The paper is devoted to research into methods for requirements elicitation. From our perspective, the use of ontology engineering is particularly interesting. The author focuses his attention on selected aspects of ontology engineering, resulting from the needs of his doctoral thesis, organized by the Design Science Research paradigm. The article seeks to recognize a certain level of interplay between ontology engineering and individual tasks of the process of requirements elicitation (a synergy). Therefore, it aims to explain possible expectations for ontology engineering, which result from the specific features of the process of requirements elicitation. In order to identify the guidelines for the developing domain ontologies based on the requirements elicitation process. The originality of the article results from the consideration of the interplay between ontology engineering and requirements engineering, relative to the context of software engineering. As a result, the software engineering layers propagated by Pressman have been extended to include another layer of management philosophy. The rationale behind this was the noticing of the impact of other concepts of philosophy on this work, such as axiology or epistemology.
missing, incomplete and changing requirements. It also means that it is important, from
the perspective of the PhD thesis, to agree on the expectations for the process of creating
domain ontologies. It consists of the following issues:
• what expectations for elicitation of requirements, does the use of ontology
engineering bring?
• how does the elicitation of requirements influence the use of the ontology
engineering workshop?
• what is the significance of the specificity of domain ontologies (various conceptual
categories) for the production of ontologies?
Despite many possible issues of interest to the doctoral thesis, the paper attempts to
define only the influence of ontology engineering on requirements engineering, to elicit
requirements based on domain ontologies. The review of the literature devoted to the
issue of the use of ontologies could be much larger if the scope of literature was not
narrowed down by using the Population, Intervention, Comparison, Outcomes, Context
(PICOC) conceptual framework [
The purpose of the paper is the study on ontology engineering guidelines for elicitation
the requirements for IT systems. In general, it has been arranged according to the
classical hierarchy of software engineering issues [
The next points of this paper have been organized by the pyramid of software engineering issue, presented on the Fig.1. The research summary has also been organized in accordance with the layers of software engineering presented in Fig. 1. 3
The work described in this article is subordinated to the guidelines of the research
paradigm of information systems in the science of creating artifacts ‒ DSR (Design
Science Research)[
In connection with the sequence of DSR steps and requirements of step 2, i.e. the definition of the objectives of the study, the author of the paper intends to achieve the following goals: • Characterize the research goal required for the third research step, i.e. Design and development. • To restrict the literature of the subject, supporting the achievement of the research objective.
Achievement of goals will be possible by choosing specific issues from the pyramid of software engineering, presented in Fig. 1. 4 4.1
In order to characterize management philosophy, the author adopted the following
approach to concepts such as: axiology, epistemology, ontology. Fig. 2 presents
cooperative issues of management philosophy. It assumed that the definition of axiological
criteria allows adopting the epistemological standpoint of knowing the organization
only in the next step. In turn, it enables the ontology production. At the same time, it is
managed by methodology, e.g. the one for requirements elicitation.
The paper is based on axiological criteria resulting from ISO / IEC 25010: 2011[
The ISO/IEC 25010: 2011 standard currently summarizes the quality attributes as well as properties for IT systems and requirements for them. Thus, this standard affects the perception of quality at the international level. The Table 1 summarizes these attributes, and the properties of these attributes. Considerations subordinated to the purpose of the article, based on the philosophy of management, pay attention to future system users. The so-called strong sociology recognizes society's knowledge as crucial. As for the aforementioned quality attributes, it should lead to reconciliation of all attributes these attributes with future users. 4.3
The relationship between ontology engineering and software production was agreed in
the context of the ODSD process (Ontology Driven Software Development)[
Fig. 3. Control flow between the envisioned activities: Creation and development of domain ontologies, Development and implementation of IT business alignment strategy, Implementation of IT project, Exploitation of an IT solution. Source: Own study The considerations of this article pay attention primarily to the fact that these tasks are performed simultaneously. From what it can be concluded that the software development process could be open to accepting design artifacts coming from the IT business alignment strategy. And thus elicitation of requirements based on domain ontologies plays an important role here. 4.4
While discussing methods related to requirements engineering, attention is paid to the
relationship of this work with system engineering and software design. In Pressman's
book [
"For requirements engineering, we suggest using a domain-independent ontology to determine the general concepts, requirements and dependencies to be detected when disclosing requirements."
However, without diminishing the role of ontology to only defining the guidelines
for the disclosure of requirements, attention here is paid to the definition that somehow
opens the possible scope of ontology. "Ontology-driven (or ontology-based) RE form
a RE process or at least a RE method comprehensively aided by ontologies. Therefore,
ontologies are involved in the RE process. ODRE clearly states the method to integrate
a proposed process on the continuous process RE” [
However, attention should be paid to the popular definition of ontology “An ontology
is a formal, explicit specification of a shared conceptualization”. This definition was
influenced by the work of the authors: Gruber [
The point of view adopted in the article distinguishes between methods and accompanying tools. In the subject literature, one can identify those ontologies whose application is top-down (e.g. OntoREM, an ontology-driven requirements engineering methodology) [28] and trying to create a kind of matrix for unidentified problems of what the ontology will present (e.g. ODM, an ontology definition meta model) [29]. The issue of marketization and industrialization of tools supporting the ontology engineering methods is also noteworthy. What is important in is whether the tooling environment provides the possibility of using, e.g. the traceability mechanism between ontology and the CIM model, as well as the wider process and even the life cycle of the software.
In the review of approaches in the use of ODSD [
DL) for ontological modeling. 3. The DSL Meta-Model Ontology Based Approach. OWL ontologies are integrated in the model-based software technology that uses that uses automated program synthesis for generating software from models [31]. 4. The Three ontology method [32] that uses domain, task and top-level ontologies for ontological modeling of a software system.
In addition, the same work has reached the very sources of the ODSD idea and it has been explained that ontologies as semantic declarative models can extend MDA. Starting in 2000, when W3C introduced Ontology Driven Architecture (ODA), this contributed to the introduction by the OMG Ontology Definition Metamodel (ODM), whose latest version is for the year 2014. 5
The purpose of the work, through the analysis of the literature based on the DSR research paradigm, is illustrated in Fig. 4. It can be concluded that revealing requirements on the basis of ontologies is primarily a reconciliation of mutual meta-models between the source ontology and the target one containing requirements.
Domain ontology
elicitation •Meta-model compability with requirements Requirements •Meta-model compability with domain ontology The literature on the subject describes specific solutions relevant to specific fields of application (e.g. OntoREM), as well as those open to various variants of the application of ontologies (e.g. ODM). An objective choice between dedicated and universal solutions could be problematic, if it were not the software development process and subprocesses. The methods are directly dependent on this process (ODSD was selected in this paper). On the other hand, the classic layered approach to software engineering issues allows to look comprehensively to justify the choice. Nevertheless, the analysis of the literature allowed to formulate the conclusions contained in Table 3. The multiplicity of possible categories of ontologies indicates the reconciliation of meta-models between the source ontology and the target ontology. Therefore, a solution was chosen that would allow defining both ontologies and meta-models. The ODM (Ontology Definition Meta Model) standard seems to meet these expectations. An example of analogous application is the MOST Project (Marrying Ontology and Software Technology). The software that supports the entire software production process is important here, which allows to use the functionality of, among others, traceability.
Methods of elicitation of requirements should be based on restrictive features of ontologies according to Gruber definition (conceptualization, formality, explicity), also ensuring compliance with the environment adequate to the selected development process.
The ODSD process specifies the use of the CIM model. With this model, CIM cooperates under the business analysis process, i.e. requirements elicitation, where tasks are specified: 1) Understanding the application field; 2) Identification of sources of requirements; 3) Caring for quality Management philosophy 6
Shareholder analysis; 4) Selection of techniques, methods, tools; 5) Task to requirements elicitation.
Agreeing with project stakeholders about the quality attributes: functional suitability, performance, compliance, usability, reliability, protection, ease of maintenance, portability.
The context of management philosophy, suggests: axiological criteria based on the ISO / IEC 25010: 2011 standard, epistemological stand: strong sociology program, open ontology for any area of application, methods as flexible as possible for changes and ad-hoc expectations.
The paper tried to explain the expectations of ontology engineering with respect to the process and tasks devoted to requirements elicitation. As a result, it provides knowledge used in the PhD thesis devoted to the production of domain ontologies. Based on a review of direct literature and other similar reviews, it can be concluded that there are no uniform recommendations. At the same time, it determines the adoption of certain criteria, based on this paper in the position towards management philosophy. The whole of the issues raised in the paper has been arranged according to the classic layers of software engineering issues. As a result, it was assumed that aligning ontology engineering expectations for the process of elicitation requirements should be based on the position on these indissoluble issues: • Reconciliation of stakeholders' expectations; • Adoption of the position on quality attributes; • Consolidation of the development process with the business analysis itself; • Subordination of work methods to the required ontology features; • Ensuring compatibility between meta-models of domain ontologies and requirements. 7 28. Kossmann, M., Wong, R., Odeh, M., Gillies, A.: Ontology-driven Requirements Engineering: Building the OntoREM Meta Model. In: 2008 3rd International Conference on Information and Communication Technologies: From Theory to Applications. pp. 1–6.
IEEE (2008) 29. OMG Board: About the Ontology Definition Metamodel Specification Version 1.1, https://www.omg.org/spec/ODM/1.1/ 30. Katasonov, A.: Ontology-driven software engineering: beyond model checking and transformations. . Int. J. Semant. Comput. 6(2), 205–242. (2012) 31. Haav, H.-M., Ojamaa, A.: Semi-automated integration of domain ontologies to DSL metamodels. Int. J. Intell. Inf. Database Syst. (IJIIDS), 10(1-2), 94–116. (2017) 32. Hoehndorf, R., Ngonga Ngomo, A.-C., Herre, H.: Developing consistent and modular software models with ontologies. Frontiers in Artificial Intelligence and Applications. pp. 399-412, IOS Press (2009)