КОРПУСНАЯ ЛИНГВИСТИКА-2008

Blanco X.

Using NooJ for Multipurpose Analysis of Romance Languages Corpora

NooJ, developed by Max Silberztein (Franche-Comte University), is a powerful linguistic engineering environment allowing the user to formalize a large amount of linguistic phenomena (orthography, morphology, transformational syntax, lexical productivity, etc.). It includes different computational devices (from finite-state automata to augmented recursive transition networks) and provides tools to construct, test, apply and maintain broad coverage electronic dictionaries and grammars. The large NooJ scientific community is continuously developing linguistic resources for NooJ in more than a dozen languages (http://nooj4nlp.net).

We will discuss the main NooJ features and functionalities for the analysis of romance language corpora. We will focus on the Spanish and Catalan modules, that we developed in fLexSem (Laboratory of Phonetics, Lexicology and Semantics) (http://www.flexsem.net), but we will present the French, Italian and Portuguese modules as well.

Let us emphasize that NooJ is widely used in the Autonomous University of Barcelona for computational linguistics teaching in the Master in Information Processing and Multilingual Communication and in the doctoral program Romance Languages and Cultures.

Bibliography
Blanco, X. (2007): "El Master TICOM. Un paradigma integrado de ensenanza de lenguas y tecnologias de la informacion", V Congreso Internacional de Investigaciones Linguistico-Filologicas, Academia Peruana de la Lengua, Lima.
Blanco, X; Silberztein, M. (forthcoming): Proceedings of the 2007 International NooJ Conference, Cambridge Scholars Publishing. Koeva, S.; Maurel, D.; Silberztein, M. (2007): Formaliser les langues avec l'ordinateur : de INTEX a NooJ, Presses Universitaires de Franche-Comte. Silberztein, M. (2006): NooJ Manual, (pdf file, http://nooj4nlp.net).

Назад