We report on the Evalita 2009 PoS Tagging task, an initiative for the evaluation of automatic PoS Taggers for Italian. The challenge of this year’ task consisted in dealing with a large tag set, including morphological traits, and the fact that the training data were extracted from a newspapers corpus while the test data were from a different domain, the Italian Wikipedia. Considering these difficulties, the performance of the participating systems is quite high, compared with state of the art taggers for other languages, with small differences in accuracy between the systems
Overview of the Evalita 2009 POS Tagging Task
ATTARDI, GIUSEPPE;SIMI, MARIA
2009-01-01
Abstract
We report on the Evalita 2009 PoS Tagging task, an initiative for the evaluation of automatic PoS Taggers for Italian. The challenge of this year’ task consisted in dealing with a large tag set, including morphological traits, and the fact that the training data were extracted from a newspapers corpus while the test data were from a different domain, the Italian Wikipedia. Considering these difficulties, the performance of the participating systems is quite high, compared with state of the art taggers for other languages, with small differences in accuracy between the systemsFile in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
