论文信息 - Fast and Flexible MWE Candidate Generation with the mwetoolkit

Fast and Flexible MWE Candidate Generation with the mwetoolkit

We present an experimental environment for computer-assisted extraction of Multiword Expressions (MWEs) from corpora. Candidate extraction works in two steps: generation and filtering. We focus on recent improvements in the former, for which we increased speed and flexibility. We present examples that show the potential gains for users and applications.

Carlos Ramisch | Aline Villavicencio | Vitor de Araújo

[1] Oliver Christ,et al. A Modular and Flexible Architecture for an Integrated Corpus Query System , 1994, ArXiv.

[2] Carlos Ramisch,et al. Multiword Expressions in the wild? The mwetoolkit comes in handy , 2010, COLING.

[3] Stefan Evert,et al. The Statistics of Word Cooccur-rences: Word Pairs and Collocations , 2004 .

[4] Timothy Baldwin,et al. Multiword Expressions: A Pain in the Neck for NLP , 2002, CICLing.

[5] Pavel Rychlý,et al. Manatee, Bonito and Word Sketches for Czech , 2004 .

[6] Satanjeev Banerjee,et al. The Design, Implementation, and Use of the Ngram Statistics Package , 2003, CICLing.

[7] Carlos Ramisch,et al. Towards the Construction of Language Resources for Greek Multiword Expressions: Extraction and Evaluation , 2010, LREC 2010.

[8] Carlos Ramisch,et al. mwetoolkit: a Framework for Multiword Expression Identification , 2010, LREC.

[9] Carlos Ramisch,et al. Web-based and combined language models: a case study on noun compound identification , 2010, COLING.