RDFS/OWL

Supported ontology formats Change types RDF/OWL, further formats via import plugins RDF/OWL... 

In the Semantic Web, the tasks of (i) validating RDF data against a set of integrity constraints and (ii) performing RDFS/OWL reasoning, are grounded on different semantics. While the latter operates under the Open World Assumption (OWA) (i.e., a statement cannot be assumed to be false if it cannot be proven to be true) and the Nonunique Name Assumption (nUNA) (i.e., the same object/resource can have... 

Introduction of XML formats was a very important step toward better intercomputer communication, but it is not a miraculous solution to all problems. Not every possible relation is easily expressed in XML (Wang et al. 2005), so specifications usually contain many implicit assumptions that are not properly formalized. The Resource Description Framework (RDF) provides a very powerful yet simple model for this formalization (Manola and Miller 2004). In this framework, any information is transformed to basic units called triplets that are combined to map the available information. This unifying mechanism can be used to express hierarchical vocabularies for domain knowledge description, as in RDF Schema (Brickley and Guha 2004) or its extension, Web Ontology Language (OWL) (Smith et al. 2004), both standardized by the W3C. 

BioPAX (http //biopax.org/), the group developing a common exchange format for biological pathways data, uses OWL. BioPAX can capture molecular binding interactions and manage small molecules (represented by InChl). Another example of RDF usage in chemistry is provided by the CombeChem project (Taylor et al. 2006). 

In this chapter we shall first describe RDF and OWL, the grammar of the Semantic Web, and then move on to identifiers, the vocabulary of the Semantic Web, and ontologies, which represent the real-world knowledge required to make use of the grammar and vocabulary. We will concentrate on practical chemistry- and biochemistry-orientated deployments of Semantic Web technology rather than the computer science behind it. Then we will cover some case studies Web services, databases, and semantic publishing in the forms of Semantic Eye and RSC Project Prospect. We will briefly cover what the Semantic Web has to offer for experimental data before finishing up with some possible future directions. 

As for representing scientific articles as a whole purely in terms of OWL, this is more complicated than might be supposed. Attentive readers of scientific texts will notice that there is a subtle and comprehensive system for making assertions in terms of possibilities and necessities rather than the simple subject-object-predicate triple of RDF. Authors use phrases such as It is not impossible that or We can assume that to distance themselves from absolute certainty and to weaken or strengthen their statements. 

The KAON system [871] was chosen instead. It is based on the Resource Description Framework (RDF) that also forms the base of OWL. KAON enables semantic queries directly on the backend repository (stored in a relational database) by transforming the query into SQL, at the cost of loosing some of the expressiveness of OWL. Translation of OWL ontologies, such as OntoCAPE, into the KAON system has been realized, based on their common RDFS characteristics. 

The underlying data format that is used for specifying the ontology also introduces potential problems. The language used (e.g., OWL, RDF, XSD) constrains the expressiveness of the data representation. For example, many formats lack information relating to units of measure or intended usage [Bernstein and Melnik 2007],... 

OWL (see earlier) is a more expressive ontology language than RDFS and can be used for building terminologies that can then be represented in RDFS. 

RDF itself can be used to create vocabularies, and the most prominent of these is RDFS. RDFS allows descriptions of classes, subclasses, and arbitrary relationships between classes. This means that ontologies described using OWL can also be represented in RDFS and that tools built to support OWL will also support RDFS. 

Table 3.2 Overview of the key RDF/RDF(S)/OWL modeling constracts described in this chapter and their definitions adapted from the corresponding language ref nce documentation...

The Web Ontology Language (OWL) ° was introduced as a more expressive ontology language than RDF(S). Since 2012, W3C recommends the use of OWL 2, an extended and improved version of OWL 1. OWL 2 enhances the expressivity of RDF(S) in several ways. We provide a few examples of OWL 2 constructs that are used in the chapters of the book, but refrain from a complete introduction to this language. 

For classes, OWL 2 provides constructs beyond simple subsumption hierarchies declared with rdfs subClassOf. One can specify, e.g., classes that have exactly the same set of instances with owl equivalentClass or specify classes that share no instances at aU with owl dis j ointwith. Further, complex classes can be declared as intersections of (owl intersectionOf), unions of (owl unionOf), or complements of other (sets of) classes (owl con jle-mentOf). The meaning of these modeling constructs is equivalent to that of the corresponding set theory operators. 

Local Data Model Definition. Based on our experience in the domain, the W3C standard of RDF(S) and OWL languages are sufficient to define local data models. For a more detailed explanation of data model definition and some examples of engineering ontologies, we refer readers to Chap. 5. 

---