This book investigates how machine learning techniques to improve statistical machine translation, currently at the forefront of research in the field. This book looks first at the technology allows? technologies that solve problems that no machine translation is correct but is closely related to the development of machine translation systems. The book then presents a new or improved statistical machine translation techniques, including a framework for discriminative training to take advantage of syntactic information, the use of semi-supervised learning method and kernel-based, and a combined output of some machine translation in order to improve overall translation quality. Contributors: Srinivas Bangalore, Nicola Cancedda, Josep M. And Melamed, Ion Muslea, Hermann Ney, Bruno Pouliquen, Dan Roth, Anoop Sarkar, John Shawe-Taylor, Ralf Steinberger, Joseph Turian, Nicola Ueffing, Masao Utiyama, Wang Zhuoran, Benjamin Wellington, Kenji Yamada Neural Information Processing series