A multistage PoS-tagger at the EVALITA 2009 PoS-tagging Task

This paper presents an experimental system architecture for Part-Of-Speech Tagging for the Italian language, able to manage a large tagset to provide both lexical and morphological information. The tagger was built as a cascade of four classifiers where each classifier in the cascade accepts data from an initial input or the guesses of the previous one, executes its annotation, and sends the resulting data to the next stage, or to the output of the cascade. At the EVALITA 2009 PoS-tagging task the combined classifier attained an accuracy of 96.06% on the open task and of 93.54% on the closed one.