Garo Garabedyan's Divergent Thinking Blog

Semantic Web

leave a comment »

I like the above interview and I have decided to write down the foundations of Semantic web according to it as a struggle to make the first popular.
Software Engineering Radio Episode 116: The Semantic Web with Jim Hendler

We have spent a lot of our times in thinking of very fast ways to explain Semantic web and the best I like is:
Right now, if I have a document, I can put it on the web and instant point to any other document with a simple link. But if I have a database or a spreadsheet I can not put that on the web and have it point another database or spreadsheet.
So, the sense is figuring out what is the database and the infrastructure by which data and definitional information can comes to the web rather than these documents is the heart of the Semantic web.

Web for machines, software, programs:
The key concept is right now the information on the web is primarily described for human inns, but if I am using data or something like that it is very hard to describe that for human use, because what I meant to do is aggregated, collated, queried, all things that pure happen back behind somewhere in the infrastructure, than when I present it to the human it is going to look like a table or a text or a web site, so it is really about building a new infrastructure for the web, to make far more the data work together and of course that will require more machines and machine integration things.

Three things you need:
1. Data format that can include links- If I need to integrate DBs I need someone to knowing what can be merged and what can not be merged, which things link to which other things. RDF is primarily the language that gives you that. So we need a standard for the kind of data description.

2. Means for describing terms of that data- If I have a DB, I look at a column called C27 column 19 and I see that it is filled with the number 15, that doesn’t help me much, if I want a merge with another DB. If I know that that represents the age of a person, then I can find the another DB which have either other properties of that person or other ages of people. Now I know how to put them together. An Ontology or Domain description in a formal way to represent it.
These Ontologies define the meanings of things and here is where the term semantic comes.
If I want to link data together I need to know what is it about, and it is often about something that is not in the DB.
We need to take data out of the database into a place where it can be merged or you need query dynamically the DB.
RDF gives you data in triple stores- graph. If you take two table you can not guarantee that you will get table as a result of merging, if you have two trees I can not guarantee that I will have a tree as a result. Merging graphs always gives you graphs.
Directed labeled graph is RDF.
OWL (higher level data schema) to represent the Ontologies.

3. Standard way to query RDF- SPARQL.


Written by garabedyan

July 27, 2009 at 19:07

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s