MSTParser is a non-projective dependency parser (see McDonald et al., 2005a, 2006) that searches for maximum spanning trees over directed graphs. Models of dependency structure are based on large-margin discriminative training methods (see McDonald et al., 2005b). Projective parsing is also suppo...
This tool assigns a part-of-speech tag and base form to each token in a text. It operates on text that has previously been tokenised and morphologically analysed. The POS tagger is a module of Apertium machine translation system. The provided tool can currently operate on a subset of the language...
LX-UDParser is a UD parser for Portuguese, which adopts the Universal Dependency framework, with an initial performance of 90.87 for UAS and 88.01 for LAS under a ten-fold cross validation scheme. It is described in this article: António Branco, João Ricardo Silva, Luís Gomes and João Rodri...
The present tool, that was built to deal with Portuguese-specific issues concerning a few non-trivial cases that involve tokenization-ambigous strings, segments text into lexically relevant tokens, using whitespace as the separator. Note that, in these examples, the | (vertical bar) symbol is use...
Uplug (see Tiedemann, 2003a) is a collection of tools and scripts for processing text-corpora, for automatic alignment and for term extraction from parallel corpora. Several tools have been integrated in Uplug. Pre-processing tools include a sentence splitter, a general tokenizer and wrappers a...
A NER-classifier based on memory-based learning, trained on the CINTIL dataset, a corpus that contains part of the Corpus de Referência do Português Contemporâneo - CRPC (Reference Corpus of Contemporary Portuguese). https://portulanclarin.net/repository/browse/cintil-corpus-internacional-do-por...
LX-UTagger is a POS tagger for Portuguese that adopts the Universal Part-of-Speech tagset (UPOS), related to the Universal Dependency framework, with an initial performance of 99.06% under a ten-fold cross validation scheme. It is described in this article: António Branco, João Ricardo Silv...
Treat is a toolkit for natural language processing and computational linguistics in Ruby. The Treat project aims to build a language- and algorithm- agnostic NLP framework for Ruby with support for tasks such as document retrieval, text chunking, segmentation and tokenization, natural language pa...
The present tool, that was built to deal with specific issues concerning orthographic conventions adopted for Portuguese, marks sentence boundaries with <s>…</s>, and paragraph boundaries with <p>…</p>. Unwraps sentences split over different lines. A f-score of 99.94% was obtained when testing o...
This is a workflow that is designed especially for use in the UIMA-based U-Compare workbench (see separate META-SHARE record). The workflow is in "ucz" format (specific to U-Compare) and can be imported via the "Import Workflow" item in the "Workflows" menu of the U-Compare interface. It include...