Learning Machine Translation
The Internet gives us access to a wealth of information in languages we don't understand, and instant online translation leaves much to be desired. The investigation of automated or semi-automated approaches to translation has become a thriving research field with enormous commercial potential. Statistical approaches dominate the field, and this volume investigates how Machine Learning techniques can improve aspects of Statistical Machine Translation (SMT). The book looks first at enabling technologies--technologies that solve problems that are not Machine Translation proper but are linked closely to the development of a Machine Translation system. These include the acquisition of bilingual sentence aligned data from comparable corpora, automatic construction of multilingual name dictionaries, and word alignment. Chapters then present new or improved statistical Machine Translation techniques, including a discriminative training framework for the problem of including syntactic information in the translation model, the use of semi-supervised learning to improve Machine Translation output, and combining outputs of multiple Machine Translation systems in order to improve overall translation quality.