email: sht25@cl.cam.ac.uk.
Research
My research is on summarization, IR, and on the use
of NLP techniques for web-based applications. My PhD (Argumentative Zoning: Information Extraction from
Scientific Articles) was concerned with robust rhetorical
processing of scientific articles, in order to facilitate
scientific-specific searches. The integration of citation indexing, IE
techniques and more traditional summarization techniques results in
"rhetorically zoned" articles. I am also interested in cognitive
experiments to prove the use of this type of robust processing in a
real user environment. Another ongoing interest is in evaluation of
summarization systems (which is a hard problem plaguing the
community), particulary task-based evaluation.
My current research plans involve robust generation from sentence
fragments and the integration of IE, IR and citation indexing for the
specific searches that a scientist might want to do on large
repositories of scientific articles.
Projects
I am/was involved in the following projects at the Computer Lab:
My publications are online here.
Teaching
I have taught two halves of two courses this year,
"Computing
and the Web" and Internet
Applications on the CSTIT MPhil course,
which is run jointly by the Engineering Department and the Computer
Laboratory.
Biography
I got my PhD from CogSci in 1999, where I was mostly concerned with summarization and the structure of scientific
arguments (cf. Argumentative Zoning). I also worked at the HCRC Language Technology Group.
During a Postdoc at Columbia University (2000-2001), I worked on the
Digital Libraries Project PERSIVAL whose
aim it is to provide patient-specific access to large collections of
scientific articles, amongst others. In a subpart of the project, we
reranked the output of searches in the field of cardiology to those
articles which are of relevance to one particular patient the
cardiologist is currently considering. I also worked on the TIDES
project on multilingual summarization at Columbia.
Previously, at IMS, University
Stuttgart, where I got my first degree, I was involved in the work
of the
EAGLES corpus and lexicon standardisation group. Part of my work
was a tagset conversion tool, another part the testing of the
interaction of text type, used lexicon and automatic tagger on the
results. I also spent some time at XRCE Xerox in Grenoble, working on the
extraction of nominalizations and collocations.
3
Simone Teufel
Created: October 29, 2001