Architecture for Including Personalization in a Mobile GIS via Semantic Web Techniques - Geographical Information Systems: Trends and Technologies - page 214

Linked Data . Today, perhaps the main problem of current tourist applications
is interoperability. A possible solution to interoperability problems could be
the use of well-known ontologies adopted by the majority of the community,
such as the ones within the Open Linked Data (OLD) project (Bizer et al.
2009). OLD is an initiative to interconnect the information from the web
in a way that is understandable to computer programs. A set of ontologies
are defi ned, where each ontology describes the information structure of a
different domain. However, the power of OLD is not in their ontologies,
but in their interrelationships. Ontologies are related to each other, allowing
programs to navigate from one concept to other concepts, either in the
same domain or in different domains. For example, the system links
LinkedGeoData (LGD) ontology, which contains geographical data from
OpenStreetMap, with DBpedia, which contains data on Wikipedia.
Semantic Languages
Once we know how to represent an ontology from a conceptual point of
view, we need to know the language used to represent it. As ontologies
are conceptual schemas focused on computers, they should be written in
a language that a program should be able to understand. Today, the most
used ontology languages are Resource Description Framework (RDF) (W3C
2004) and Web Ontology Language (OWL) (W3C 2007).
RDF/OWL
RDF is a standard model for data interchange on the Web. This language
is based on Triples that represent expressions of the kind subject-predicate-
object ( who does what with something ). For example, a triple could be “Jack
has a house”, where “Jack” is the who , “has” is the what and “a house” is
the something . Triplets can be semantically processed by machine agents
and most of the current Semantic Web applications apply reasoning using
such triplets (Stojanovic et al. 2002). OWL language extends RDF with the
possibility to express additional constraints.
RDF and OWL languages use different standard terms to represent
concepts and relationships of ontologies. Firstly, there are Classes , which
provide an abstraction mechanism for grouping resources with similar
characteristics. Secondly, there are the instances , which are the individuals
in the class. Thirdly, there are the properties that relate information and
can be defi ned as: Object Properties , which link individuals to individuals,
and Datatype Properties , which link individuals to data values. These terms
are very important, because for every ontology we have to defi ne every
single aspect interesting to the knowledge domain by using these terms
consistently.