Domain adaptation of MT systems through automatic post-editing

Journal of Artificial Intelligence Research, 2016

We describe a hybridisation strategy whose objective is to integrate linguistic resources from shallow-transfer rule-based machine translation (RBMT) into phrase-based statistical machine translation (PBSMT). It basically consists of enriching the phrase table of a PBSMT system with bilingual phrase pairs matching transfer rules and dictionary entries from a shallow-transfer RBMT system. This new strategy takes advantage of how the linguistic resources are used by the RBMT system to segment the source-language sentences to be translated, and overcomes the limitations of existing hybrid approaches that treat the RBMT systems as a black box. Experimental results confirm that our approach delivers translations of higher quality than existing ones, and that it is specially useful when the parallel corpus available for training the SMT system is small or when translating out-of-domain texts that are well covered by the RBMT dictionaries. A combination of this approach with a recently pro...

2014

System architecture, experimental settings and evaluation results of the EIWA in the WAT2014 Japanese to English (jaen) and Chinese to Japanese (zh-ja) tasks are described. Our system is combining rule-based machine translation (RBMT) and statistical post-editing (SPE). Evaluation results for ja-en task show 19.86 BLEU score, 0.7067 RIBES score, and 22.50 human evaluation score. Evaluation results for zh-ja task show 33.57 BLEU score, 0.8114 RIBES score, and 15.00 human evaluation score.

Machine Translation, 2014

The effective integration of MT technology into computer-assisted translation tools is a challenging topic both for academic research and the translation industry. Particularly, professional translators feel crucial the ability of MT systems to adapt to their feedback. In this paper, we propose an adaptation scheme to tune a statistical MT system to a translation project using small amounts of post-edited texts, like those generated by a single user in even just one day of work. The same scheme can be applied at the larger scale in order to focus general purpose models towards the specific domain of interest. We assess our method on two domains, namely information technology and legal, and four translation directions, from English to French, Italian, Spanish and German. The main outcome is that our adaptation strategy can be very effective provided that the seed data used for adaptation is enough related to the remaining text to translate; otherwise, MT quality does neither improve nor worsen, thus showing the robustness of our method.

Most current Machine Translation (MT) systems do not improve with feedback from post-editors beyond the addition of corrected translations to parallel training data (for statistical and example-base MT) or to a memory database. Rule based systems to date improve only via manual debugging. In contrast, we propose a largely automated method for capturing more information from human post-editors, so that corrections may be performed automatically to translation grammar rules and lexical entries. This paper introduces a general framework for incorporating a refinement module into rule-based transfer MT systems. This framework allows for generalizing post-editing efforts in an effective way, by identifying and correcting rules semi-automatically on order to improve coverage and overall translation quality.

Proc. of the 10th Annual Conf. of the …, 2005

Most current Machine Translation (MT) systems do not improve with feedback from post-editors beyond the addition of corrected translations to parallel training data (for statistical and example-base MT) or to a memory database. Rule based systems to date improve only via manual debugging. In contrast, we introduce a largely automated method for capturing more information from human post-editors, so that corrections may be performed automatically to translation grammar rules and lexical entries. This paper introduces a general framework for incorporating a refinement module to rule-based transfer MT systems. This framework allows to generalize post-editing efforts in an effective way, by identifying and correcting rules semi-automatically to improve coverage and overall translation quality.

translution.com

1 Executive Summary We evaluated the French-to-English versions of two rules based machine translation (MT) systems (referred to as s01 and s02 in this document) in order to assess the quality of their output and to determine whether updating the system dictionaries brought about an improvement in performance.

Computational Linguistics, 2003

We have developed an example-based machine translation (EBMT) system that uses the World Wide Web for two different purposes: First, we populate the system's memory with translations gathered from rule-based MT systems located on the Web. The source strings input to these systems were extracted automatically from an extremely small subset of the rule types in the Penn-II Treebank. In subsequent stages, the source, target translation pairs obtained are automatically transformed into a series of resources that render the translation process more successful. Despite the fact that the output from on-line MT systems is often faulty, we demonstrate in a number of experiments that when used to seed the memories of an EBMT system, they can in fact prove useful in generating translations of high quality in a robust fashion. In addition, we demonstrate the relative gain of EBMT in comparison to on-line systems. Second, despite the perception that the documents available on the Web are of questionable quality, we demonstrate in contrast that such resources are extremely useful in automatically postediting translation candidates proposed by our system.

2010

In this paper we describe the statistical machine translation system of the RWTH Aachen University developed for the translation task of the Fifth Workshop on Statistical Machine Translation. Stateof-the-art phrase-based and hierarchical statistical MT systems are augmented with appropriate morpho-syntactic enhancements, as well as alternative phrase training methods and extended lexicon models. For some tasks, a system combination of the best systems was used to generate a final hypothesis. We participated in the constrained condition of German-English and French-English in each translation direction.

Current state-of-the-art Statistical Machine Translation systems are based on log-linear models that combine a set of feature functions to score translation hypotheses during decoding. The models are parametrized by a vector of weights usually optimized on a set of sentences and their reference translations, called development data. In this paper, we explore a (common and industry relevant) scenario where a system trained and tuned on general domain data needs to be adapted to a specific domain for which no or only very limited in-domain bilingual data is available. It turns out that systems can be adapted successfully by re-tuning model parameters using surprisingly small amounts of parallel in-domain data, by cross-tuning or no tuning at all. We show in detail how and why this is effective, compare the approaches and effort involved. We also study the effect of hyperparameters (such as maximum phrase length and development data size) and their optimal values in this scenario.

2007

The special challenge of the WMT 2007 shared task was domain adaptation. We took this opportunity to experiment with various ways of adapting a statistical machine translation systems to a special domain (here: news commentary), when most of the training data is from a different domain (here: European Parliament speeches). This paper also gives a description of the submission of the University of Edinburgh to the shared task.