The field of machine translation (MT) - the automation of translation between human languages - has existed for more than 50 years. MT helped to usher in the field of computational linguistics and has influenced methods and applications in knowledge representation, information theory, and mathematical statistics.

Many applications within natural language processing involve performing text-to-text transformations, i.e., given a text in natural language as input, systems are required to produce a version of this text (e.g., a translation), also in natural language, as output. Automatically evaluating the output of such systems is an important component in developing text-to-text applications. Two approaches have been proposed for this problem: (i) to compare the system outputs against one or more reference outputs using string matching-based evaluation metrics and (ii) to build models based on human feedback to predict the quality of system outputs without reference texts. Despite their popularity, reference-based evaluation metrics are faced with the challenge that multiple good (and bad) quality outputs can be produced by text-to-text approaches for the same input. This variation is very hard to capture, even with multiple reference texts. In addition, reference-based metrics cannot be used in production (e.g., online machine translation systems), when systems are expected to produce outputs for any unseen input. In this book, we focus on the second set of metrics, so-called Quality Estimation (QE) metrics, where the goal is to provide an estimate on how good or reliable the texts produced by an application are without access to gold-standard outputs. QE enables different types of evaluation that can target different types of users and applications. Machine learning techniques are used to build QE models with various types of quality labels and explicit features or learnt representations, which can then predict the quality of unseen system outputs. This book describes the topic of QE for text-to-text applications, covering quality labels, features, algorithms, evaluation, uses, and state-of-the-art approaches. It focuses on machine translation as application, since this represents most of the QE work done to date. It also briefly describes QE for several other applications, including text simplification, text summarization, grammatical error correction, and natural language generation.

Learn how to build machine translation systems with deep learning from the ground up, from basic concepts to cutting-edge research.

This is an integrated and structured set of progressive readings from translation and related disciplines, which provides a comprehensive overview of the field and how it is developing.

This book is about the limits of machine translation. It is widely recognized that machine translation systems do much better on domain-specific controlled-language texts (domain texts for short) than on dynamic general-language texts (general texts for short). The authors explore this general domain distinction and come to some uncommon conclusions about the nature of language. Domain language is claimed to be made possible by general language, while general language is claimed to be made possible by the ethical dimensions of relationships. Domain language is unharmed by the constraints of objectivism, while general language is suffocated by those constraints. Along the way to these conclusions, visits are made to Descartes and Saussure, to Chomsky and Lakoff, to Wittgenstein and Levinas. From these conclusions, consequences are drawn for machine translation and translator tools, for linguistic theory and translation theory. The title of the book does not question whether language is possible; it asks, with wonder and awe, why communication through language is possible.

Lynne Bowker and Jairo Buitrago Ciro introduce the concept of machine translation literacy, a new kind of literacy for scholars and librarians in the digital age. This book is a must-read for researchers and information professionals eager to maximize the global reach and impact of any form of scholarly work.