Publications

Biomedical Literature text mining is the extraction of useful information
from large volumes of unstructured biomedical text.  The text mining process
can identify what biomedical S&T is being performed globally, who is doing
it, and where it is being performed.  Additionally, a major cutting-edge
application of text mining, that has been applied to the medical literature
only, is discovery and innovation from complementary literatures.  In total,
text mining is the essential core of any successful document-based
Technology Watch activity.

The document (see links) overviews the status of biomedical literature text
mining, within the broader context of science and technology (S&T) text
mining.  It concludes that there are deficiencies in each of the four
areas required for successful science and technology text mining:
motivation, personnel, protocols, and databases.  It recommends some
near-term steps necessary to overcome the deficiencies.

Biomedical literature text mining in general, and the attached document in
particular, should be useful to clinicians, lab experimentalists,
theoreticians, and other people with research, development and acquisition
responsibilities, in outlay/ intelligence/ documentation and other medical
informatics activities.  This includes S&T managers, evaluators, performers,
administrators, and biomedical literature editors and publishers.

RNK
http://www.dtic.mil/dtic/kostoff/index.html
http://www.sciquest.com/cgi-bin/ncommerce3/ExecMacro/sci_kostoff2.d2w/report?nav_banner=special&Tmstmp=25884