I am pleased to report that Jason has submitted the final version of his thesis, entitled "Semantic Relatedness Applied to All Words Sense Disambiguation". You can find that online here:
http://www.d.umn.edu/~tpederse/Pubs/jason-thesis.pdf Jason is now officially graduated (yippee!). Before he moves on it seems appropriate to mention his very significant accomplishments, most of which involved WordNet-Similarity, which as you may know may be our most widely used tool in the NLP community. Jason spent much of 2003-2004 engaged in an overhaul of the internals of WordNet-Similarity, which I know was very taxing work at times, and certainly a bit tedious as well (especially when I got into one of my "why is the semi-colon here and not there" moods, which happens more often than I'll care to admit). In addition to re-engineering the package, he also arrived at a very principled way of handling strange situations like multiple inheritance, which have largely been ignored in previous discussion of the measures we support. In 2004-2005 the mission was to use WordNet-Similarity, and Jason arrived at a method of all words disambiguation which includes a number of very nice features. This was always the missing link in our tool set here - we have many many methods for lexical sample disambiguation, but until this work we had nothing for all words. That hole has been plugged now, and I am getting reports from the field that this method is catching on and attracting users. In short, Jason accomplished a lot in two years, and his work here has clearly had an impact on users around the world. That's a rather good legacy I think. :) Congratulations! Ted -- Ted Pedersen http://www.d.umn.edu/~tpederse Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/nlpatumd/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
