This site, established by Bob Futrelle in January 2001, is to be a collection of resources and links to resources about all aspects of natural language processing (NLP) and in particular its use in information extraction from biology texts. It is a non-commercial site dedicated to research in this field. It is not meant to teach NLP. But it does attempt to give you organized access to resources from which you can learn. I'll also be including links to on-line materials for courses at various schools, so you could actually work through courses in this area, on-line.
And who is Bob Futrelle? I got involved with natural language (digital speech synthesis) via my undergraduate thesis at MIT, 1959 (just a bit early, maybe too early!). My first paper, 1962, was on parsing of natural language (also pretty early). Then linguistics remained a hobby for the next twenty years as I pursued my career, first in theoretical physics, then in biology (ten years on the Life Sciences faculty at U. Illinois, Urbana/Champaign). In the early 80's I got involved with NLP and after coming to Northeastern University in 1986, I set up the Biological Knowledge Laboratory, which I run, starting in 1989, in the College of Computer Science. I have been heavily involved with diagram understanding in more recent years (my current NSF funding). Now I've begun to work more in NLP; this site is evidence in that regard. Lately, I've been trying to develop a comprehensive theory of discourse for documents that have interrelated text and graphics, as most do. A PDF version of our meeting paper on the topic is available at the meeting site, the first paper in Session 3 on this page. (or a cached copy) Another is submitted, and implementations based on the theory are in progress. For more info, see my homepage and my CV.
Although this site is officially known as bionlp.org and can be accessed using http://www.bionlp.org, all the material is kept in my directories at the College of Computer Science (CCS) at Northeastern University in Boston, http://www.ccs.neu.edu/home/futrelle/bionlp/. Using the bionlp.org URL transparently redirects you to the CCS site.
One of the frustrating things about the web is that links often "die". As a general rule, whenever I find a particularly valuable resource, I will copy it to this site and mark it "(cached)". But be warned that such cached material can get out of date. I will not attempt to scour all sites continually looking for updates to material I have cached here. On the other hand, a quick search of the web using a set of the most specific terms from any cached document should locate the current versions or closely related material. I am aware that there can be complex issues of copyright involved when pages are cached on a site. See also google.com's policy on caching. But in the spirit of fair use, basic research, non-commercial operation, convenience, efficiency, etc., I do cache some material.
To build, maintain and improve a site such as this, input from everyone is needed. So get in touch with me with your suggestions and materials. Use email or send a form-based message, no email required.
-- Bob Futrelle
Newton, MA
27 February 2001