Introduction to Ontologies

When modeling across a diverse group of organizations that are exchanging information, you need to have consistent definitions of terms and the relationships between terms. Using an Ontology, you can define what you know about your domain and refer to these definitions within your information models. Originally, the Web Ontology Language (OWL) and the Resource Definition Framework (RDF) were defined purely to provide a machine-to-machine interchange of metadata and semantics. Now, OWL and RDF have been combined to form the Ontology Definition Metamodel (ODM), which supports a more human-readable, abstract modeling of these languages. Enterprise Architect provides a wide range of modeling capabilities for interacting with metadata in a variety of ways, for a variety of purposes. The following topics introduce the MDG Technology for ODM, providing a definition of ontologies and showing how they can be integrated with other Enterprise Architect features to address the key challenges of developing large-scale ontologies within the fully-integrated modeling environment. The ontologies can be related to other modeling languages such as the Unified Modeling Language (UML), the Systems Modeling Language (SysML) and UPDM.

Benefit of Ontologies

The real value of ontologies lies in their ability to formally define and share standardized knowledge in an industry domain and between domains. Creating such ontologies, however, can quickly become a daunting task both logistically and technically. Four key challenges are:

• Stakeholder Enablement: The ability to engage subject matter experts, analysts, managers, practitioners and implementers to contribute value to ontologies
• Maintainability: The ability to update, compare and revise ontologies against predecessor baselines
• Reusability: The ability to define ontology templates for repeated use across development projects
• Traceability: The ability to ‘drill down’ or ‘drill up’ within ontology elements to visualize their interrelationships.

Addressing such challenges within the ontology development environment very quickly goes beyond the scope of traditional, XML-based ontology tools. It requires the investment of additional applications that operate outside the core development activity. This is why Enterprise Architect supports the integration of ontology development within the visual modeling platform using ODM, to simplify modeling on a more abstract level whilst maintaining the necessary details defined in the OWL and RDF specifications and providing connectivity and interoperability with other modeling languages used in system development.

Four-Level Ontology Stack

The Web Ontology Language OWL provides a way of representing knowledge and describing taxonomies and classification graphs. The nouns (vertices) represent classes of objects, and the verbs (edges) model the relationships or connections between the objects. The OWL language is the upper-most element in a stack and is based on the W3C's schema definition language Resource Description Framework Schema (RDFS) which specializes the basic metadata markup language Resource Description Framework (RDF).

Relationship to Class Models

Resource Description Framework (RDF) data model are similar to classical conceptual modeling approaches such as entity–relationship or unified modeling language (UML) class diagrams. RDF is based on the notion of making statements about resources (in particular web resources) in expressions of the form:

subject–predicate–object

These statements are known as triples because they contain three elements or parts. The subject denotes the resource, the predicate denotes traits or aspects of the resource and expresses a relationship between the subject and the object.

For example, one way to represent the notion "The car has the color red" in RDF is as the triple: a subject denoting "the car," a predicate denoting "has the color", and an object denoting "red."

Therefore, RDF uses subject instead of an object (or entity) in contrast to the typical approach of an entity–attribute–value model in object-oriented design: entity (Car), attribute (Color), and value (Red).

In a UML Class diagram, a modeler would represent this notion as a Class Car with an Attribute Color. An Instance (Object) of this class could be assigned the value Red for the Color attribute.