Wednesday, April 16, 2008

What is the Semantic Web?

I just finished working on a poster about my research for the Research Forum at UT. The public will be the College of Natural Sciences in general, so it wasn't worth getting to technical. I knew this was my opportunity to do a great "What is the Semantic Web" poster, and then on the side talk about my research. And now, I have the opportunity to share with y'all what the Semantic Web is, or at least, what I consider the Semantic Web to be. I would really appreciate your comments!

Obviously, we need a definition and this is mine: The Semantic Web is an extension of the current web which provides a common framework that allows effective discovery, integration and reuse of data in which the semantics of the data is defined, making it possible for the web to understand the request of people and machines to use the data.

So what next? Oh... that famous Semantic Web Layer Cake. As long as you know what URI, XML, you can then research what RDF, RDFS and OWL is and even understand it! If not, I'll explain it very briefly.

A URI (Uniform Resource Identifier) is a compact string of characters used to identify a name or a resource. A URL ( is a URI. When I met Tim Berners-Lee, he specifically said that URL should not be used anymore. Instead, everybody should use URI. If you have a problem with that, you should take it up with him. XML (eXtensible Markup Language) is like HTML, but you can also make up your own tags.

RDF, RDFS and OWL deserve their own post, and that will happen in the future (maybe next week). Briefly RDF is a framework for describing resources. What??? RDF represents triples. What??? If you the sentence "Juan is a Student", this represents a subject-predicate-object, and that is the triple that RDF represents. RDFS and OWL are ontology languages. So what is an ontology? The best definition I have seen is a "specification of a conceptualization". In other words, it represents the concepts within a specific domain and the relationship between the concepts.

Well, now that I kind of explained the layer cake, let me go on. In this poster I wanted to show some how what the Semantic Web is going to be. So obviously the terms Web 2.0 and Web 3.0 came to mind. The three-tier architecture represents the Web 2.0 while the Semantic Web will be part of the Web 3.0. I don't believe that it is the Web 3.0 There are a lot of definitions of what Web 3.0 and that also deserves another post. But with that in mind, I thought of this image:

Somehow what I wanted to do explain that the current three tier architecture that supports the Web 2.0 should be integrated with the Semantic Web. The problem is that the layer cake is very overwhelming and so complicated to explain. The interesting thing is that the three-tier architecture is very similar to the following...

If you compare the bottom layer of both images, you can see that the three-tier architecture only has databases while the Semantic Web has a web of data (html, xml, relational databases, etc). In the application layer, RDF and the ontologies need to integrated with the application layer of the three-tier architecture. Applications will use SPARQL and inference engines to do the sweet stuff of the Semantic Web. The red stuff shows my research interest. There is no standard way of integrating relational database content with the Semantic Web, but again, that is for another post.

Another way to explain how the Web 2.0 should be "transformed" to the "Web 3.0" (or Semantic Web)...

And this is left for an open discussion.

No comments: