 |
|
|
|
| HAL : hal-00510999, version 1 |
| Fiche détaillée |  Récupérer au format |
|
|
| International Conference on Lexis and Grammar, Belgrade : Serbie (2010) |
|
|
|
|
| Fast Development of Basic NLP Tools: Towards a Lexicon and a POS Tagger for Kurmanji Kurdish |
|
|
| Géraldine Walther 1Benoît Sagot 2 |
|
|
| (15/09/2010) |
|
|
|
| The development of basic NLP resources for minority languages is still a challenge to both formal and computational linguists. In this paper, we show how we were able to develop a medium-scale morphological lexicon for Kurmanji Kurdish in a few days time using only freely accessible resources. We also developed a preliminary POS tagger that shall be used as a pre-annotation tool for developing a POS-annotated corpus, based solely on raw text and on our morphological lexicon. |
|
|
|
|
|
|
|
|
|
|
|
| 1 : | Laboratoire de Linguistique Formelle (LLF) |
|
|
|
CNRS : UMR7110 – Université Paris VII - Paris Diderot |
|
|
| 2 : | ALPAGE (INRIA Rocquencourt) |
|
|
|
INRIA – Université Paris VII - Paris Diderot |
|
|
| 3 : | Institut de l'information scientifique et technique (INIST) |
|
|
|
CNRS : UPS76 |
|
|
|
|
|
|
|
|
|
| Domaine | : | Informatique/Traitement du texte et du document Sciences de l'Homme et Société/Linguistique |
|
|
| Liste des fichiers attachés à ce document : | |||||
|
|||||
|
|
|
| hal-00510999, version 1 | |
| http://hal.archives-ouvertes.fr/hal-00510999 | |
| oai:hal.archives-ouvertes.fr:hal-00510999 | |
| Contributeur : Karën Fort | |
| Soumis le : Lundi 23 Août 2010, 11:40:54 | |
| Dernière modification le : Lundi 23 Août 2010, 14:04:40 | |
