RACAI's Natural Language Processing pipeline for Universal Dependencies

Abstract

This paper presents RACAI's approach, experiments and results at CONLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies. We handle raw text and we cover tokenization, sentence splitting, word segmentation, tagging, lemmatization and parsing. All results are reported under strict training, development and testing conditions, in which the corpora provided for the shared tasks is used ``as is'', without any modifications to the composition of the train and development sets.

Tasks

Reproductions