handling translation divergences in generation-heavy hybrid.ppt

Generation-Heavy Hybrid Machine Translation Nizar Habash Postdoctoral ResearcherCenter for Computational Learning SystemsColumbia University Columbia University NLP Colloquium The IntuitionGeneration-Heavy Machine Translation IntroductionResearch Contributions A general reusable and extensible Machine Translation (MT) model that transcends the need for large amounts of deep symmetric knowledge Development of reusable large-scale resources for English A large-scale Spanish-English MT system: Matador; Matador is more robust across genre and produce more grammatical output than simple statistical or symbolic techniques Roadmap Introduction Generation-Heavy Machine Translation Evaluation Conclusion Future Work IntroductionMT Pyramid IntroductionMT Pyramid IntroductionMT Pyramid IntroductionWhy gisting is not enough IntroductionTranslation Divergences 35% of sentences in TREC El Norte Corpus (Dorr et al 2002) Divergence Types Categorial (X tener hambre ? X be hungry) Conflational (X dar pu?aladas a Z ? X stab Z) Structural (X entrar en Y ? X enter Y) Head Swapping (X cruzar Y nadando ? X swim across Y) Thematic (X gustar a Y ? Y like X) Roadmap Introduction Generation-Heavy Machine Translation Evaluation Conclusion Future Work Generation-Heavy Hybrid Machine Translation Problem: asymmetric resources High quality, broad coverage, semantic resources for target language Low quality resources for source language Low quality (many-to-many) translation lexicon Thesis: we can approximate interlingual MT without the use of symmetric interlingual resources Relevant Background Work Hybrid Natural Language Generation Constrained Overgeneration ? Statistical Ranking Nitrogen (Langkilde and Knight 1998), Halogen (Langkilde 2002) FERGUS (Rambow and Bangalore 2000) Lexical Conceptual Structure (LCS) based MT (Jackendoff 1983), (Dorr 1993) LCS-based MTExample Generation-Heavy HybridMachine Translation MatadorSpanish-English GHMT GHMTAnalysis Source lan