Dear all,
It is my pleasure to announce the release of HeidelTime, a new temporal
tagger.
HeidelTime performed best at the TempEval-2 challenge for both, the
extraction and the normalization of temporal expressions (for English).
Currently, HeidelTime supports two languages, English and German. Due to
the strict separation of the source code and language resources like
patterns and rules, HeidelTime is easily extensible to further
languages. For the normalization of temporal expressions, different
normalization strategies are applied depending on the document type. In
particular, HeidelTime distinguishes between news-style and
narrative-style documents. We evaluated HeidelTime on several corpora
and achieve very good results.
HeidelTime is available as UIMA-component and as standalone version.
Additional information and a download links can be find on:
http://dbs.ifi.uni-heidelberg.de/heideltime
An online demo of HeidelTime is available at:
http://dbs-projects.ifi.uni-heidelberg.de/heideltime/
If you have any questions or run into any problems, please don't
hesitate to contact me.
Best regards,
Jannik Strötgen
--
Jannik Strötgen
Institute of Computer Science
Database Systems Research Group
Im Neuenheimer Feld 348
69120 Heidelberg
Germany
Phone: +49 (0) 6221 / 54-5709
eMail: [email protected]
www: http://dbs.ifi.uni-heidelberg.de/
