Statistical machine translation

By Philipp Koehn. 2010

101 citations recorded by the source corpus.

This introductory text to statistical machine translation (SMT) provides all of the theories and methods needed to build a statistical machine translator, such as Google Language Tools and Babelfish. In general, statistical techniques allow automatic translation systems to be built quickly for any language-pair using only translated texts and generic software. With increasing globalization, statistical machine translation will be central to communication and commerce. Based on courses and tutorials, and classroom-tested globally, it is ideal for instruction or self-study, for advanced undergraduates and graduate students in computer science and/or computational linguistics, and researchers in natural language processing. The companion website provides open-source corpora and tool-kits.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Assessing gender bias in machine translation: a case study with Google Translate (2018, neutral): “sing Advisory Committee) report( [19]. Such research was then disfavoured in the United States, making a re-entrance in the 1970s before the 1980s surge in statistical methods for machine translation [25, 26]. Statistical and example-based machine translation have been on the rise ever since [2, 8, ”
Gmail Smart Compose: Real-Time Assisted Writing (2019, neutral): “Language modelling is a fundamental and indispensable component of many natural language processing (NLP) and automatic speech recognition (ASR) systems [23, 27, 32, 35].”
Learning to Mine Aligned Code and Natural Language Pairs from Stack Overflow (2018, positive): “ty of the intent given the snippet P(I |S), and the probability of the snippet given the intent P(S |I). Specifically, we follow previous work that has noted that models from machine translation (MT; [19]) are useful for learning the correspondences between natural language and code for the purposes”
An Empirical Study on Learning Bug-Fixing Patches in the Wild via Neural Machine Translation (2018, neutral): “nerally use data-driven methods to translate text or speech from one language to another. Machine translation systems are trained on translated texts, or “parallel corpora”, for particular text types [38] in both natural languages and formal languages such as programming languages. Manually migratin”
Improving Pronoun Translation for Statistical Machine Translation (2012, unknown): “This justiﬁes recent interest (Le Nagard & Koehn, 2010; Hard-meier & Federico, 2010) and motivates the work presented in this paper.”
Apertium: a free/open-source platform for rule-based machine translation (2011, unknown): “Development and research on corpus-based MT, mainly statistical machine translation (SMT; Koehn 2010), has drastically increased since the nineties, due to the significant rise in computational power and storage capacity of modern computers, and the growing availability of parallel texts.”
Semantic Matching in Search (2014, unknown): “The machine learning techniques concerned with in this survey include statistical language model [204], statistical machine translation [99], learning to rank [128, 115, 116], graphical model [24], topic model [25], matrix factorization [103], kernel methods [158], sparse methods 3, and deep learnin”
Book Reviews: Data-Intensive Text Processing with MapReduce by Jimmy Lin and Chris Dyer (2010, unknown): “We begin by giving a brief introduction to statistical machine translation and the phrase-based translation approach; for more a comprehensive introduction, refer to [57, 69].”
Computer-Assisted Text Analysis for Comparative Politics (2015, unknown): “Simultaneously, large, mature software companies like IBM, Microsoft, and Google have also developed their own machine translation systems (Koehn, 2009).”
Neural Machine Translation: A Review (2019, unknown): “In contrast, traditional (symbolic) MT systems can easily use very large vocabularies (Heafield et al., 2013; Lin & Dyer, 2010; Chiang, 2007; Koehn, 2010).”