Output details
11 - Computer Science and Informatics
University of Sheffield
Return to search
Output 0 of 0 in the submission
Output title
Authoring Technical Documents for Effective Retrieval
Type
E - Conference contribution
Name of conference/published proceedings
Knowledge Engineering and Management by the Masses. Lecture Notes in Computer Science
Volume number
6317
Issue number
-
First page of article
287
ISSN of proceedings
16113349
Year of publication
2010
Number of additional authors
1
Additional information
<27>In technical domains such as aerospace, terminology is very complex and authors introduce large numbers of term variations, raising significant issues for information retrieval. We describe a novel approach that uses high accuracy terminology recognition (TR) and interaction during authoring to ensure terms are linked to unique concepts in a domain ontology. Experiments on Rolls-Royce data show the TR method outperforms the state-of-the-art by 100%. Rolls-Royce filed a patent on the methodology (EP2400401) and now use it as a core component in a data mining program. It has saved £millions and has been delivered to over 10,000 engineers (colin.cadas@rolls-royce.com).
Interdisciplinary
-
Cross-referral requested
-
Research group
None
Citation count
1
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
