So, instead, let me give you some web search problems. They all came up in my real life, so they are not artificially created. It seems to me that the Semantic Web should make answering them easier, but with the existing web resources, they are really difficult.

• Find all papers having received best paper awards at ISWC conferences. I did that today, and it took me more than 30 minutes. And I’m not sure if I got all of them – indeed I would have missed one of them if I hadn’t known beforehand about that specific paper having received the award. Isn’t this a typical Semantic Web problem? (The results of my search are further below.)
• There’s an owl-like bird in southern German woods, and in colloquial german it’s called Käuzchen. Try to find out the english name for this bird. I actually failed, though I think I got close to the answer when I merged web search with an external knowledge base (in form of a biologist I happen to know). And actually, simply going to Wikipedia and clicking on the English link is not enough, since I’m not looking for the Strix genus of owls, but rather for a particular bird …
• Who is this researcher with the russian looking name who worked on resolution-based methods for the description logic EL? This also looks like a typical Semantic Search problem, which shouldn’t be too difficult if you have the corresponding knowledge (and background knowledge) available. I admit I failed on this one using traditional methods (unless you consider it a traditional method to ask Franz Baader by email about it.)
• Are lobsters spiders? I.e. are lobsters classified as spiders by biologists? This one is actually tougher than you would think using traditional methods. Should be easy using Semantic Web knowledge bases and some simple reasoning, shouldn’t it?

For all these tasks (and many others), it seems to be apparent that Semantic Web Reasoning – and the availability of corresponding knowledge bases – would make the finding of answers much easier. The current reality of the Semantic Web is still quite a bit away from this. But we’re working on it.

Finally, as promised, the results of my inquiry about the ISWC best paper awards:

• 2004: Y. Guo, Z. Pan, and J. Heflin, An Evaluation of Knowledge Base Systems for Large OWL Datasets.  This one compares different semantic web reasoners.
• 2005: The award was split that year. One half was awarded to P. Mika, Ontologies Are Us: A Unified Model of Social Networks and Semantics. The other half was awarded to B. Motik, On the Properties of Metamodeling in OWL. Edit: Apologies to Peter Mika. I accidently forgot his paper and mentioned instead a paper which won the best paper award at the International Symposium on Wearable Computers 2005, which is also abbreviated ISWC. It’s now been corrected. We definitely need Semantic Search!
• 2006: H. Chen, Y. Wang, H. Wang, Y. Mao, J. Tang, C. Zhou, A. Yin, Z. Wu, Towards a Semantic Web of Relational Databases: a Practical Semantic Toolkit and an In-Use Case from Traditional Chinese Medicine.
• 2007: D. Zeginis, Y. Tzitzikas and V. Christophides, On the Foundations of Computing Deltas Between RDF Models. This one is about a diff for RDF, and it involves RDF inferencing.
• 2008: M. Horridge, B. Parsia and U. Sattler, Laconic and Precise Justifications in OWL.

So why did I dig these awards out? Because I noticed that among these 6 papers there are 3 which are explicitly concerned with OWL. And the 2007 paper involves RDF inferencing. Talk about the importance of reasoning for the Semantic Web …

Author: Pascal Hitzler, AIFB, University of Karlsruhe (TH), Germany

And indeed, at ISWC08 I have seen a number of things which fit exactly into this corner (while at the same time the authors/programmers might not even be aware of it).

• As part of the Billion Triple Challenge, Axel Polleres presented the SAOR system, which does approximate OWL reasoning by means of forward chaining rules. Now you can’t do OWL reasoning (in a sound and complete way) with forward chaining rules (and Axel knows this), so in the end you’re losing some consequences. But at the same time you do get some consequences when having to deal with large amounts of data.
• Eyal Oren, also at the Billion Triple Challenge, presented the MARVIN system which performs approximate RDF reasoning by means of massive parallelisation. MARVIN comes out of the EU project LarKC, which is actually pursuing approximate reasoning on a large scale (pun intended). Edit: This one actually won the 3rd prize at the challenge.
• Among the results presented at ISWC08, I found those by Claudia D’Amato on Statistical Learning for Inductive Query Answering on OWL Ontologies really amazing. She and her collaborators managed to do OWL instance retrieval without any deduction algorithm. Instead they used Support Vector Machines and learned which (named) OWL classes individuals belong to. The learning was done from a small sample set (generated by a reasoner), but the network was able to generalise from the data to achieve about 90% of coverage. In my opinion, this is something conceptually new and it is really remarkable that it works.
• In a regular paper Eyal Oren also reported on using Evolutionary Algorithms for RDF query answering.

The above is only a selection of approximate reasoning related things at ISWC08. There was also the Workshop on Nature inspired Reasoning for the Semantic Web where related ideas were discussed. At the colocated Web Reasoning and Rule Systems conference, RR2008, there will be two papers on approximate reasoning (incidentially, with me as coauthor).

I foresee the importance of such approaches rising substiantially in the future (and I think it’s a safe guess since Frank also seems to think so). The Billion Triple Challenge series could become one of the driving forums for this. There are exciting times ahead!

Author: Pascal Hitzler, AIFB, University of Karlsruhe (TH), Germany