What are Ontologies Useful For?

What are Ontologies Useful For?

Anna Goy (Università di Torino, Italy) and Diego Magro (Università di Torino, Italy)
DOI: 10.4018/978-1-4666-5888-2.ch734
OnDemand PDF Download:
No Current Special Offers

Chapter Preview



The term ontology originates in Philosophy, where it is usually written with the capital initial letter and refers to the branch of this discipline that studies the basic categories of existing entities and their relationships.

In Computer Science, the term ontology does not usually refer to a global and unambiguous characterization of reality, but instead to the representation of a particular “viewpoint” about a portion of reality. This means that, in Computer Science, there can be many “ontologies,” often “partial” (i.e., referring to a part of the existence). For these reasons, the term is usually written without the capital initial letter and used in the plural. In Computer Science, ontologies are considered more interesting if they can be implemented by “computational objects,” exploitable in software applications.

Despite these considerations, Computer Scientists do not actually agree about the meaning of the term “ontology” and in the literature different definitions of the term “ontology” can be found. In Gruber (1993) it is defined as “an explicit specification of a conceptualization”; Borst narrows the definition by introducing the aspects of formal specification and sharing: an ontology is thus “a formal specification of a shared conceptualization” (Borst, 1997); in (Studer et al., 1998) the authors propose to merge these two definitions: “An ontology is a formal, explicit specification of a shared conceptualisation.” According to Guarino and Giaretta an ontology is “a logical theory which gives an explicit partial account of a conceptualization” (Guarino & Giaretta, 1995). This definition further narrows the meaning, by considering ontologies only logical theories and, in particular, stressing the fact that they are usually “partial” accounts of a conceptualization.

Since this last aspect is one of major importance, in this article we will (mainly) refer to the following definition of “ontology,” which integrates this concept within the previously mentioned definition provided by Studer and colleagues (Studer et al., 1998):

An ontology is a formal, explicit (possibly partial) specification of a shared conceptualisation.

In the following, we will explain this notion of ontology by adopting a rigorous, but informal, approach. For a formal characterization, refer to (Guarino et al., 2009), among the others.

All the previously mentioned definitions mention the notion of conceptualization. A conceptualization is a set of elements, considered as existing in some portion of reality, together with a set of concepts and relationships which characterize (or enable to understand or to describe) that portion of reality, from a particular perspective. Notice that here we use the term “reality” in a broad sense (including physical entities, counterfactual ones, imaginary entities, and so on).

Key Terms in this Chapter

Semantic Search Engine: A semantic search engine in a search engine which exploits semantic knowledge (e.g., ontologies) to improve its “understanding” of the user query, and thus the search mechanism itself, in order to provide better results, i.e., more structured and focused answers.

Web Services: The W3C defines a Web Service as a software system supporting machine-to-machine interaction over a network. A Web Service is characterized by an interoperable interface, described in WSDL (Web Services Description Language), and communicates using the message-based SOAP (Simple Object Access Protocol) protocol.

Linked Data: The term Linked Data is used to refer to a set of methods and best practices for publishing and linking structured data on the Web. Such data are heterogeneous, machine-readable, and can be linked to ontologies.

OWL: The Web Ontology Language (OWL) is an ontology language for the Semantic Web with formally defined meaning, usually exploited to express ontologies. The primary OWL syntax is RDF/XML. OWL specifications are managed by the WWW Consortium (W3C). The current specification, OWL 2, includes three different sub-languages, called profiles , suited to different applicative goals.

RDFS: Resource Description Framework (RDF) is a general-purpose language for representing information on the Web. RDF Schema (RDFS - www.w3.org/TR/rdf-schema/ ) is a specification enabling the definition of RDF vocabularies. The semantic knowledge represented in RDFS can be stored in triplestores and queried through SPARQL, the query language (and protocol) for RDF ( www.w3.org/TR/rdf-sparql-query ).

Inference: An inference is the derivation of logical consequences from formally represented premises. The inference process is usually referred to as reasoning .

Semantic Web: The Semantic Web can be seen as an evolution of the WWW in which machines can “understand” the meaning of the information and services available on it. This goal is enabled by the usage of languages and technologies that support a description of Web resources in terms of concepts and relations they refer to.

Complete Chapter List

Search this Book: