<div dir="ltr">Greetings all,<div><br></div><div>You can find the slides for the tutorial that came about with the help of your responses. Thanks very much for your help in putting this together. </div><div><br></div><div>
<a href="http://www.slideshare.net/duluthted/micai-2013tutorialpedersen">http://www.slideshare.net/duluthted/micai-2013tutorialpedersen</a><br></div><div><br></div><div>Like most things this is a work in progress, so any suggestions or comments can certainly be included in the next iteration. </div>
<div><br></div><div>Cordially,</div><div>Ted</div><div><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Sun, Oct 6, 2013 at 10:45 AM, Ted Pedersen <span dir="ltr"><<a href="mailto:tpederse@d.umn.edu" target="_blank">tpederse@d.umn.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Greetings all,<br>
<br>
I'm preparing a tutorial on measuring semantic similarity and<br>
relatedness between concepts, My particular focus is on methods that<br>
do this using ontologies or other (at least somewhat) structured<br>
resources (like Wikipedia, folksonomies, etc.) and that also have<br>
freely available software associated with them (or at least a web<br>
demo).<br>
<br>
While it's a very interesting area, this particular tutorial won't<br>
include purely distributional approaches (due to time constraints), so<br>
I'm looking for methods and software that use some sort of resource<br>
like WordNet, Wikipedia, medical ontologies, Freebase, etc. to arrive<br>
at measurements of semantic similarity or relatedness between pairs of<br>
concepts.<br>
<br>
What follows is my current list, based not only on projects I have<br>
heard of but have used in the not too distant past - so I guess I'm<br>
particularly interested in projects you have used or created yourself<br>
(and can therefore vouch for to some extent).<br>
<br>
Based on WordNet, provide path, depth, info content based measures,<br>
may include relatedness measures like lesk, vector, hso<br>
<br>
WordNet::Similarity<br>
<a href="http://wn-similarity.sourcforge.net" target="_blank">http://wn-similarity.sourcforge.net</a><br>
<br>
NLTK<br>
<a href="http://nltk.org" target="_blank">http://nltk.org</a><br>
<br>
ws4j<br>
<a href="https://code.google.com/p/ws4j/" target="_blank">https://code.google.com/p/ws4j/</a><br>
<br>
Based on UMLS (Unified Medical Language System), provide path, depth,<br>
info content measures, includes relatedness measures lesk, vector<br>
<br>
UMLS::Similarity<br>
<a href="http://umls-similarity.sourceforge.net" target="_blank">http://umls-similarity.sourceforge.net</a><br>
<br>
Based on (GO), provide path, depth, and info content measures<br>
<br>
Proteinon<br>
<a href="http://lasige.di.fc.ul.pt/webtools/proteinon/" target="_blank">http://lasige.di.fc.ul.pt/webtools/proteinon/</a><br>
<br>
I will post a summary of whatever I hear about after some period of<br>
time. Any hints or suggestions will be very gratefully received.<br>
<br>
Many thanks,<br>
Ted<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Ted Pedersen<br>
<a href="http://www.d.umn.edu/~tpederse" target="_blank">http://www.d.umn.edu/~tpederse</a><br>
</font></span></blockquote></div><br><br clear="all"><div><br></div>-- <br>Ted Pedersen<br><a href="http://www.d.umn.edu/~tpederse">http://www.d.umn.edu/~tpederse</a>
</div>