Designing Features of Applied Ontology of Integrated Expert

Systems’ Typical Architectures Using the Intelligent Software

Environment of the AT-TECHNOLOGY Workbench

Galina V. Rybina

, Alexandr A. Slinkov

and Andrey A. Grigoryev

Department of Cybernetics, National Research Nuclear University «MEPhI», Kashirskoe highway, Moscow, Russia

Keywords: Integrated Expert Systems, IES, Intelligent Software Environment of the AT-TECHNOLOGY Workbench,

Technological Knowledge Base, Applied Ontology of Typical IES Architectures, SDP, RUC, Tutoring IES.

Abstract: A new stage of experimental research is described, devoted to the application of an ontological approach to

automate the design of software for various classes of intelligent systems, in particular, integrated expert

systems (IES) developed on the basis of an original problem-oriented methodology and an intelligent software

environment of the AT-TECHNOLOGY workbench. The main emphasis is on the features of building an

extended technological knowledge base by including a new component in its composition in the form of an

applied ontology of typical IES architectures, which provides a logical interconnection of all components of

the technological knowledge base and effective interaction with an intelligent planner in the process of

prototyping IES. As an example, a fragment of an applied ontology of typical architectures, constructed for

tutoring IES, is given.

1 INTRODUCTION

The concept of the ontology of standard architectures

of integrated expert systems (IES) (Rybina, 2021) is

an evolution of the problem-oriented methodology

for designing IES and the intelligent software

environment of the AT-TECHNOLOGY workbench

(Rybina, 2008), designed to automate and

intellectualize software design processes of various

classes of IES, especially at time-consuming stages of

system analysis.

A complete description of the considered

methodology and technology of designing and

developing applied IES is given (Rybina, 2008), and

it is necessary to indicate that the principal feature of

this methodology is the conceptual and software

modelling of the architectures of the developed IES at

each level of integrational processes in the IES, which

is effectively supported by the powerful functionality

of the intelligent planner using great opportunities of

the technological knowledge base (KB) (Rybina,

2019), which includes a huge set of specifications of

https://orcid.org/0000-0002-4077-3660

https://orcid.org/0000-0002-8688-4163

https://orcid.org/0000-0003-1188-6815

different standard design procedures (SDP) for the

design of the most common IES architectures (static,

dynamic and tutoring IES), a set of operational and

information reusable components (RUC), as well as

an applied ontology of typical IES architectures.

The main prerequisites for the expansion of

technological KB by creating an ontology of different

typical IES architectures were, on the one hand, a

significant amount of accumulated information and

software of various classes of IES developed in recent

years, and on the other hand, in the context of

interaction with an intelligent planner, the need to

reduce the accessibility to semantically

heterogeneous RUC with implicit functionality when

their search and initialization in the conditions of the

implementation of a specific SDP.

In general, if we consider new approaches to

automation and intellectualization of software system

design processes using or under the control of

ontologies (Rybina, 2021), then the place and role of

ontologies here is significantly determined by the

level of complexity of the architecture models of the

452

Rybina, G., Slinkov, A. and Grigoryev, A.

Designing Features of Applied Ontology of Integrated Expert Systemsâ

Z Typical Architectures Using the Intelligent Software Environment of the AT-TECHNOLOGY Workbench.

DOI: 10.5220/0011951100003612

In Proceedings of the 3rd International Symposium on Automation, Information and Computing (ISAIC 2022), pages 452-456

ISBN: 978-989-758-622-4; ISSN: 2975-9463

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

systems being developed, the availability of adequate

life cycle models (LC) and semantically correct

reflection of the basic design processes at all stages of

LC taking into account the ontological representation

of the projected architecture, composition, structure

and specifications of individual components and the

relationships between them.

Nevertheless, despite a wide range of works on

ontological engineering (Calero, 2006), (Happel,

2006), (Bossche, 2007), (Jabar, 2019), (Horoshevskij,

2019), (Erzhenin, 2020), (Negoda, 2021), etc., issues

related to the development of a significant and

semantically adequate ontological model of software

system design processes, in particular, intelligent

systems, are poorly considered. So the expansion of

research in the framework of intelligent technology

creation for designing an IES as a common class of

intelligent systems with extended scalable

architectures is especially important today, including

the details of combining the methods of intelligent

planning with an ontological approach (Rybina,

2019).

This paper discusses the results of an

experimental software study of the actual structure of

the considered in main theme applied ontology, the

model and design methods of which allow us to

jointly take into account the semantic features of the

architecture models of the designed IES and the

features of component-by-component functionality in

the form of a set of RUCs for each SDP. It should be

noted that since the greatest number of software

components and tools that have been tested and

reengineered as part of the AT-TECHNOLOGY tool

workbench and designed in the form of operational

and information RUCs have been accumulated for the

implementation of the tutoring IES architectures, the

corresponding SDP "Designing tutoring IES and

web-IES" was selected as the base test field for

various studies and experimental software modelling.

2 BRIEF DESCRIPTION OF THE

ONTOLOGY MODEL AND

FEATURES OF APPLIED

ONTOLOGY DESIGN

As it was shown in (Rybina 2021), the choice of the

basic model of the ontology of typical architectures

was greatly influenced by the positive experience of

creating several ontologies of courses/disciplines for

tutoring and practical use in the educational process,

the development of which was carried out on the basis

of a model in the form of a semantic network of a

special type (Rybina, 2017), (Rybina, 2022).

Therefore, a modified semantic network is used

here as an ontology model, but of a simpler form

(Rybina, 2021) (Figure 1):

Figure 1: Ontology model in from of semantic network.

Accordingly, the actual applied ontology (Rybina,

2021) is reproduced in the form, shown on Figure 2:

Figure 2: Applied ontology.

Designing Features of Applied Ontology of Integrated Expert Systemsâ

Z Typical Architectures Using the Intelligent Software

Environment of the AT-TECHNOLOGY Workbench

453

Now let's briefly consider some aspects of

designing the actual version of the applied ontology

of various typical IES architectures, the main purpose

and purpose of which is to take into account and

reflect at different levels of the ontology the features

and logical relationships of the architecture model of

the projected IES, as well as the specification of

methods and algorithms implemented in the

corresponding RUC.

It should be noted here that the architecture model

of the prototype of the IES, represented in the form of

a hierarchy of extended data flow diagrams (EDFD)

(Rybina, 2008), is one of the important components

of the process of prototyping of the IES, due to the

fact that its order has a strong effect on the prototype

of the IES and its capabilities for the implementation

of a specific class of tasks to be solved. The

emergence of architectural elements at different

levels of nesting due to the implementation of multi-

level integration processes reproduced in the EDFD

hierarchy required the use of different non-trivial

approaches and solutions, including a set of plural

information and operational RUCs included in

various software tools of the AT-TECHNOLOGY

workbench.

Therefore, the general structure and basic levels

of the considered applied ontology are given in

(Rybina, 2021), as well as connection (relation) types

presented in that research and at Figure 1 above.

Taking into account these conceptual and

functional features of IES prototyping (Rybina

2008), (Rybina, 2019), all algorithms and procedures

for designing, storing and maintaining the considered

in main title applied ontology as an significant section

of the technological KB were developed in such a

way as to make the possibility of access and

comprehensive customization the appropriate RUC to

perform all planned tasks with the help of an

intelligent planner, depending on the features of the

model the architecture of the designed IES (at the

same time, a knowledge engineer can perform some

tasks independently or jointly with an expert).

For automated support of the processes of

designing an applied ontology of typical IES

architectures, modified tools were used that function

as part of the basic tools of the AT-TECHNOLOGY

workbench and allow to fully implement the

necessary functionality, as well as for modeling

interaction with an intelligent planner (Rybina, 2019)

developed tools for managing interaction with

technological KB were involved.

3 APPLIED ONTOLOGY OF

TYPICAL IES

ARCHITECTURES

FRAGMENT’S EXAMPLE (SDP

"DESIGNING TUTORING IES

AND WEB-IES")

As an example, we will give a fragment of the applied

ontology of typical IES architectures, which shows

the conceptual (logical) and program relationship

between the set of standard processes for tutoring IES

related to displaying the current student model, built

as a result of web testing, on the ontology of a specific

course/discipline and the formation of an individual

strategy (plan) depending on the results obtained

tutoring type (Rybina, 2008), (Rybina, 2017).

As noted above, the overall management of these

and other processes in the two basic modes of

operation of tutoring IES (DesignTime and RunTime)

(Rybina, 2008), (Rybina, 2019), (Rybina, 2017) is

supported by an intelligent planner, the SDP

"Designing tutoring IES and web-IES" and a set of

RUC implemented using a significant amount of

various software and information tools registered in

different years in the AT-TECHNOLOGY

workbench and included in the subsystem of support

for the design of tutoring IES.

Therefore, issues related to preliminary analysis,

structural and functional identification and formal

representation of all software tools and components

in accordance with the basic RUC model are of great

importance for designing an applied ontology of

typical IES architectures (Rybina 2008). Below are

the formal descriptions (concretizations) of two

conceptually related RUC implementing the

functionality mentioned above.

Specification of the RUC "Mapping the current

student model to the ontology of the

course/discipline" in accordance with the basic model

of the RUC <N, Arg, F, PINT, FN> (Rybina, 2008) is

defined as:

N - the name of the registered component

"Mapping the current model of the student on the

ontology of the course/discipline";

Arg = <Arg1, Arg2>, where Arg1 is the

course/discipline ontology, (previously built in

DesignTime mode (Rybina, 2008), (Rybina, 2017));

Arg2 - current student model (M1cur);

F: Arg1 х Arg2 → R - method for evaluating the

results of testing students, where R = {rj}, (j = 1÷m)

- a set of "problem areas" of a particular student

(Rybina, 2008), (Rybina, 2017);

ISAIC 2022 - International Symposium on Automation, Information and Computing

454

PINT - RUC interface "Formation of tutoring

strategies";

FN - the function of forming the current level of

the student’s competencies (Rybina, 2017).

So, the specification of the RUC "Formation of

tutoring strategies" is defined as:

N - the name of the registered component

"Formation of tutoring strategies";

Arg = <Arg1, Arg2>, where Arg1 = {arg1i} is the

set of course/discipline ontologies, (i=1÷n), (n is the

number of course/discipline ontologies);

Arg2={arg2j} - set M1tech, (j=1÷p) where p -

number of M1tech;

F = <F1, F2>, where F1: Arg1 х Arg2 → A -

method of forming a set of tutoring strategies (plans),

where A = {ai} (i=1÷k) - set of tutoring plans, where

k is the number of plans; F2: Arg1 × A → B is a

method for generating a set of learning influences

based on plans, where B = {Bi}(i=1÷m) is a set of

learning influences, where m is the number of

learning influences;

PINT - RUC interface "Management of the

implementation of learning influences";

FN - formation of a set of tutoring strategies

containing an ordered set of learning influences.

Figure 3: Fragment of the applied ontology.

Figure 3 shows a fragment of the applied ontology

of typical IES architectures, the upper levels of which

reflect the functional and structural features of the

subsystem for supporting the design of tutoring IES,

which operates in the basic version of the AT-

TECHNOLOGY workbench (Rybina, 2008),

(Rybina, 2017). The middle and lower levels of the

ontology are formed on the basis of formal

specification of the RUC.

The greatest variety of connections is observed at

the levels of methods/operations, where such types of

connections are used, such as, for example,

connections of the RA type (aggregation) between the

method of forming a set of learning influences and the

method of forming a set of tutoring strategies, as well

as connections of the RS type (strong) between the

method of formation of educational training task

"Filling in the gaps in the text” the RUC “Formation

of educational training tasks” and others.

4 CONCLUSION

The ongoing research and experimental software

modeling of software design methods for applied

power plants with various architectural typologies

under the control of ontologies are quite new for

artificial intelligence technologies and software

engineering in general.

It is too early to expect practical results in the field

of creating effective ontological models and powerful

tools and platforms for automating and

intellectualizing software development processes of

intelligent systems. Nevertheless, the accumulated

experience and the constantly developing

technological base in the form of the environment of

the AT-TECHNOLOGY workbench allow us to

solve the scientific and practical tasks set in stages.

REFERENCES

Rybina G. V., Slinkov A. A. The Implementation of the

Ontological Approach to Control of the Processes of

Designing Integrated Expert Systems Based on the

Problem-Oriented Methodology. // Artificial

Intelligence. 19th Russian Conference, RCAI 2021,

Taganrog, Russia, October 11-16, 2021. Proceedings.

Springer Nature Switzerland AG 2021. pp. 354-364.

Rybina G. V. Teoriya i tekhnologiya postroeniya

integrirovannyh ekspertnyh sistem. Monografiya. – M.:

Nauchtekhlitizdat, 2008.

Rybina G. V., Blokhin Y.M. Methods and Software

Implementation of Intelligent Planning for Integrated

Expert System Design // Scientific and Technical

Information Processing, 2019, 46 (6), pp. 434 - 445

Ontologies for Software Engineering and Software

Technology. 1st Ed. / C. Calero, F. Ruiz, M. Piattini –

Springer, 2006.

Happel H.J. KOntoR: An Ontology-enabled Approach to

Software Reuse / H.J. Happel, A. Korthaus, S. Seedorf,

P. Tomczyk // Proc. SEKE 2006: the 18th International

Designing Features of Applied Ontology of Integrated Expert Systemsâ

Z Typical Architectures Using the Intelligent Software

Environment of the AT-TECHNOLOGY Workbench

455

Conference on Software Engineering & Knowledge

Engineering (July 5-7, 2006, California, USA, 2006).

Bossche M.V. Ontology Driven Software Engineering for

Real Life Applications / M.V. Bossche, P. Ross, I.

MacLarty, B. van Nuffelen, N. Pelov // Proc. 3rd

International Workshop on Semantic Web Enabled

Software Engineering, SWESE 2007 (Innsbruck,

Austria, 2007).

Jabar M.A. General Domain Ontology in Enterprise

Software Development Process / M.A. Jabar, M.S.

Khalefa // International Journal of Engineering and

Advanced Technology. – 2019. - Vol. 8, Is. 3S.

Horoshevskij V.F. Proektirovanie sistem programmnogo

obespecheniya pod upravleniem ontologij: modeli,

metody, realizacii //Ontologiya proektirovaniya. 2019.

T.9. №4. S.429-448.

Erzhenin R.V., Massel' L.V. Ontologicheskij podhod k

predstavleniyu znanij o metodologii modelirovaniya

slozhnoj sistemy upravleniya. // Ontologiya

proektirovaniya. 2020. T.10.№4.S. 463-474.

Negoda V. N.,Kulikova A.A. Skvoznoe proektirovanie

avtomatizirovannyh sistem na osnove ontologij //

Ontologiya proektirovaniya. 2021.T.11.№4.S.450-463.

Rybina G. V. Intellektual'naya tekhnologiya postroeniya

obuchayushchih integrirovannyh ekspertnyh sistem:

novye vozmozhnosti // Otkrytoe obrazovanie. 2017.

№4. S. 43-57.

Rybina G. V., Nikiforov A. Y., Slinkov A. A., Grigoryev

A. A. Automated Formation of the Unified Ontological

Space of Students’ Knowledge and Skills to Implement

Intellectual Tutoring Tasks Based on Tutoring

Integrated Expert Systems. // 2022 VI International

Conference on Information Technologies in

Engineering Education (Inforino), 2022, pp. 1-6.

ISAIC 2022 - International Symposium on Automation, Information and Computing

456