The promise of unsupervised learning methods lies in their potential to use vast amounts of unlabeled data to learn complex, highly nonlinear models with millions of free paramete...
In this paper, we propose a novel method for semi-supervised learning of nonprojective log-linear dependency parsers using directly expressed linguistic prior knowledge (e.g. a no...
We develop a new approach to learning phrase translations from parallel corpora, and show that it performs with very high coverage and accuracy in choosing French translations of ...
We describe our contribution to the Generation Challenge 2010 for the tasks of Named Entity Recognition and coreference detection (GREC-NER). To extract the NE and the referring e...
This paper presents a method that assists in maintaining a rule-based named-entity recognition and classification system. The underlying idea is to use a separate system, construc...
Georgios Petasis, Frantz Vichot, Francis Wolinski,...