I have a very non-technical team that would like to build an RDF vocabulary. So I thought a little about what I learned about RDF, which is mainly that RDF triples are made out of URIs. Google Sites offers the ability of creating Pages and Sub-Pages that generate URIs such as https://sites.google.com/site/crops/tomato, https://sites.google.com/site/crops/tomato/collecting-source-environment and so on.

Would this be a good way to start building a vocabulary without any extra programming or development, or am I missing something? I imagine my team would simply create a page for each of the vocabulary terms we have.

Eventually I want this vocabulary to be used with RDF. Any ideas? What would happen if I wanted to build relationships such as the ones defined in OWL and RDFS?

asked 31 Jan '13, 11:53

Luca%20Matteis's gravatar image

Luca Matteis
1.5k39
accept rate: 11%

edited 31 Jan '13, 11:54

What do you want to use the vocabulary for? Some use cases would help to guide your choice for many of the issues you are considering. Also, if you can limit the vocabulary to a particular domain, you can build something, create an application that uses the data and demonstrates its value, and then build out from there iteratively.

(31 Jan '13, 12:56) bobdc bobdc's gravatar image

I want to use the vocabulary to document all of our "crop descriptors" that we have. We're a research organization that has worked on these so called descriptors for over 20 years, so we're trying to move forward with technologies and embark into the semantic web. So the particular domain of the vocabulary would be crops. Definitely, we also would like to do things iteratively... we essentially want map much of the crop-related data that exists, using our vocabulary, and be able to make it queriable in the semantic web.

(31 Jan '13, 13:04) Luca Matteis Luca%20Matteis's gravatar image

@Signified all of those tools are too complex for my team. Classes, properties... I just tried using webprotege and neologism and I also couldn't figure it out. I'm talking about a very non-technical team here.

(31 Jan '13, 14:20) Luca Matteis Luca%20Matteis's gravatar image

In a very brief nutshell: RDF triples represent graphs of structured or semi-structured data. The URI is used to provide a unique naming scheme for classes and play a part in constructing triples, but only a part. You should have a domain unique to your organization for URIs. URIs should be persistent as well, this will give you an idea of how that works. Google sites, unless using your own domain name, aren't appropriate.

Read the W3 primer for RDF, it's a good place to start. You might want to hire or contract in help if this is beyond your team's technical ability, taken that an RDF vocabulary is really what you need.

link

answered 31 Jan '13, 19:11

Phil's gravatar image

Phil
1536
accept rate: 28%

edited 31 Jan '13, 19:27

The approach you've chosen for building a vocabulary is a very elegant idea, however this will not result you an RDF graph by default (assuming your team will create human content on these URLs). The URLs would be the nodes, but you need edges too. There are two options to build it up to a graph:

You will need either a triplestore or one should be able to dereference your URLs to download RDF about the resource. The ideal solution is both, but you need to make sure to keep these synchronized. You can check out how the dbpedia was built out: http://wiki.dbpedia.org/Architecture

If you consider this image, here the box "HTML" would be equivalent to your handmade pages: dpbedia architecture

In dbpedia the HTML representation is automatically generated from the database. As you can see, you will need a lot more in order to serve up RDF. Having your own domain (and not leveraging the google URL) has also some benefits, so you should consider some equivalent solution running on your domain.

link

answered 02 Feb '13, 14:24

Barna's gravatar image

Barna
62827
accept rate: 0%

Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "Title")
  • image?![alt text](/path/img.jpg "Title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Tags:

×852
×593
×136
×98
×22

Asked: 31 Jan '13, 11:53

Seen: 1,240 times

Last updated: 02 Feb '13, 14:24