=Paper=
{{Paper
|id=Vol-2353/paper43
|storemode=property
|title=A Structure of Semantic Service in a Distributed Knowledge Based System
|pdfUrl=https://ceur-ws.org/Vol-2353/paper42.pdf
|volume=Vol-2353
|authors=Nataliia Kulykovska,Artur Timenko
|dblpUrl=https://dblp.org/rec/conf/cmis/KulykovskaT19
}}
==A Structure of Semantic Service in a Distributed Knowledge Based System==
https://ceur-ws.org/Vol-2353/paper42.pdf
A Structure of Semantic Service in a Distributed
Knowledge Based System
Nataliia Kulykovska1[0000-0003-4691-5102], Artur Timenko2[0000-0002-7871-4543]
1
Zaporizhzhia National Technical University, Zhukovsky str., 64,Zaporizhzhia, 69063, Ukraine
natalya.gontar@gmail.com
2
Zaporizhzhia National Technical University, Zhukovsky str., 64,Zaporizhzhia, 69063, Ukraine
timenko.artur@gmail.com
Abstract. The main difference in distributed systems based on knowledge is the
use of the service approach and ontologies in knowledge engineering. Semantic
service acts as a software agent that regulates the interaction of all components
of the system. Semantic service provides: semantic principle, which defines a
formal description of information and allows you to define the following char-
acteristics of services: scalability, semantic interoperability, formal models of
services and ontologies; the principle of decision making; the principle of dis-
tribution, which allows you to aggregate the capabilities of several computing
objects through cooperation. The communication and collaboration mechanism
of the distributed systems based on knowledge covers the message transfer be-
tween services, information exchange between semantic service and distributed
knowledge bases, and the semantic service behaviors in the entire process from
request submission, request handle, to result return. The knowledge base is
formed from three types of ontologies. The rules for recognizing performance
problems are converted into ontology and the system knowledge base is placed.
The working memory of a semantic service contains facts that correspond to
simple events, composite events, and identified problems and recommenda-
tions. The algorithm for using semantic service is discussed in the article. To
assess the performance of the SS attached, create an information system of two
components: event trace generator and event trace analyzer.
Keywords: distributed system, semantic service, components of a distributed
system, knowledge engineering, data, ontology, services, semantic web, struc-
ture of semantic service
1 Introduction
Economic globalization and global networks have become the trend of world devel-
opment. Especially with the development of information technology and the innova-
tion of management theory, modern enterprises need to go beyond the boundaries of
traditional companies to achieve for a rapid and effective integration of resources on a
global scale, thus can produce high-quality products that meet the needs of the market
or provide services the customer need in a very short period of time [1].
In this case, a new distributed computer system in terms of development of modern
technologies put the emphasis on the properties of interoperability and scalability.
This direction is connected with the rapid growth of blockchain technology and the
Internet of things (IoT). As analysts predict that, by 2025, the share of block chain
apps will account for 10% of world gross domestic product [2]. By 2020 will create
more than 30 billion IoT devices. IoT will affect every industry, from retail to health
care [3].
At the moment in the literature there are a large number of definitions of the con-
cept of "distributed system". The most comprehensive definition proposed by AS
Tanenbaum [4]: “Distributed system (DS) is a set of independent computers, which is
perceived by its users as the only consistent system.” Another definition is proposed
in [5]: DS are software and hardware systems, in which execution of operations (ac-
tions, calculations) necessary to ensure the target functionality of the system is dis-
tributed (physically or logically) between different performers. In the computing field,
under our computer, in our study, we will understand the software and hardware sys-
tem created for a specific practical application, the functionality of which is distrib-
uted on various nodes.
On the one hand, DS are tools that allow solving a large number of complex tasks,
most of which are not solved by other methods. DS can eliminate the main drawback
of centralized systems - the limitations of increasing computing power. At the same
time, the analysis revealed a number of problems that need to be solved. The limited
use of DS is complicated by the use of equipment from different manufacturers with
different types of architectures. Due to the wide variety of aspects of building com-
puting systems, as well as the variety of existing operating systems, it becomes neces-
sary to create methods for adaptive planning of distribution of flows in a distributed
system, which will significantly speed up the processing of incoming requests for
services and increase the overall system performance.
Modern representations of data are changing the ways and forms of communica-
tion, production and consumption of information [6]. The dominance of horizontal
relations, structure-role of information decentralization of all types of data available at
any time on every device. The user should not care about the specific technology used
to provide computing capacity or data storage, so you can say that a user has some
information about the remote resource. Distributed systems based on knowledge
(DKBS) in order to study the data, their processing and use evolving technologies of
the semantic web.
The formal model of the DKBS consists of a set of ontologies; lots of services; a
set of events that describe the processes of the system; semantic service (SS); a set of
composite services and knowledge base. The completeness and effectiveness of the
system is determined by a multitude of ontologies. The main difference in DKBS is
the use of the service approach and ontologies in knowledge engineering. SS acts as a
software agent that regulates the interaction of all components of the system.
2 Literature review
Software agents originally were discussed in the 70’s, and in the mid 90’s briefly
gained some momentum but then stalled. The ”software agent” term has found its
way into a number of technologies and has been widely used, for example, in artificial
intelligence, databases, operating systems and computer networks literature. Although
there is no single definition of an agent [7, 8, 9] all definitions agree that an agent is
essentially a special software component that has autonomy that provides an interop-
erable interface to an arbitrary system and/or behaves like a human agent, working for
some clients in pursuit of its own agenda. Most discussions on agents focus on their
autonomy, intelligence, mobility and interaction [10, 11, 12, 13, 14]. Agent-based
systems [15, 16] claim to be next generation software capable of adapting dynami-
cally to changing business environment and of solving a wide range of knowledge
processing application. Although sophisticated software agents can be difficult to
build from scratch due to the skills and knowledge needed, the widely available agent
construction toolkits may provide a quick and easy start to building software agents
without much agent expertise. Significant research and development into multi-agent
systems (MAS) has been conducted in recent years [17, 18, 19], and there are many
architectures available today [20, 21]. Nevertheless, several issues still need to be
faced to make the multi-agent technology widely accepted: secure and efficient exe-
cution supports; standardization; appropriate programming languages and coordina-
tion models.
Decentralization and openness are inherent properties of multiagent systems
(MAS). The technologies they provide are thus the right abstraction for developing
Web-oriented applications. Moreover, different works have been proposed to use
Semantic Web technologies (SWT) for representing various dimensions of MAS (e.g.,
interaction protocols, norms, organizations).
Consequently, recent research in MAS have seen an intensive use of Knowledge
representation together with increasing use of SWT. We envision that SWT will ulti-
mately play a central role in all parts of MAS. Thus far, work combining MAS and
SWT have only been concerned about addressing one dimension of MAS at a time.
Also, they were mostly tackling the agent [22, 23] and interaction dimensions [24, 25]
to ease communications, especially on domain knowledge. Other works have used
those technologies to model part of the organization structure [26], norms and com-
mitment [27], reputation [28] and more. The situation shows that it is time to go be-
yond these ad hoc solutions and integrate the pieces into a complete Semantic-Web-
based infrastructure. We observe too that none of the mentioned contributions were
really taking advantage of the Web aspect of these technologies, except some Web
service integration. We also want to provide models and specifications for the SS so
that services can uniformly query and reason about DKBS, web services, data, and
ontologies.
3 A Distributed Knowledge Based System
The proposed architecture is based on services and is designed to support all the proc-
esses of the system life cycle. That is, the proposed architecture is meant to include all
the concepts necessary to perform all activities related to the life cycle. We shall
adopt a generic, knowledge-based architecture. The enterprise members are repre-
sented by autonomous agents, geographically scattered, which are able to cooperate to
achieve a common business goal.
The various persons constituting the DKBS, in order to process distributed knowl-
edge bases, assume the following roles:
Knowledge Manager: is in the top of the hierarchy, and acts as a project manager,
but in higher levels. It's like a knowledge strategist, cooperating, defining, and dis-
tributing the knowledge to coordinate all the other roles;
Knowledge Provider: is the owner of human knowledge. It is typically an expert in
the application domain, but could be another person in the organization who does
not have the expert status;
Knowledge Analyst: uses a range of methods and tools that make the analysis of a
standard knowledge-intensive task relatively straightforward;
Knowledge System Developer: is responsible for the design and implementation.
The developer must have a basic background of analysis methods. In knowledge
system development, the main knowledge problems have been solved by the
knowledge analyst. Therefore, this role must have some skills of software design-
ers;
Knowledge User: makes use, directly or indirectly, of a knowledge system. Its
interaction with the knowledge base system is important for the project develop-
ment and validation;
Project Manager: manages the project, specially the knowledge engineer and the
knowledge system developer.
In turn, the structure of the system based on knowledge, is shown in Fig. 1. The
main components are [29, 30,31]:
base of knowledge. Base of knowledge intended for storage of knowledge about
the subject. Its concrete form depends strongly on the chosen model of knowledge
representation. The presence of this component is the main difference between the
systems based on knowledge from other types of programs;
output machine. Output machine generates a response to the user request using the
knowledge base. Its principle of operation is also dependent on the chosen model
of knowledge representation;
editor knowledge base – program to change the contents of the knowledge base;
user interface – the mechanism by which the communication user and the system.
Fig. 1. Structure of systems based on knowledge.
In addition to the agents representing the distributed knowledge base, a SS is intro-
duced in the MAS community. This agent reacts by seizing deal opportunities present
in the DKBS, and proceed thus to establish the corresponding virtual entities.
The principle of the Distributed Knowledge Base proposed in this paper is to de-
pends on a number of ontologies that provide semantic principle: ontologies of events
DKBS (Task Ontology); ontologies of services (Application Ontology); ontology
specialized areas of data systems based on (Domain Ontology). (fig. 2).
Fig. 2. Structure of DKBS.
DKBS can be represented as a model of interaction of the main artifacts: clients,
services and SS. Each component in the system is associated with a specific ontology.
SS provides:
semantic principle, which defines a formal description of information and allows
you to define the following characteristics of services: scalability, semantic inter-
operability, formal models of services and ontologies;
the principle of decision making;
the principle of distribution, which allows you to aggregate the capabilities of sev-
eral computing objects through cooperation.
Considering all the types of ontologies [32, 33, 34], we chose it for three, because
the Application Ontology is a model of a service describes its functional and non-
functional characteristics. Domain Ontology is a certain base set of knowledge for
each subject area. Task Ontology keeps information about all the functions and ac-
tions of the system.
Each ontology has its own expressive capabilities depending on their functional
purpose. Ontology-oriented for subject area (Domain Ontology), describe the diction-
ary of terms concrete and formal set four end subsets: concepts, relations, axioms and
interpretation functions. Task Ontology consists of a Glossary of terms, specialize
tasks and actions in subject area. Every task has a different status and stages of its
implementation. The main feature for all concepts of the ontology problem is the
time. Application Ontology is the most specific ontology, in addition to all the basic
concepts, contains specialized terms and instances subject area.
SS includes domain ontology and a set of modules to operate the services and their
ontological description a lot of events we have formed as a set of claims that can be
applied to DKBS and a set of axioms of their appearance.
4 Structure of Semantic Service in a DKBS
In each interaction of the SS, the ontology functionality is necessarily preserved. Each
function is characterized by the input (1) and output arguments (2). Correspondingly,
each action SS (3) is characterized by a function of the work with the ontology, the
values of its arguments when you call and the completion of the system.
f SS FAkIN ( fa1IN ,..., fa kIN ); k 1,..., K . (1)
f SS FAnOUT ( fa1OUT ,..., fa KOUT ); k 1,..., K . (2)
aij { f k , FAvaliIN , FAval OUT
j ); i 1,..., K ; j 1,...K ; f k F . (3)
In the DKBS, the communication and collaboration mechanism between services is
a problem that must be well considered in the design of the SS structure. The commu-
nication and collaboration mechanism covers the message transfer between services,
information exchange between SS and distributed knowledge bases, and the SS be-
haviors in the entire process from request submission, request handle, to result return.
These procedures can be summarized as follows:
SS deploys appropriate services on its local Knowledge Base, performs the regis-
tration on the knowledge bases;
The retrieval request accesses the system periodically. SS gets retrieval requests
and searches its Knowledge Base.
SS returns the retrieval result to the system after the search finishes.
The SS implies a classification and a sort order to the retrieval result according to
its corresponding retrieval request.
User views the retrieval results on the interface and selects the needed Knowledge.
With the information about the Knowledge Base embedded in the retrieval result,
the user then contacts the relevant enterprise to obtain the Knowledge.
The principle of operation of the SS is illustrated in Fig. 3. The knowledge base is
formed from three types of ontologies. The rules for recognizing performance prob-
lems are converted into ontology and the system knowledge base is placed. The work-
ing memory of a SS contains facts that correspond to simple events, composite events,
and identified problems and recommendations. The algorithm for using SS to perform
an analysis is as follows. The initial data is the sequence of events of the system, let's
call it the trace. Trace contains simple events. E The algorithm processes the events of
the route in the order of the time of their occurrence and performs the following cycle:
1. Read the next event from the trace E.
2. Convert it to the concept of ontology and add it to the working memory of the SS.
3. Start the SS. If the new event is a search query, then pick up the relevant answers
in the knowledge base. For an indefinite concept of a simple event, the output en-
gine consistently executes the rules for constructing composite events, for con-
structed fact-composite events, the rules for identifying performance problems.
The result of triggering a SS can be:
(a) adding a new fact-simple event to some of the constructed fact-composite
events;
(b) identification of fact-composite events of working memory for compliance or
non-compliance with the performance problem;
(c) delete or save a new event in the working memory;
(d) search for a saved event.
4. Retrieve the search results from the working memory or offer recommendations to
solve the problem.
Fig. 3. Structure of Semantic Service.
To assess the performance of the SS attached, create an information system of two
components:
event trace generator DKBS;
event trace analyzer for SS.
Events are recorded in the track when calling action functions. A simple event is
represented as:
e f , et , EPval , t , d , cs , (4)
where f F - action function;
et - the type of event that sets its parameters et EP (ep1 ,..., epK ) ;
EPval (v1 ,..., vK ) - event parameter values;
t - time of the event;
d - event duration;
cs - service code.
The implementation scheme of the tracer generator is presented in fig. 4. The tracer
generator consists of the following components:
OntologyGenerator program converts messages into ontology concepts;
library SM is designed to track messaging between services;
library Timer provides functions for obtaining current system time with high accu-
racy;
library TracerWriter saves simple events created by the tracer to a trace file.
Presumably all generator modules are implemented in the Java programming lan-
guage.
Fig. 4. The tracer generator.
The implementation scheme of the SS trace analyzer is shown in fig. 5. The trace
analyzer consists of the following components:
library UI implements the CC user interface.
library TraceReader provides reading simple events from trace files.
library Ontology implements event conversion to ontology concepts.
library Query generates queries to the SS.
Fig. 5. The trace analyzer.
5 Conclusion
Modern representations of data are changing the ways and forms of communication,
production and consumption of information. The dominance of horizontal relations,
structure-role of information decentralization of all types of data available at any time
on every device. The user should not care about the specific technology used to pro-
vide computing capacity or data storage, so you can say that a user has some informa-
tion about the remote resource. DKBS in order to study the data, their processing and
use evolving technologies of the semantic web. The formal model of the DKBS con-
sists of a set of ontologies; lots of services; a set of events that describe the processes
of the system; SS, a set of composite services and knowledge base. The completeness
and effectiveness of the system is determined by a multitude of ontologies. The main
difference in DKBS is the use of the service approach and ontologies in knowledge
engineering. SS acts as a software agent that regulates the interaction of all compo-
nents of the system. SS provides: semantic principle, which defines a formal descrip-
tion of information and allows you to define the following characteristics of services:
scalability, semantic interoperability, formal models of services and ontologies; the
principle of decision making; the principle of distribution, which allows you to aggre-
gate the capabilities of several computing objects through cooperation. The communi-
cation and collaboration mechanism of the DKBS covers the message transfer be-
tween services, information exchange between SS and distributed knowledge bases,
and the SS behaviors in the entire process from request submission, request handle, to
result return. The knowledge base is formed from three types of ontologies. The rules
for recognizing performance problems are converted into ontology and the system
knowledge base is placed. The working memory of a SS contains facts that corre-
spond to simple events, composite events, and identified problems and recommenda-
tions. The algorithm for using SS is discussed in the article. To assess the perform-
ance of the SS attached, create an information system of two components: event trace
generator and event trace analyzer.
References
1. Ruzhi Xu, Peiguang Lin, Cheng Liu: Research on Distributed Knowledge Base System
Architecture for Knowledge Sharing of Virtual Organization. Atlantis Press, pp. 349-357
(2010)
2. WEB 3.0 budet pitat'sya stekom tekhnologii blokcheyn [WEB 3.0 will be powered by
blockchain technology stack]. Available at: https://101blockchains.com/ru/web-3-0/ (ac-
cessed 01.03.2019)
3. Deloitte projects by 2025 that 10% of global GDP to be built on blockchain applications.
Available at: https://www.freightwaves.com/news/2017/9/2/deloitte (accessed 02.09.2017)
4. Tanenbaum A., Van Steen M.: Distributed systems. Pearson Prentice Hall (2007)
5. Burns B.: Designing Distributed Systems: Patterns and Paradigms for Scalable, Reliable
Services. O'Reilly Media, Inc (2018)
6. Allemang D, Hendler J.: Semantic web for the working ontologist modeling in RDF,
RDFS and OWL. Elsevier Inc. (2008)
7. Russell, S., Norvig, P.: Artificial Intelligence: A Modern Approach. Prentice-Hall, Engle-
wood Cliffs (1995)
8. Genesereth, M.R., Ketchpel, S.P.: Software Agents. Communications of the ACM 37(7):
48–53 (1994)
9. Wooldridge, M.J., Jennings, N.R.: Intelligent Agents: Theory and Practice. Knowledge
Engineering Review 10(2): 115–152 (1995)
10. Stoian, G., Popirlan, C.I.: A proposal for an enhanced mobile agent architecture (EMA).
In: Annals of the University of Craiova, Mathematics and Computer Science Series, vol.
37(1), pp. 71–79 (2010)
11. Tandareanu, N., Popirlan, C.I.: A Mobile Agents Approach for Knowledge Bases Process-
ing. In: Samarati, P., Yung, M., Martinelli, F., Ardagna, C.A. (eds.) ISC 2009. LNCS, vol.
5735, pp. 27–32. Springer, Heidelberg (2009)
12. Popirlan, C.I.: Knowledge Processing in Contact Centers using a Multi-Agent Architec-
ture. WSEAS Transactions On Computers 9(11), 1318–1327 (2010)
13. Popirlan, C.I., Stefanescu, L.: Mobile Agents System for Intelligent Data Analysis. In:
Proceedings of WSEAS Applied Computing Conference (ACC 2009), Athens, Greece, pp.
663–668 (2009)
14. Popirlan, C.I., Dupac, M.: An Optimal Path Algorithm for Autonomous Searching Robots.
In: Annals of University of Craiova, Mathematics and Computer Science Series, vol.
36(1), pp. 37–48 (2009)
15. Sycara, K.P.: Multiagent Systems. AI Magazine, 79–92 (1998)
16. Wooldridge, M.: Introduction to Multiagent Systems. John Wiley and Sons, UK (2002)
17. Kabbaj, A.: Development of Intelligent Systems and Multi-Agents Systems with Amine
Platform. In: Schärfe, H., Hitzler, P., Øhrstrøm, P. (eds.) ICCS 2006. LNCS (LNAI), vol.
4068, pp. 286–299. Springer, Heidelberg (2006)
18. Balachandran, B., Enkhsaikhan, M.: Developing Multi-agent E-Commerce Applications
with JADE. In: Apolloni, B., Howlett, R.J., Jain, L. (eds.) KES 2007, Part III. LNCS
(LNAI), vol. 4694, pp. 941–949. Springer, Heidelberg (2007)
19. Rodrguez, S., Fernndez, V., Julin, V., Corchado, J.M., Ossowski, S., Botti, V.: A
THOMAS Based Multi-Agent System for Recommendations and Guidance in Malls. Jour-
nal of Physical Agents 3(2), 21–26 (2009)
20. Hagras, H., Callaghan, V., Colley, M.: Intelligent Embedded Agents. Journal of Informa-
tion Sciences, 289–292 (2005)
21. Bouchachia, A., Prossegger, M.: A Bi-Clustering Agent-based Approach for Map Segmen-
tation. In: Proceedings of the IEEE Symposium on Intelligent Agents, IA 2009, Part of the
IEEE Symposium Series on Computational Intelligence, pp. 99–105 (2009)
22. T. Klapiscak and R. Bordini. JASDL: A Practical Programming Approach Combining
Agent and Semantic Web Technologies. In Proc. of DALT, pp 91–110, (2009)
23. M. Hadzic, E. Chang, and P. Wongthongtham: Ontology-Based Multi-Agent Systems.
Springer (2009)
24. OWL-DL as a FIPA-ACL content language. In Proc. of FOCA 2006, (2006).
25. Y. Zou, T. Finin, Y. Peng, A. Joshi, and R. S. Cost: Agent Communication in DAML
World. In Proc. of WRAC 2002, pp. 347–354 (2002)
26. D. Okouya, L. Penserini, S. Saudrais, A. Staikopoulos, V. Dignum, and S. Clarke: Design-
ing MAS Organisation through an Integrated MDA/Ontology Approach. In Proc. of
TWOMDE: pp. 55–60 (2008)
27. N. Fornara and M. Colombetti: Representation and monitoring of commitments and norms
using OWL. AI Comm., 23(4):341–356 (2010)
28. L. Nardin, A. Brand˜ao, and J. Sichman: Experiments on semantic interoperability of
agent reputation models using the SOARI architecture. Eng. Appl. of AI, 24(8), pp. 1461–
1471 (2011)
29. Dzharratano Dzh.: Ekspertnyye sistemy: printsipy razrabotki i programmirovaniye: per. s
angl. K. A. Ptitsyna. – 4-ye izd. [Expert systems: principles of development and program-
ming]. Moskva, Vil'yams (2007)
30. Smolin D. V.: Vvedeniye v iskusstvennyy intellekt: konspekt lektsiy [Introduction to Arti-
ficial Intelligence]. Moskva (2004)
31. Behnam Azvine, Nader Azarmi, Detlef D. Nauck: Intelligent Systems and Soft Comput-
ing: Prospects, Tools and Applications. Springer (2006)
32. Gruber Thomas R.: A translation approach to portable ontology specifications. Appeared
in Knowledge Acquisition, vol. 5(2), pp. 199-220 (1993)
33. Yu Liyang: Introduction to Semantic Web and Semantic Web services, CRC Press (2007)
34. John Davies, Rudi Studer, Paul Warren: Semantic Web technologies: trends and research
in ontology-based systems. John Wiley & Sons (2006)