1999), studies have shown that existing search sys-
Current question answering systems aspire to
tems, e.g., PubMed, the U.S. National Library of
address this shortcoming by gathering relevant
Medicine’s search engine, are often unable to sup-
“facts” from multiple documents in response to
information needs. The so-called “definition”
ply physicians with clinically-relevant answers in
or “other” questions at recent TREC evalua-
a timely manner (Gorman et al., 1994; Cham-
tions (Voorhees, 2005) serve as good examples:
bliss and Conley, 1996). Clinical information
841
Disease:Chronic ProstatitisIanti-microbial
Bạn đang xem 1999) - TÀI LIỆU BÁO CÁO KHOA HỌC ANSWER EXTRACTION SEMANTIC CLUSTERING AND EXTRACTIVE SUMMARIZATION FOR CLINICAL QUESTION ANSWERING PDF