<div class="gmail_extra">Sorry for the late reply; was busy with EulerGUI release.<br><br><div class="gmail_quote">2012/4/14 Kaarel Kaljurand <span dir="ltr"><<a href="mailto:kaljurand@gmail.com" target="_blank">kaljurand@gmail.com</a>></span><br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br> <div class="im"><br> On Thu, Apr 12, 2012 at 14:54, Jean-Marc Vanel <<a href="mailto:jeanmarc.vanel@gmail.com">jeanmarc.vanel@gmail.com</a>> wrote:<br> > How to define OWL datatype properties ?<br> > An example is "Every man has a name." , but is there a way to declare that<br> > name is always a string ? And if yes, does it translate as expected to OWL ?<br> <br> </div>The OWL datatype support in the ACE->OWL/SWRL converter (that is part<br> of APE) works so that if the argument of a transitive verb or<br> of-construct or comparative/transitive adjective is a data item<br> (number or string) then the relation is interpreted as a data<br> property. Examples:<br> <br> /*1*/ Paul has "Paul".<br> /*2*/ Paul's name is "Paul".<br> /*3*/ A name of Paul is "Paul".<br> /*4*/ Every man whose name is "Paul" is younger than 1000.<br> <br> The data property support in OWL goes beyond that: (1) in addition to<br> talking about numbers and strings, one can talk about the more general<br> data ranges (e.g. a number between 13-19), (2) one can talk about the<br> data properties themselves (functionality, keyness). This is not<br> supported in the ACE->OWL/SWRL mapping because we did not find a nice<br> corresponding representation for these things in ACE.<br> <br> Your options are:<br> <br> 1. Augment your ACE text with an OWL syntax that can express these<br> things (e.g. the Manchester syntax)<br> <br> 2. Post-process the OWL form which ACE->OWL/SWRL delivers, e.g. you<br> could tag in the lexicon the words which you want to stand for data<br> properties and data types:<br> <br> $ ./ape.exe -text "Every name is a string." -ulextext "noun_sg(name,<br> dp_name, neutr). noun_sg(string, d_string, neutr)." -solo owlfsspp<br> <br> And then rewrite the resulting OWL:<br> <br> SubClassOf(<br> :dp_name<br> :d_string<br> )<br> <br> into the OWL that you actually want:<br> <br> DataPropertyRange(<br> :name<br> xsd:string<br> )<br> <br> In this case it seems quite trivial but I'm not sure about the general<br> case. If there is a clear and general way to do it and we can come up<br> with conventions how to tag the lexicon then this could be also<br> integrated into the ACE->OWL/SWRL converter. </blockquote><div><br></div><div>I tried this expression that both expresses waht I wads searching, and actually works (with the git version ) :</div><div><br></div><div><div> ./ape.exe -text 'If X v:has-a-name Y then Y is a n:xsd_string.' -solo owlfsspp</div><div>Prefix(:=<<a href="http://attempto.ifi.uzh.ch/ontologies/owlswrl/test#">http://attempto.ifi.uzh.ch/ontologies/owlswrl/test#</a>>)</div> <div>Ontology(<<a href="http://attempto.ifi.uzh.ch/ontologies/owlswrl/test">http://attempto.ifi.uzh.ch/ontologies/owlswrl/test</a>></div><div> ObjectPropertyRange(</div><div> :has-a-name</div><div> :xsd_string</div> <div> )</div><div>)</div></div><div><br></div><div>And this nicely brings me to the next question I planed to ask (I asked it before, but nothing happened AFAIK ).</div><div>What is the best practice for refering to OWL classes and properties in another RDF namespace ?</div> <div><br></div><div>I this particular case, I want to obtain directly ( no hacking and regular expressions ) :</div><div><br></div><div><div>ObjectPropertyRange(</div><div> :has-a-name</div><div> xsd:string</div> <div> )</div></div><div><br></div><div>( I don't particularly recommend the underscore notation, it's just an example; some syntax must be standardized, and the OWL translator has to leverage on it )</div><div><br> </div><div>I consider that this is main missing feature in ATTEMPTO tool chain.</div><div>Defininig an ontology from scratch is the current ATTEMPTO use case, but extending one or more ontologies is also very important ( reuse is good practice ! ).</div> <div>ATTEMPTO was born before the Semantic Web, and it has not yet fully adapted to it .</div><div><br></div><div>I'll enter this in the GitHub site in issues :</div><div><a href="https://github.com/Attempto/APE/issues">https://github.com/Attempto/APE/issues</a></div> <div><br></div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Btw, APE is now on GitHub<br> and easily forkable. </blockquote><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> Everybody's ideas are welcome in the form of<br> GitHub pull requests. :)<br> <br> <a href="https://github.com/Attempto/APE" target="_blank">https://github.com/Attempto/APE</a><br> <br> </blockquote><div>Very good news !</div><div>It should have been announced on the list !</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Best regards,<br> Kaarel<br> _______________________________________________<br> attempto mailing list<br> <a href="mailto:attempto@lists.ifi.uzh.ch">attempto@lists.ifi.uzh.ch</a><br> <a href="https://lists.ifi.uzh.ch/listinfo/attempto" target="_blank">https://lists.ifi.uzh.ch/listinfo/attempto</a><br> </blockquote></div><br><br clear="all"><div><br></div>-- <br>Jean-Marc Vanel<br>Déductions SARL - Consulting, services, training,<br>Rule-based programming, Semantic Web<br><a href="http://jmvanel.free.fr/" target="_blank">http://jmvanel.free.fr/</a> - EulerGUI, a turntable GUI for Semantic Web + rules, XML, UML, eCore, Java bytecode<br> +33 (0)6 89 16 29 52<br>chat : irc://<a href="http://irc.freenode.net#eulergui" target="_blank">irc.freenode.net#eulergui</a><br> </div>