Bilingual word alignment forms the foundation of most approaches to statistical machine translation. Current word alignment methods are predominantly based on generative models. In this paper, we demonstrate a discriminative approach to training simple word alignment models that are comparable in accuracy to the more complex generative models normally used. These models have the the advantages that they are easy to add features to and they allow fast optimization of model parameters using small amounts of annotated data. 1

Statistical machine translation (SMT) treats the translation of natural language as a machine learning problem. By examining many samples of human-produced translation, SMT algorithms automatically learn how to translate. SMT has made tremendous strides in less than two decades, and many popular techniques have only emerged within the last few years. This survey presents a tutorial overview of state-of-the-art SMT at the beginning of 2007. We begin with the context of the current research, and then move to a formal problem description and an overview of the four main subproblems: translational equivalence modeling, mathematical modeling, parameter estimation, and decoding. Along the way, we present a taxonomy of some different approaches within these areas. We conclude with an overview of evaluation and notes on future directions.

This paper presents the task definition, resources, participating systems, and comparative results for the shared task on word alignment, which was organized as part of the ACL 2005 Workshop on Building and Using Parallel Texts. The shared task included English--Inuktitut, Romanian--English, and English--Hindi sub-tasks, and drew the participation of ten teams from around the world with a total of 50 systems.

Abstract. This paper investigates the combination of word-alignments computed with the competitive linking algorithm and well-established IBM models. New training methods for phrase-based statistical translation are proposed, which have been evaluated on a popular traveling domain task, with English as target language, and Chinese, Japanese, Arabic and Italian as source languages. Experiments were performed with a highly competitive phrase-based translation system, which ranked at the top in the 2005 IWSLT evaluation campaign. By applying the proposed techniques, even under very different data-sparseness conditions, consistent improvements in BLEU and NIST scores were obtained on all considered language pairs. 1

Abstract—Bilingual semantic term association is very useful in cross-language information retrieval, statistical machine translation, and many other applications in natural language processing. In this paper, we present a method, named SBA-term, which applies sparse linear regression (Lasso, Least Squares with l1 penalty) and L 2 rescaling for design matrix to the task of bilingual term association. The approach hinges on formulating the task as a feature selection problem within a classification framework. Our experimental results indicate that our novel proposed method is more efficient than co-occurrence at extracting relevant bilingual terms semantic associations. In addition, our approach connects the vibrant area of sparse machine learning to an important problem of natural language processing. I.

The best systems for machine translation of natural language are based on statistical models learned from data. Conventional representation of a statistical translation model requires substantial offline computation and representation in main memory. Therefore, the principal bottlenecks to the amount of data we can exploit and the complexity of models we can use are available memory and CPU time, and current state of the art already pushes these limits. With data size and model complexity continually increasing, a scalable solution to this problem is central to future improvement. Callison-Burch et al. (2005) and Zhang and Vogel (2005) proposed a solution that we call translation by pattern matching, which we bring to fruition in this dissertation. The training data itself serves as a proxy to the model; rules and parameters are computed on demand. It achieves our desiderata of minimal offline computation and compact representation, but is dependent on fast pattern matching algorithms on text. They demonstrated its application to a common model based on the translation of contiguous substrings, but leave some open problems. Among these is a question: can this approach match the performance of conventional methods despite unavoidable differences that it induces in the model? We show how to answer this question affirmatively. The main

This degree is submitted to the University of Cambridge

c ○ H.M. Cooper 2010”I may not have gone where I intended to go, but I think I have ended up where I needed to be.” The aim of this thesis is to find new approaches to Sign Language Recognition (SLR) which are suited to working with the limited corpora currently available. Data available for SLR is of limited quality; low resolution and frame rates make the task of recognition even more complex. The content is rarely natural, concentrating on isolated signs and filmed under laboratory conditions. In addition, the amount of accurately labelled data is minimal. To this end, several contributions are made: Tracking the hands is eschewed in favour of detection based techniques more robust to noise; for both signs and for linguistically-motivated sign sub-units are investigated, to make best use of limited data sets. Finally, an algorithm is proposed to learn signs from the inset signers on TV, with the aid of the accompanying subtitles, thus increasing the corpus of data available. Tracking fast moving hands under laboratory conditions is a complex task, move this to real world data and the challenge is even greater. When using tracked data as a base for SLR, the errors

Abstract not found