Motivation: Chemical named entity recognition is used to automatically
identify mentions to chemical compounds in text, and is the
basis for more elaborate information extraction. However, only a
small number of applications are freely available to identify such
mentions. Particularly challenging and useful is the identification of
IUPAC chemical compounds, which due to the complex morphology
of IUPAC names requires more advanced techniques than that of
brand names.
Results: We present CheNER, a tool for automated identification of
systematic IUPAC chemical mentions. We evaluated different systems
using an established literature corpus to show that CheNER
has a superior performance in identifying IUPAC names specifically,
and that it makes better use of computational resources.
Availability: http://metres.udl.cat/index.php/9-download/4-chener,
http://ubio.bioinfo.cnio.es/biotools/CheNER/
Supplementary information: Both web sites above include the
user manual for the software. Supplementary materials accompany
this publication.
AU is a Ph.D. scholar funded by Generalitat de Catalunya. AU, FS & RA are partially supported by the Spanish government (grants TIN2011-28689-C02-02 and BFU2010-17704) and Generalitat de Catalunya (research groups 2009SGR809 and 2009SGR145). MV & AV were supported by the Spanish Government (grant BIO2007-66855) and eTOX (grant IMI/115002).