If you have been paying close attention, you'll see that I have added a "new" research area on my list of active research projects on my home page - that's Medical Informatics.
Of course this isn't all that new, many of you have been involved in some of that work, and I guess what I finally realized is that the volume of work I've done in that area over the last few years is enough to merit listing it alongside other projects like SenseClusters, WordNet-Similarity, SenseRelate, NSP, and Supervised WSD (the big 5 I like to call them). Well, now there are six, you'll see that on my home page, and I have also created a page that lists my medical informatics publications separately (like I have for the big 5). http://www.d.umn.edu/~tpederse/medical-pubs.html You'll note there two publications from 2007, one is an article in the Journal of Biomedical Informatics based on work that Sid, Serguei and I have done over years, starting with Sid's Mayo internship way back in 2003. Measures of Semantic Similarity and Relatedness in the Biomedical Domain (Pedersen, Pakhomov, Patwardhan, and Chute), Journal of Biomedical Informatics, 40(3), 288-299, June 2007. http://www.d.umn.edu/~tpederse/Pubs/jbi2007.pdf The other is a paper by Bridget, Serguei and me that extends some of her MS thesis work and Mayo internship work into the medical domain. This will appear at the BioNLP workshop at ACL, which turned out to have a fierce acceptance rate, about 24%, so that acceptance is very impressive (and the credit for that lies with Bridget.) Determining the Syntactic Structure of Medical Terms in Clinical Notes (McInnes, Pedersen, and Pakhomov) - Appears in the Proceedings of BioNLP 2007, June 29, 2007, Prague, Czech Republic. http://www.d.umn.edu/~tpederse/Pubs/bionlp07.pdf Medical Informatics will remain an active area for me as it is the focus of Bridget's PhD dissertation in the Twin Cities (I am her co-advisor on that, with John Carlis) and Serguei and I continue to work on things together, particularly the issue of using measures of similarity and relatedness in the medical domain. In some ways you can think of our goal there to port WordNet-Similarity to UMLS...you see some preliminary steps in that direction in the above mentioned JBI article. Anyway, stay tuned. :) Ted -- Ted Pedersen http://www.d.umn.edu/~tpederse
