| | |
Summary: 12 1094-7167/03/$17.00 © 2003 IEEE IEEE INTELLIGENT SYSTEMS
Published by the IEEE Computer Society
N a t u r a l L a n g u a g e P r o c e s s i n g
ExtrAns:
Extracting Answers
from Technical Texts
Diego Mollá and Rolf Schwitter, Macquarie University
Fabio Rinaldi, James Dowdall, and Michael Hess, University of Zurich
It's Friday at 12:20 p.m. in Nagoya, Japan, andAkiraWatanabe, pilot of anAirbus pas-
senger jet, is completing the final safety checks before requesting clearance, closing
the doors, and taking off for Sapporo. However, there's a slight problem:Although the first
officer's electronic centralized aircraft monitor (ECAM) is working fine, Watanabe's is
completely blank. Clearly, theAirbus has an electri-
cal problem. Watanabe radios the control tower and
maintenance engineers are dispatched. They must
diagnose the problem quickly, as the passengers are
onboard and getting restless.
The maintenance team's ultimate knowledge
source is a 15,000-page maintenance manual. Using
a keyword-based information retrieval (IR) system,
|
