Learn More
Evaluation results recently reported by Callison-Burch et al. (2006) and Koehn and Monz (2006), revealed that, in certain cases, the BLEU metric may not be a reliable MT quality indicator. This happens, for instance , when the systems under evaluation are based on different paradigms, and therefore , do not share the same lexicon. The reason is that, while(More)
In this work we revise the application of discriminative learning to the problem of phrase selection in Statistical Machine Translation. Inspired by common techniques used in Word Sense Disambiguation, we train classifiers based on local context to predict possible phrase translations. Our work extends that of Vickrey et al. (2005) in two main aspects.(More)
This paper describes a set of experiments carried out to explore the domain dependence of alternative supervised Word Sense Disam-biguation algorithms. The aim of the work is threefold: studying the performance of these algorithms when tested on a different corpus from that they were trained on; exploring their ability to tune to new domains, and(More)
sota la direcció del doctor Horacio Rodríguez Hontoria Barcelona, Maig 1998 ii iii iv v Acknowledgments This thesis wouldn't have been possible without the aid and collaboration of many people whom I wish to thank. The hardest and longest task has been carried out by Horacio Rodríguez. His devotion and patience has gone far beyond what one expects from an(More)
This work studies Named Entity Classification (NEC) for Catalan without making use of large annotated resources of this language. Two views are explored and compared, namely exploiting solely the Catalan resources, and a direct training of bilingual classification models (Span-ish and Catalan), given that a large collection of annotated examples is(More)
In this paper we show h o w m a c hine learning techniques for constructing and combining several classiiers can be applied to improve t h e accuracy of an existing English POS tagger (MM arquez and Rodr guez, 1997). Additionally, the problem of data sparseness is also addressed by applying a technique of generating convex pseudo{data (Breiman, 1998).(More)