Goal

The global objective of the Domain Ontology Description is to provide an overview of the problem domain. Stakeholders naturally express requirements in their own terms and with implicit knowledge of their own work [71]. Therefore the aim of this activity is deepening the understanding of the problem by complementing the usual requirements description in terms of use cases with a description of the concepts that compose the problem domain. It describes concepts used in the specific language of the application domain and users. Results of this work can sometimes imply modifications in uses cases. The design of the domain ontology occurs very earlier in our methodological process and this denotes the substantial role played by it in our process, and especially in the organisation identification and capacity identification activities.

Input

The problem’s ontology is deduced from the text descriptions of system usage scenarios, from interviews to stakeholders and domain experts, and from the previously identified requirements.

Output

The ontology is described in terms of concepts (categories, entities of the domain), predicates (assertions on concepts properties), actions (performed in the domain, and affecting the status of concepts) and their relationships. An profile for UML class diagrams is used to describe the ontology. Table 1 details mapping between ASPECS metamodel and UML constructs used to describe ontology. This diagram can be translated into an XML schema that can be used to obtain a Resource Description Framework (RDF4) encoding of the ontological structure.

ASPECS Item UML Construct UML Stereotype Concept Class concept Predicate Class predicate Action Class action Ontology Package ontology

Table 1: UML Profile Detail for Ontology Description

MAS Metamodel Elements

Define(Ontology), Define(Concept), Define(Action), Define(Predicate), Relate(Concept, Concept), Relate(Action, Concept), Relate(Predicate, Concept).

Work to be done & Methodological guidelines

The first task in the ontology description, aims at identifying major concepts. It can be done by looking at use cases (textual) description to find recurrent nouns. These concepts are then arranged by means of inheritance and abstraction. New higher level concepts will be so identified. When major concepts have been identified, the third task consists in determining concepts relationships and especially composition ones. There are two ways available to determine these relations. The first consists in looking into existing and well known ontologies to determine if a reuse is possible. The second is interviewing domain experts and looking into the textual description of usage scenario and other available documentation (see subsection 4.4).

Example

The ontology associated to the domain of the FIRA Robot soccer is partially detailed in figure 6. This ontology represents knowledge about robot soccer matches such as how time periods are handled, the different mobile (and thus situated) elements, rules handling, etc.

Figure 6: Ontology of the problem domain associated to FIRA Robot soccer