232
B. Di Martino and A. Esposito
defines a virtual Global Data Space where applications freely read and write data by
using a name and key pair, which are used to uniquely identify them and make them
discoverable by interested services.
2.1 Semantic Technologies
Semantic technologies are mostly based on the definition of Ontologies to provide a
common reference for terms and concepts. In particular, an Ontology can be defined
as a formalized representation of knowledge, a sort of standardly presented vocabulary composed of concepts and meaningful entities that are in relation with each
other. One could be tempted to compare them with taxonomies, because of their
innate hierarchical structures, but they differ because ontologies allow for the definition of complex relationships among concepts, which greatly surpasses a merely
hierarchical categorization. Terms used in an ontology have a specific meaning, at
least in the domain in which the ontology itself has been developed. That is why two
different kinds of ontologies can be taken in consideration.
Upper Ontologies provide a more general description of knowledge, in which the
represented terms, concepts and relations can be applied to several domains, without
losing meaning. Information contained in Upper Ontologies can thus be applied in
different context, and used as a common reference.
Domain Ontologies focus on a specific domain, and the concepts they represent
can have different meaning from a context to another. It is common to have the same
term representing different information in different Domain Ontologies.
Through the years, ontologies have been expressed through a series of different
standards, generally based on XML, such as RDF and more recently OWL [4],
which became the most used. Based on OWL several ontologies have been defined
to support data and services interoperability. As an instance, OWL-S is a complex
ontology framework used to describe services, their relations and compositions [13].
Through the use of a Service class, OWL-S defines a reference to a specific semantic
service, which is in turn composed of three other parts:
• the Profile class, essentially meant to be human readable, offers a natural language
based description of what the specific service offers. The Profile is connected to
a series of properties which emphasize the service’s name, complete description,
quality of services, context applicability and possible limitations.
• the Model class offer a precise description of the interfaces of the Service, including its input and output parameters, message formats, adopted communication
protocols or port numbers.
• the Grounding class allows the mapping between the abstract descriptions offered
by Model and Profile to concrete implementation of the service, exposed via standard languages, namely WSDL.
All the connections between the Service class and the Grounding, Model and Profile
Classes are obtained via specific properties.
B. Di Martino and A. Esposito
defines a virtual Global Data Space where applications freely read and write data by
using a name and key pair, which are used to uniquely identify them and make them
discoverable by interested services.
2.1 Semantic Technologies
Semantic technologies are mostly based on the definition of Ontologies to provide a
common reference for terms and concepts. In particular, an Ontology can be defined
as a formalized representation of knowledge, a sort of standardly presented vocabulary composed of concepts and meaningful entities that are in relation with each
other. One could be tempted to compare them with taxonomies, because of their
innate hierarchical structures, but they differ because ontologies allow for the definition of complex relationships among concepts, which greatly surpasses a merely
hierarchical categorization. Terms used in an ontology have a specific meaning, at
least in the domain in which the ontology itself has been developed. That is why two
different kinds of ontologies can be taken in consideration.
Upper Ontologies provide a more general description of knowledge, in which the
represented terms, concepts and relations can be applied to several domains, without
losing meaning. Information contained in Upper Ontologies can thus be applied in
different context, and used as a common reference.
Domain Ontologies focus on a specific domain, and the concepts they represent
can have different meaning from a context to another. It is common to have the same
term representing different information in different Domain Ontologies.
Through the years, ontologies have been expressed through a series of different
standards, generally based on XML, such as RDF and more recently OWL [4],
which became the most used. Based on OWL several ontologies have been defined
to support data and services interoperability. As an instance, OWL-S is a complex
ontology framework used to describe services, their relations and compositions [13].
Through the use of a Service class, OWL-S defines a reference to a specific semantic
service, which is in turn composed of three other parts:
• the Profile class, essentially meant to be human readable, offers a natural language
based description of what the specific service offers. The Profile is connected to
a series of properties which emphasize the service’s name, complete description,
quality of services, context applicability and possible limitations.
• the Model class offer a precise description of the interfaces of the Service, including its input and output parameters, message formats, adopted communication
protocols or port numbers.
• the Grounding class allows the mapping between the abstract descriptions offered
by Model and Profile to concrete implementation of the service, exposed via standard languages, namely WSDL.
All the connections between the Service class and the Grounding, Model and Profile
Classes are obtained via specific properties.