Phrase Translation Table Research Articles

Acquiring Chinese paraphrases based on random walk of $N$ steps

The conventional “pivot” approach of acquiring paraphrases from bilingual corpus has certain limitations where only candidate paraphrases within two steps are considered. In this paper, we propose a graph-based model of acquiring paraphrases from a phrase translation table. First, we describe a graph-based model representing Chinese-English phrase translation relations, a random walk algorithm based on $N$ number of steps and a confidence metric for the obtained paraphrases. Furthermore, with the aim of finding more potential for Chinese paraphrases, we augment the model so that it is able to integrate other language pairs, such as English-Japanese phrase translation relations. We performed experiments on NTCIR Chinese-English and English-Japanese bilingual corpus and compared the results to those of conventional methods. The experimental results show that the proposed approach acquires more paraphrases. In addition, the performance was improved further after the English-Japanese phrase translations were added to the graph-based model.

An Investigation of the Sampling-Based Alignment Method and its Contributions

By investigating the distribution of phrase pairs in phrase translation tables, the work in this paper describes an approach to increase the number of n-gram alignments in phrase translation tables output by a sampling-based alignment method. This approach consists in enforcing the alignment of n-grams in distinct translation subtables so as to increase the number of n-grams. Standard normal distribution is used to allot alignment time among translation subtables, which results in adjustment of the distribution of n- grams. This leads to better evaluation results on statistical machine translation tasks than the original sampling-based alignment approach. Furthermore, the translation quality obtained by merging phrase translation tables computed from the sampling-based alignment method and from MGIZA++ is examined.

A SomAgent statistical machine translation

Abstract The paper describes the process by which the word alignment task performed within SOMAgent works in collaboration with the statistical machine translation system in order to learn a phrase translation table. We studied improvements in the quality of translation using syntax augmented machine translation. We also experimented with different degrees of linguistic analysis from the lexical level to a syntactic or semantic level, in order to generate a more precise alignment. We developed a contextual environment using the Self-Organizing Map, which can model a semantic agent (SOMAgent) that learns the correct meaning of a word used in context in order to deal with specific phenomena such as ambiguity, and to generate more precise alignments that can improve the first choice of the statistical machine translation system giving linguistic knowledge.

Syntactically Lexicalized Phrase-Based SMT

Until quite recently, extending phrase-based statistical machine translation (PBSMT) with syntactic knowledge caused system performance to deteriorate. The most recent successful enrichments of PBSMT with hierarchical structure either employ nonlinguistically motivated syntax for capturing hierarchical reordering phenomena, or extend the phrase translation table with redundantly ambiguous syntactic structures over phrase pairs. In this paper, we present an extended, harmonized account of our previous work which showed that incorporating linguistically motivated lexical syntactic descriptions, called <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">supertags</i> , can yield significantly better PBSMT systems at insignificant extra computational cost. We describe a novel PBSMT model that integrates supertags into the target language model and the target side of the translation model. Two kinds of supertags are employed: those from lexicalized tree-adjoining grammar and combinatory categorial grammar. Despite the differences between the two sets of supertags, they give similar improvements. In addition to integrating the Markov supertagging approach in PBSMT, we explore the utility of a new surface grammaticality measure based on combinatory operators. We perform various experiments on the Arabic-to-English NIST 2005 test set addressing the issues of sparseness, scalability, and the utility of system subcomponents. We show that even when the parallel training data grows very large, the supertagged system retains a relatively stable absolute performance advantage over the unadorned PBSMT system. Arguably, this hints at a performance gap that cannot be bridged by acquiring more phrase pairs. Our best result shows a relative improvement of 6.1% over a state-of-the-art PBSMT model, which compares favorably with the leading systems on the NIST 2005 task. We also demonstrate that the advantages of a supertag-based system carry over to German-English, where improvements of up to 8.9% relative to the baseline system are observed.

HMM Word and Phrase Alignment for Statistical Machine Translation

Estimation and alignment procedures for word and phrase alignment hidden Markov models (HMMs) are developed for the alignment of parallel text. The development of these models is motivated by an analysis of the desirable features of IBM Model 4, one of the original and most effective models for word alignment. These models are formulated to capture the desirable aspects of Model 4 in an HMM alignment formalism. Alignment behavior is analyzed and compared to human-generated reference alignments, and the ability of these models to capture different types of alignment phenomena is evaluated. In analyzing alignment performance, Chinese-English word alignments are shown to be comparable to those of IBM Model 4 even when models are trained over large parallel texts. In translation performance, phrase-based statistical machine translation systems based on these HMM alignments can equal and exceed systems based on Model 4 alignments, and this is shown in Arabic-English and Chinese-English translation. These alignment models can also be used to generate posterior statistics over collections of parallel text, and this is used to refine and extend phrase translation tables with a resulting improvement in translation quality.