Comunicación

The practice of Evidence Based Medicine (EBM) requires the
physician to access the most up to date clinical evidence
reported in the medical research literature. However
good practice of EBM is seriously affected by the
enormous amount of medical literature available and
the limited time that the physician has. In this talk I
will focus on my research group's current work on natural
language processing for EBM. We address the task as one of
query-focused multi-document summarisation where a clinical
question retrieves several documents, and the clinical
evidence must be extracted, summarised and presented in
the most effective manner. We have recently gathered a
corpus of clinical questions, their evidence summaries,
and the related references, and we are currently using it
to produce and test various tasks related to the ultimate
goal of finding and summarising the evidence.

Dr. Diego Molla-Aliod is a senior lecturer at Macquarie
University. His research focuses in query-based information
extraction systems. He was the principal researcher in the
ExtrAns answer extraction system of unix documentation
at University of Zurich, and then the project leader of
AnswerFinder, a question-answering system developed at
Macquarie University. Since 2009 his research has focused
on the medical domain and recently he has made available
a corpus for summarisation in evidence based medicine. He
is a co-founder and past secretary and president of the
Australasian Language Technology Association (ALTA), and
was the leader of the Search Technology priority area of
the HCSNet ARC Research Network.