bigdata, OWL, SWRL
Now that bigdata is handling billions of triples with ease, we are ready to venture into higher expressivity as well. There is always a tradeoff between the expressiveness of the ontology and the computational complexity of computing the entailments. So far, bigdata has focused on the data scale, now we are ready to look at the reasoner complexity. To do this we are exploring some integration options, including partnering with Clark & Parsia to develop an integration with the Pellet2 OWL reasoner [2].
Speak out and let us know what combination of data scale and ontology complexity you need. Do you want datalog [1], OWL2 profiles (RL, QL, EL)[3], Horn-SHIQ[4]? Do you need SWRL [5], and how you want to use it? Example ontologies, data scale and use caselets would all help.
[1] http://www.iris-reasoner.org/
[2] http://clarkparsia.com/pellet/
[3] http://www.w3.org/TR/owl2-profiles/
[4] http://logic.aifb.uni-karlsruhe.de/wiki/Horn-SHIQ
[5] http://www.w3.org/Submission/SWRL/
Speak out and let us know what combination of data scale and ontology complexity you need. Do you want datalog [1], OWL2 profiles (RL, QL, EL)[3], Horn-SHIQ[4]? Do you need SWRL [5], and how you want to use it? Example ontologies, data scale and use caselets would all help.
[1] http://www.iris-reasoner.org/
[2] http://clarkparsia.com/pellet/
[3] http://www.w3.org/TR/owl2-profiles/
[4] http://logic.aifb.uni-karlsruhe.de/wiki/Horn-SHIQ
[5] http://www.w3.org/Submission/SWRL/
posted by Bryan Thompson at
8:15 AM
5 Comments:
Thanks for a great project, I'm impressed with what you've created to date.
I would like to see OWL DL support. I'm currently looking at a solution involving OWLIM for reasoning, based on its performance. But due to the single-threaded nature of reasoning/writes in OWLIM, I would prefer a solution that allowed for concurrent updates of the entailments and concurrent reads. If I could do the inferencing and persistence with one solution (e.g. bigdata), that would be preferable.
We are building a system that essentially is a data center graph and we'd like to be able to navigate from node to node and view all nodes in a network path, using our ontology, associated asserted and inferred statements and SPARQL (e.g. find all nodes of type X connected to node Y).
We are using inverse, inverse functional and transitive properties and would like to use owl:intersectionOf and owl:unionOf.
Based on my understanding of the OWL profiles, OWL RL support is what we might be needing.
Chris,
Higher expressivity is squarely on our radar. We've been talking with Clark & Parsia about options for supporting OWL profiles. We may do a simple proof of concept soon, but ideally we will find a customer with pressing high expressivity needs to fund a full-featured integration sooner rather than later.
You may want to speak with Clark & Parsia directly now to see if they have a solution currently that might fit your needs.
Thanks,
Mike
Thanks Mike. I will contact C&P.
Once I dig a little deeper into bigdata I'd like to have a discussion about our needs and your roadmap.
Partnering with Clark and Parsia would be a huge win IMHO. Pellet is a really great reasoner and not having it work with Sesame out of the box has been a bit of a bummer. If you combine that with their work on OWL integrity constraints, it gets even better.
Ryan-
What is the current state of your exploration of integrating OWL reasoning?
I'm constructing a system for enterprise information integration. A key part is the knowledge warehouse that should be based on RDF and OWL. I would like to use bigdata because of it's impressive data scale feature but require OWL DL support.
As Chris already said, bigdata is a great project. Thanks for your work.
Post a Comment
<< Home