The items mentioned are all a part of Semantic Web technology in which RDF (the Resource Description Framework) is used. Most uses now involve OWL (the Web Ontology Language), a more robust ontology that includes RDF. Here are the basic definitions:

The data (graph) can be stored in a trivial database (TDB) as it is a fairly simple system. The properties are normally indexed.

Dear Amany,

I am not able to completely understand your question. However, here are my two pennies:

A triple (as the name suggests) consists of three elements.

* subject - can be an (URI/IRI)/Blank node

* predicate - can be an URI/IRI

* object - can be an (URI/IRI)/Literal/Blank node

A triple is made up of all these three elements and not the other way around.

You may find all the terms defined In Keven's reply, which gives you an apt overview.

Hi Amany,

It seems to me that you are asking not just how the fundamentals of RDF work, but what RDF /is/. Is that right? I think that because the other answers, however correct, didn't seem to satisfy you.

RDF is the Resource Description Framework. The name is important because it gives a clue to the original thinking of the inventors. RDF is a framework (that is, a model) to describe resources. Which resources? Originally, and still often, they were resources on the Web. That is, Web pages. These days, the (very general) RDF data model is used to describe all sorts of things whether they are on the Web or not.

Here is an important, but subtle and underused, point: You cannot say anything in RDF less than one triple's worth of information. That is, a single triple is the least content possible in an RDF graph.

RDF is therefore an odd sort of data model in that you cannot just name a resource explicitly. You can only name it implicitly by assigning it an IRI used in a triple. The triples that refer to a resource can collectively provide enough information about the resource that everyone (should) agree on what the resource is.

A triple has three components, as Keven and Harsh described. The easiest way to think about an RDF triple is to note that triples are one of two kinds:

1) A triple may describe two /things/ and the relationship between them. Here is an example of a single RDF triple that names two things and links them together with a relationship:

(the first thing, me) (the relationship, in this case "authored") (the second thing, this answer)

You can read that triple as "David Hyland-Wood authored this answer".

In this case, IRIs are used to name the two things and also to name the relationship. That's almost always the way that RDF triples are written.

2) A triple might also provide a piece of data about a resource. In this case, the second thing has some form of data type, such as a string or a number or a date. That ability allows the RDF model to contain measurements, human-readable descriptions, and such. It is very useful when you want to describe your resources in great detail. For example:

(the first thing, me) (the relationship, in this case "has eye colour") (the date "1963-08-28")

You can read that triple as "David Hyland-Wood was born on 28 August 1963".

So to get back to your original question, we use IRIs to represent things (nouns) and relationships/concepts (verbs). We use other non-IRI data types such as strings, numbers, and dates to represent data.

If we have a lot of RDF triples describing resources, they form a conceptual graph because we presume the IRIs are unique in the world. If there are two RDF triples that both start with the same IRI, then we say they are giving two pieces of information about the same thing.

The edges of RDF graph are the data, because the RDF model doesn't allow you to make any further statements about a string, date, number, etc. used in a data field. You can make all the statements you want about things that are named with IRIs.

Does that help?

Regards,

Dave