The Semantic Web in Science and Education

Wednesday, January 21st, 2009

Innovation & ICT, Web 3.0

(77 votes, average: 3.71 out of 5)

Loading ...

The world wide web is growing every day and is getting ever more complex. Search engines, robots and algorithms have reached an impressive complexity and ability to create relevant results to search terms but are still not able to answer concrete questions. When asking Google “what is the fastest way from Barcelona to Lyon right now” the results would not take different vehicles and routes into account and combine them to current events like traffic jams or bad weather to propose useful decisions for web users. While the information is all there already, it often needs a lot of human effort to bring it all together to solve a simple task.

southpark

Picture from http://ablvienna.wordpress.com/2008/04/17/ it%C2%B4s-a-web-of-data-stupid/

The semantic web is aimed to solve this problem of missing contexts. It is not a new trend but a vision that has been existing for a long time and can already be found in very early web theories. Tim Berners Lee, one of the founders of the world wide web is cited as one of the fathers of a semantic web, often also referred as a part of a Web 3.0 (see Berners-Lee, Tim/Fischetti, Mark, 1999: Weaving the Web: The Original Design and Ultimate Destiny of the World Wide Web by Its Inventor).

A Web of Meaning

A semantic web is supposed to be a web of meanings not only a web of content. It’s a web of data not only data on the web. It is the idea of intelligent systems that take contexts and situations into account to help solve problems or answer questions and which will also give science and education new ways of dealing with knowledge.

Coming closer to a semantic web of meaning and context, we will see developments from two directions: First, Markup Languages which set standards about how to declare meta-information and second, intelligent software, or “agents”, which is able to make use of these meta-information standards.

Meta Information standards and agents

To understand human language, software needs meta-information in a certain standardized way to be able to distinguish between situations and to refer to contexts. It needs ways to create context themselves to create hierarchical relations between words – like a truck is a car but a not a SUV or compact.

Such semantic classifications are added to the net as metadata, not meant to be read by humans but by machines. Thus, they need standards like the Web Ontology Language (OWL) or the Resource Description Framework (RDF), languages readable by machines to describe multimedia contents formally. Very few parts of the web of today exist with this kind of semantic description. These frameworks, very basically explained, pretty much work like human language, using so called “triples”: Subject – predicate – object: Hans (subject) has the address (predicate) Mainstreet 4, London (object).

Software like browsers, search engines, web sites or other applications can then make use of this meta information to understand plain questions, combine information or to keep track of developments.

Sematic web for dummies

From http://web.media.mit.edu/~stefanm/ commonsense/SemanticWeb.ppt

Semantic Web in Science and Education

The semantic web is mainly a scientific subject still. The structures for different ways to declare content are being developed in different workgroups consisting of researchers, developers, scientists, engineers, and the W3C, yet it is clear that no single framework will be able to classify all web data available. There is a need for structural organization of the context of learning on the net, also there is a metatag-dilemma for scientists and educators. How will all the meta tags for human knowledge be created and maintained?

Rather than expecting a universal semantic web, it is arguable that there will be different systems for different knowledge domains such as medicine e.g. Different fields of knowledge will have to find their own ways to step into the world of semantic information until it might one day be joined altogether.

Moreover, some visions of a educational semantic web already can be described: Knowledge construction will hopefully work in a different way with semantic web technologies, helping researchers gathering useful information much faster and more efficient than today. With personal learning networks, consisting of people and agents connected via semantic information makes dynamic learning much easier and might connect people who work on the same subject but would not have had any idea. Moreover, learning and educational organization such as courses, classes, universities, tests, working groups, will be easier to organize with agents to automatically track e.g. events, deadlines and progress.

One thing should not remain unsaid in the end still: the learning, the knowledge creating and the writing of course, will not be done by machines, even in a semantic web age.