BibSonomy publications for /user/diego_ma/machine_translationFri Dec 14 02:48:00 CET 2007Computational Linguistics3421-457\emph{wEBMT}: Developing and Validating an Example-Based Machine Translation System Using the World Wide Web292003machine_translation web We have developed an example-based machine translation (EBMT) system that uses the World Wide Web for two different purposes: First, we populate the system's memory with translations gathered from rule-based MT systems located on the Web. The source strings input to these systems were extracted automatically from an extremely small subset of the rule types in the Penn-II Treebank. In subsequent stages, the <source,target> translation pairs obtained are automatically transformed into a series of resources that render the translation more successful...Fri Dec 14 02:45:40 CET 2007Computational Linguistics3349-380The Web as a Parallel Corpus292003web machine_translation Fri Dec 14 02:45:36 CET 2007Machine Translation and the Information SoupParallel Strands: A Preliminary Investigation into Mining the Web for Bilingual Text1998machine_translation resources Fri Dec 14 02:42:19 CET 2007Lori Levi's homepage?Using Context in Machine Translation of Spoken LanguageXXXXambiguity context machine_translation Fri Dec 14 02:42:08 CET 2007Alon Lavie's homepage?Translation of Conversational Speech with {JANUS-II}YYYYmachine_translation Fri Dec 14 02:42:07 CET 2007Alon Lavie's homepage?Dialogue Processing in a Conversational Speech Translation SystemZZZZmachine_translation Fri Dec 14 02:41:27 CET 2007MayPresentation at IFIChart Translation2000machine_translation The introduction of chart parsing in the early 1960's made it practicable for the first time (1) to separate the grammar from the algorithm that would apply it, (2) to find all the structures assigned to a long sentence by a context-free grammar and, hence, (3) to separate the process of finding analyses from that of choosing those most likely to be correct in context. The idea can be generalized to larger language-processing problems and, in particular, to translation. The result is very liberating because it makes it possible to avoid in large measure the premature elimination of paths from the search space.Fri Dec 14 02:37:38 CET 2007Leuven, BelgiumProc. TMI95Translation Using Minimal Recursion Semantics1995MRS machine_translation We describe minimal recursion semantics (MRS), a framework for semantics within HPSG, which considerably simplifies transfer and generation. We discuss why, in general, a semantic representation with minimal structure is desirable for transfer and illustrate how a descriptively adequate representation with a non-recursive structure may be achieved. The paper illustrates the application of MRS to transfer with a series of examples and compares the approach to others which have been previously adopted within unification based frameworks. Our account involves the use of both language-specific and interlingual predicates or relations and we illustrate how this may be exploited to allow MRS to...Fri Dec 14 02:35:56 CET 2007Computational Linguistics139-71A Maximum Entropy Approach to Natural Language Processing221996maxent machine_translation The concept of maximum entropy can be traced along multiple threads to Biblical times. Only recently, however, have computers become powerful enough to permit the widescale application of this concept to real world problems in statistical estimation and pattern recognition. In this paper, we describe a method for statistical modeling based on maximum entropy. We present a maximum-likelihood approach for automatically constructing maximum entropy models and describe how to implement this approach efficiently, using as examples several problems in natural language processing.Fri Dec 14 02:35:21 CET 2007Santa Cruz, CAProc. ACL-94Available at arXiv:cmp-lg/9607005167-176Head Automata and Bilingual Tiling: Translation with Minimal Representations1996machine_translation DG automata We present a language model consisting of a collection of costed bidirectional finite state automata associated with the head words of phrases. The model is suitable for incremental application of lexical associations in a dynamic programming search for optimal dependency tree derivations. We also present a model and algorithm for machine translation involving optimal ``tiling'' of a dependency tree with entries of a costed bilingual lexicon. Experimental results are reported comparing methods for assigning cost functions to these models. We conclude with a discussion of the adequacy of annotated linguistic strings as representations for machine translation.