQL Parser
PARSER
Owl file Query
Result
RDF triplets
Knowledge base
Fig 2. HyLAR Architecture.
HyLAR is used as the reasoning engine
combined with the OWL and SPARQL pars-
ers. The reasoner brings results in the format
of triples that is well seen on Fig 2. The knowl-
edgebase with ready-done available triples can
be applied together with the HyLAR reasoning
engine for conversion and creation of enter-
prise-centralized big data projects with quali-
tative and checked information.
Apache Jena. The Jena open-source
Java framework includes a special-purpose
RDF API. Jena contains information only in the
format of RDF triples. The collection of RDF
triples forms the general database and is in-
cluded in the Jena data structure called Model.
Jena is optimal for ontology-based knowledge
representation model management because the
data structure model of this framework eas-
ily determines PDF graphs and provides them
relations. The database is well-structural and
easy-to-navigate which is required for big and
complex data [15].
The way on how the relationships go in
the one-direction mode through the triple cod-
ing exemplifi ed in Fig 3 [16].
Fuseki
Applica on code
HTTP
Applica on code
RDF API
Ontology API SPARQL API
Parser
RDF/XML
Turtle
N-triples
RDFa
Inference API
None Build-in reasoner External reasoner
Store API
In-memory SDB TDB Custom
SQL database Tuple store
Fig 3. Jena Architecture
Another benefi t of Jena in the context
of big data is the availability of both RDF and
Ontology APIs. The distinct concepts of the
framework with the RDF-based triple collec-
tion are its opportunity to build up direct rela-
tions between graphs (nodes) in the structure,
rich-in-functions APIs that provide suffi cient
management tools for the ontology-based
knowledge representation models, and big data
24
Моделі та засоби систем баз даних і знань
orientation with Jena’s in-memory structures
in combination with intended methods of the
complex data simplifi cation.
RDFox is the semantic reasoning en-
gine with the functionality of the RDF triple
store. This one of the core systems for big data
with its unique conception of shared memory
parallel reasoning [17].
RDFox is notable with memory-eco-
nomical properties that are suitable for enter-
prise-centric knowledgebase and big data proj-
ects. About 1.5 billion triples can be stored in
50 GB of the RDFox RDF store. The following
table presents the main characteristics of the
RDFox reasoning engine.
Description
The latest version of the former
RDFox semantic reasoning
engine. The latest version was
launched in 2021. It contains
a triple store that is suitable
for knowledge representation
purposes.
Additional
Features
Rule reasoner
OWL reasoner
RDFS Reasoner
Semantics
RDF
OWL
SPARQL
Table 4. RDFox Characteristic.
The ontology-based knowledge repre-
sentation model management in the context
of big data can be undertaken through the
RDFox semantic reasoning engine. The key
benefi t of the system is its triple store with
memory-effi cient stock. Additionally, big
data projects can benefi t from RDFox named
graphs, Data-log extensions, and incremen-
tal update & aggregation.
Conclusions
The ontology-based knowledge repre-
sentation models bring strong benefi ts to the
digital world. The big data sphere is not the
exception. Essential relationships between
concepts automated data reasoning, and se-
mantic advantages are key benefi ts of the
ontology-based models for big data projects.
But ontologies require proper management
when it comes to accurate knowledge repre-
sentation. The current research paper faces
the problematics of the poor ontology-based
knowledge representation model management
in the context of big data.
The most top-ranking systems, rea-
soners, and other digital solutions were
overview by the author. The best ones in
the article are described in the research pa-
per. Besides theoretical information given
in the general paragraphs about reasoners,
shape constraint language, and triple data-
base, there is an analytical background for
each reviewed solution. Pros & cons are
presented under the description of the rea-
soners and reasoning engines. According to
the undertaken research, the ontology-based
knowledge representation models in the con-
text of big data can be easily managed by
the reasoning engines, reasoner extensions,
query languages, hyper-tableaux algorithms,
SHACL implementation, RDF database us-
age. The future prospects of digital transfor-
mation and new technique and method de-
velopment with a focus on big data are real.
The ontology-based knowledge represen-
tation models are successfully managed by
the digital solutions now. The huge progress
in the big data-driven direction is predicted
over the coming decade.
Reference
[1] B. Mouad, «An Evaluation and Compara-
tive study of massive RDF Data manage-
ment approaches based on Big Data Tech-
nologies,» International Journal of Emerg-
ing Trends in Engineering Research, т. 7,
pp. 48-53, 2019.
[2] Y. Sure-Vetter, S. Staab та R. Studer, «Meth-
odology for Development and Employment of
Ontology Based Knowledge Management Ap-
plications.,» ACM SIGMOD Record 3, т. 4, №
31, pp. 18-23, 2002.
[3] P. Haase та L. Stojanovic, «Consistent Evolu-
tion of OWL Ontologies,» The Semantic Web:
Research and Applications, pp. 182-197, 2005.
[4] T. Dmitry , «Incremental and Persistent Rea-
soning in FaCT++,» в ORE, 2014.
[5] T. Dmitry та H. Ian , «FaCT++ description
logic reasoner: System description,» в Interna-
tional joint conference on automated reason-
ing, Berlin, 2006.
[6] R. Shearer, B. Motik та I. Horrocks, «HermiT:
A Highly-Effi cient OWL Reasoner,» Owled, т.
432, p. 91, 2008.
25
Моделі та засоби систем баз даних і знань
[7] Data and knowlege group. University of Ox-
ford., «HermiT OWL Reasoner,» [online].
Available: http://www.hermit-reasoner.com/.
[Date: 05 2021].
[8] D. Michel , G. Birte , G. Rafael , H. Matthew,
J.-R. Ermesto, N. Matentzoglu та P. Bijan ,
«ORE Live Competition,» 05 2021. [online].
Available: http://dl.kr.org/ore2015/vip.cs.man.
ac.uk_8008/reasoners.html.
[9] V. Haarslev, «The RacerPro knowledge rep-
resentation and reasoning system.,» Semantic
Web, т. 3, № 3, pp. 267-277, 2012.
[10] M. J. Lawley та C. Bousquet, «Fast clas-
sifi cation in Protégé: Snorocket as an OWL
2 EL reasoner.,» в 6th Australasian Ontol-
ogy Workshop (IAOA’10). Conferences in
Research and Practice in Information Tech-
nology, 2010.
[11] G. Jose Emilio Labra, «Validating and De-
scribing Linked Data Portals using RDF Shape
Expressions.,» в LDQ@ SEMANTICS, 2014.
[12] J. Corman, J. L. Reutter та O. Savković, «Se-
mantics and validation of recursive SHACL,»
в International Semantic Web Conference,
Cham, 2018.
[13] W3C, «Shapes Constraint Language (SHA-
CL),» 20 July 2017. [online]. Available: https://
www.w3.org/TR/shacl/. [Date: 05 2021].
[14] M. Terdjimi, M. Lionel та M. Mrissa, «Hylar:
Hybrid location-agnostic reasoning.,» в ESWC
Developers Workshop 2015, 2015.
[15] A. Ameen, K. Ur Rahman Khan та R. B.
Padmaja, «Reasoning in semantic web using
Jena.,» Computer Engineering and Intelligent
Systems, т. 5, № 4, pp. 39-47, 2014.
[16] Apache Software Foundation, «Jena archi-
tecture overview,» [online]. Available: https://
jena.apache.org/about_jena/architecture.html.
[Date: 05 2021].
[17] Oxford Semantic Technologies, «RDFox,»
[online]. Available: https://www.oxfordseman-
tic.tech/product. [Date: 05 2021].
Received: 27.10.2021
About author:
Oleksandr Novytskyi,
PhD, Researcher.
Number of scientifi c publications
in Ukrainian journals – 13.
https://orcid.org/0000-0002-9955-7882.
Affi liation:
Інститут програмних систем
НАН України,
проспект Академіка Глушкова, 40.
Тел.: 526 5139
E-mail: alex.googl@gmail.com
|