Training state-of-the-art portuguese POS taggers without handcrafted features

Resumo

Part-of-speech (POS) tagging for morphologically rich languages normally requires the use of handcrafted features that encapsulate clues about the language's morphology. In this work, we tackle Portuguese POS tagging using a deep neural network that employs a convolutional layer to learn character-level representation of words. We apply the network to three different corpora: the original Mac-Morpho corpus; a revised version of the Mac-Morpho corpus; and the Tycho Brahe corpus. Using the proposed approach, while avoiding the use of any handcrafted feature, we produce state-of-the-art POS taggers for the three corpora: 97.47% accuracy on the Mac-Morpho corpus; 97.31% accuracy on the revised Mac-Morpho corpus; and 97.17% accuracy on the Tycho Brahe corpus. These results represent an error reduction of 12.2%, 23.6% and 15.8%, respectively, on the best previous known result for each corpus.

Ficha do documento

Tipo

Artigo científico

Ano

2014

Instituição

Springer Int Publishing Ag

Fonte

Repositório da FGV

Idioma

Inglês

Acesso

Acesso restrito

Identificador

oai:repositorio.fgv.br:10438/23485

Temas

Tecnologia

Palavras-chave

Portuguese part-of-speech taggingDeep learningConvolutional neural networksRecognitionTecnologiaLíngua portuguesaRedes sociais