Semantic Web

Sonia Sánchez-Cuadrado
 Incorporated contributions
Jorge Morato, Yorgos Andreadakis, Anabel Fraga
 Usage domain
Web sémantique
 German Semantisches Web

Tim Berners-Lee created this concept by proposing a network in which information and services are semantically defined, so that requests of both people and machines could be equally understood and satisfied. Computers would be able to analyze all Web data: content, links, transactions between persons and computers. A Semantic Web that could be able to do this is emerging; when this is possible there will be a qualitative leap in the interconnection between multiple repositories, electronic commerce, semantic queries and automatic question-answer systems.

Semantic Web has already been with us in a decade and a great effort has been invested for its development by private and academic entities; regretfully results are currently scarce, because this forward-looking approach implicates a “technical construct of protocols, process, languages, and tools.

However, three factors promote the Semantic Web as an attractive solution; these are interoperability, and the creation of semantic resources with common domain knowledge:

1.  Interoperability: some authors consider the Semantic Web as a project for creating a universal mediator for information interchange (Kalfoglou, 2007). This would be possible through the creation of interoperable documents semantically well defined for the computer applications of the World Wide Web. In other words, it is about converting the Web, and its distributed databases, into one great database. Interoperability between the documents is sustained through the use of a common language based on RDF (Resource Description Framework) (W3C, 2005), a language which is based on XML (Extensible Markup Language) (W3C, 2006). The advantages of obtaining this interoperability are obvious for knowledge reuse (Russ, Jones and Fineman, 2006), conceptual navigation, and the fusion of Knowledge Organization Systems (KOS) through multiple domains (W3C, 2006; Zeng, 2004).

2.  Semantic Resources: Semantic Web requires that the semantic knowledge should be expressed in documents written in a Web language oriented to knowledge modeling, like RDF. These documents model KOS and its instances. KOS have an important role in the Semantic Web because they support semantic knowledge management. This is necessary to perform indexing and retrieval tasks, giving as a result more relevant and noiseless information for the user. KOS define the concepts utilized for describing and representing an area of knowledge (Daconta, Smith and Obrst, 2003; Gruber, 2005). These resources are used by persons, databases, and applications that need to share information on a specific domain,  such as medicine, real estate, commercial management, etc.

Semantic Web Layers

Semantic Web proposal suggests a seven semantic web layer “cake” for its implementation. Each layer has to be compatible with previous ones. For instance, the layer three -RDF/RDFS- must be understood by XML applications, while the next layer, usually encoded by the Web Ontology Web Language (OWL), must be able to extract information from RDF documents. 

Although the following layer scheme corresponds to the most well known version of the Semantic Web Layer Cake (also known as Semantic Web Stack):

  1. Unicode-URI: basic coding and resource identification.
  2. XML-NS-XMLSchema: syntax and its relationship with name-spaces, also expressing XML structure.
  3. RDF+RDFSchema: RDF/RDFS shows primitives to represent knowledge.
  4. Ontology Vocabulary: structuring and classifying concepts by an ontology. 
  5. Logic: axioms and monotonic rules.
  6. Proof: to validate assertions
  7. Trust: trustworthiness of the data.

Other versions has also been proposed (Bratt, 2007).

Topic Maps standard and Semantic Web

The Topic Maps standard constitutes a proposal similar in purpose but has earlier roots. This standard was proposed at the beginning of the 1990s, although  the proposal have been updated to recommend XML encoding and other improvements. Although this proposal has lower inference capabilities, it is more intuitive. This standard initially used Public Subject Indicators (PSI), instead of other metadata vocabularies.

Related Resources 

New entry. Before doing a new entry, please, copy this line and the following ones and paste them at the column bottom. Next fill out the fields: 'name', 'date' and 'text', and delete this upper blue paragraph.
Name (date)
[Entry text]

Incorporated entries
Jorge Morato (1/4/2010)

Jorge Morato (27/9/2009)
[It corresponds with the first version of the article, which is now showed in the left column.]