SPARQLing RDF Repositories

I bet that the title is the first time anyone has ever attempted to make such a geeky play on the SPARQL acronym.

This post is to list a few handy findings I have just made that could be pursued more next year in my research:

That’ll do for now.

Thoughts on things Semantic-y

My current area of work is very much in improving my research project by replacing its current data layer and visualisation layers. In their place I will be writing a significantly more comprehensive layer, built atop rising standards, particularly OWL (the ‘web ontology language’). Therefore, my focus is very broad at the moment, as I must focus on the following things:

  • Connecting to the current databases.
  • Allowing for new and additional databases to be merged.
  • Abstracting all databases using a simple server that allows SPARQL queries, and automatically maps these queries to SQL, (see D2R Server).
  • Allowing for the project to query these databases, and to build up an internal OWL ontology, (see Jena).
  • Allowing for this OWL ontology to be visualised using a new visualisation engine, (see prefuse).
  • Allowing plugin writers to tap into the new OWL ontology to make more powerful plugins.
  • Allowing for the user to easily make use of this new functionality.

The main work is in linking these three technologies together. A layer between my project and the server will need to be written to allow SPARQL queries to be generated ‘on-demand’ and sent to the server. Server responses will need to be added into the clients OWL ontology. The prefuse visualisation will need an adapter written so that it can use the Jena data model natively, instead of having to duplicate its data wastefully.

Once this is done, writing plugins will be relatively easy. It is my goal to achieve this, and a few ‘to-be-announced’ plugins as my masters project next year.

More later.

Integrating relational databases with RDF/OWL

I wanted to keep tabs on the blog post that I am linking to here. It is a small portion of what I am about to be working towards as part of my masters project. Basically, this blog post talks about how you can query a database using SPARQL (a semantic web query language similar to SQL). To do this a mapping has to be made between the SPARQL query and the SQL query, which this blog writes about, and it points to a number of interesting applications that I have looked into in the past, but never really followed up (too busy with other projects I guess).

I guess I’ll clarify this post in the future 🙂

New Blog

Ok, This is my new blog – I think this’ll be better for me as it means I have a lot more control over the underlying blog software (compared to what I had on my MSN Live Spaces blog). If you’re wanting any particular content in here, let me know – you should all know my particular areas of research (but to bring everyone up to date, my research is particularly surrounding plugin-based software, and I am just beginning research into semantic web technologies for the next few years).

Hopefully over the next few days this site will be finished off, then I need to focus some attention on my other important website: the website for my company Novix Ltd.

Update: Well, all previous blog entries have had their formatting updated, and now they look much better. I still have a bit more work to do, but overall wordpress is a good piece of blogging software.

Cheers,
Jonathan.