Probabilistic Synchronous Context-free Grammars (PSCFGs) [Aho and Ullmann, 1969, Wu, 1996] define weighted transduction rules to represent translation and reordering operations. When translation models use features that are defined locally, on each rule, there are efficient dynamic programming algorithms to perform translation with these grammars [Kasami, 1965]. In general, the integration of non-local features into the translation model can make translation NP-hard, requiring decoding approximations that limit the impact of these features. In this thesis, we consider the impact and interaction between two non-local features, the n-gram language model (LM) and labels on rule nonterminal symbols in the Syntax-Augmented MT (SAMT) grammar [Zollmann and Venugopal, 2006]. While these features do not result in NP-hard search, they would lead to serious increases in wall-clock runtime if naïve dynamic programming methods are applied. We develop novel two-pass algorithms that make strong decoding approximations during a first pass search, generating a hypergraph of sentence spanning translation i derivations. In a second pass, we use knowledge about non-local features to explore

We present an interface for visualizing information from a diverse type of linguistic resources to facilitate the correction of machine translated sentences. The goal of this project is to aid users with little or no fluency in the language of the original text to gain a greater comprehension of machine translations. This paper describes our interface and the linguistic resources it exploits to help users to detect and correct translation errors.

This paper documents recent work carried out for PeEn-SMT, our Statistical Machine Translation system for translation between the English-Persian language pair. We give details of our previous SMT system, and present our current development of significantly larger corpora. We explain how recent tests using much larger corpora helped to evaluate problems in parallel corpus alignment, corpus content, and how matching the domains of PeEn-SMT’s components affect translation output. We then focus on combining corpora and approaches to improve test data, showing details of experimental setup, together with a number of experiment results and comparisons between them. We show how one combination of corpora gave us a metric score outperforming Google Translate for the English-to-Persian translation. Finally, we outline areas of our intended future work, and how we plan to improve the performance of our system to achieve higher metric scores, and ultimately to provide accurate, reliable language translation.

This degree is submitted to the University of Cambridge

Statistical Machine Translation (SMT) is an approach to automatic text translation based on the use of statistical models and examples of translations. Although Machine Translation (MT) systems developed according to other paradigms are still in use, mainly rule-based or example-based MT, SMT dominates academic

In this work, we model the writing revision process of English as a Second Language (ESL) students with syntaxdriven machine translation methods. We compare two approaches: tree-tostring transformations (Yamada and Knight, 2001) and tree-to-tree transformations (Smith and Eisner, 2006). Results suggest that while the tree-totree model provides a greater coverage, the tree-to-string approach offers a more plausible model of ESL learners’ revision writing process. 1

An enormous potential exists for solving certain classes of computational problems through rich collaboration among crowds of humans supported by computers. Solutions to these problems used to involve human professionals who are expensive to hire or difficult to find. Despite significant advances, fully automatic systems still have much room for improvement. Recent research has involved recruiting large crowds of skilled humans (“crowdsourcing”), but crowdsourcing solutions are still restricted by the availability of those skilled human participants. With translation, for example, professional translators incur high cost and are not always available; machine translation systems have been greatly improved recently, but still can only provide passable translation, and for only limited language pairs at that; crowdsourced translation is limited by the availability of bilingual humans. This dissertation describes crowdsourced monolingual translation, where monolingual translation is translation performed by monolingual people. Crowdsourced monolingual translation is a collaborative form of translation performed by two crowds of people

proportions, the flags adopted in 1889 are still in schizophrenia, the condition manifested itself before use by the imperial family. A person diagnosed with schizophrenia may The Efficacy of Human Post-Editing for Language Translation experience hallucinations (most reported are hearing voices), delusions (often bizarre or