Martin Popel

Findings of the 2016 Conference on Machine Translation.

This paper presents the results of the WMT16 shared tasks, which included five machine translation (MT) tasks (standard news, IT-domain, biomedical, multimodal, pronoun), three evaluation tasks (metrics, tuning, run-time estimation of MT quality), and an automatic post-editing task and bilingual document alignment task. This year, 102 MT systems from 24 institutions (plus 36 anonymized online systems) were submitted to the 12 translation directions in the news translation task. The IT-domain task received 31 submissions from 12 institutions in 7 directions and the Biomedical task received 15 submissions systems from 5 institutions. Evaluation was both automatic and manual (relative ranking and 100-point scale assessments). The quality estimation task had three subtasks, with a total of 14 teams, submitting 39 entries. The automatic post-editing task had a total of 6 teams, submitting 11 entries.

TectoMT: modular NLP framework

In the present paper we describe TectoMT, a multi-purpose open-source NLP framework. It allows for fast and efficient development of NLP applications by exploiting a wide range of software modules already integrated in TectoMT, such as tools for sentence segmentation, tokenization, morphological analysis, POS tagging, shallow and deep syntax parsing, named entity recognition, anaphora resolution, tree-to-tree translation, natural language generation, word-level alignment of parallel corpora, and other tasks. One of the most complex applications of TectoMT is the English-Czech machine translation system with transfer on deep syntactic (tectogrammatical) layer. Several modules are available also for other languages (German, Russian, Arabic).Where possible, modules are implemented in a language-independent way, so they can be reused in many applications.

HamleDT: Harmonized multi-language dependency treebank

AbstractWe present HamleDT—a HArmonized Multi-LanguagE Dependency Treebank. HamleDT is a compilation of existing dependency treebanks (or dependency conversions of other treebanks), transformed so that they all conform to the same annotation style. In the present article, we provide a thorough investigation and discussion of a number of phenomena that are comparable across languages, though their annotation in treebanks often differs. We claim that transformation procedures can be designed to automatically identify most such phenomena and convert them to a unified annotation style. This unification is beneficial both to comparative corpus linguistics and to machine learning of syntactic parsing.

Adaptation of machine translation for multilingual information retrieval in the medical domain

OBJECTIVE We investigate machine translation (MT) of user search queries in the context of cross-lingual information retrieval (IR) in the medical domain. The main focus is on techniques to adapt MT to increase translation quality; however, we also explore MT adaptation to improve effectiveness of cross-lingual IR. METHODS AND DATA Our MT system is Moses, a state-of-the-art phrase-based statistical machine translation system. The IR system is based on the BM25 retrieval model implemented in the Lucene search engine. The MT techniques employed in this work include in-domain training and tuning, intelligent training data selection, optimization of phrase table configuration, compound splitting, and exploiting synonyms as translation variants. The IR methods include morphological normalization and using multiple translation variants for query expansion. The experiments are performed and thoroughly evaluated on three language pairs: Czech-English, German-English, and French-English. MT quality is evaluated on data sets created within the Khresmoi project and IR effectiveness is tested on the CLEF eHealth 2013 data sets. RESULTS The search query translation results achieved in our experiments are outstanding - our systems outperform not only our strong baselines, but also Google Translate and Microsoft Bing Translator in direct comparison carried out on all the language pairs. The baseline BLEU scores increased from 26.59 to 41.45 for Czech-English, from 23.03 to 40.82 for German-English, and from 32.67 to 40.82 for French-English. This is a 55% improvement on average. In terms of the IR performance on this particular test collection, a significant improvement over the baseline is achieved only for French-English. For Czech-English and German-English, the increased MT quality does not lead to better IR results. CONCLUSIONS Most of the MT techniques employed in our experiments improve MT of medical search queries. Especially the intelligent training data selection proves to be very successful for domain adaptation of MT. Certain improvements are also obtained from German compound splitting on the source language side. Translation quality, however, does not appear to correlate with the IR performance - better translation does not necessarily yield better retrieval. We discuss in detail the contribution of the individual techniques and state-of-the-art features and provide future research directions.

Hidden Markov Tree Model in Dependency-based Machine Translation

We would like to draw attention to Hidden Markov Tree Models (HMTM), which are to our knowledge still unexploited in the field of Computational Linguistics, in spite of highly successful Hidden Markov (Chain) Models. In dependency trees, the independence assumptions made by HMTM correspond to the intuition of linguistic dependency. Therefore we suggest to use HMTM and tree-modified Viterbi algorithm for tasks interpretable as labeling nodes of dependency trees. In particular, we show that the transfer phase in a Machine Translation system based on tectogrammatical dependency trees can be seen as a task suitable for HMTM. When using the HMTM approach for the English-Czech translation, we reach a moderate improvement over the baseline.

English-Czech MT in 2008

We describe two systems for English-to-Czech machine translation that took part in the WMT09 translation task. One of the systems is a tuned phrase-based system and the other one is based on a linguistically motivated analysis-transfer-synthesis approach.

Perplexity of n-gram and dependency language models

Language models (LMs) are essential components of many applications such as speech recognition or machine translation. LMs factorize the probability of a string of words into a product of P(wi|hi), where hi is the context (history) of word wi. Most LMs use previous words as the context. The paper presents two alternative approaches: post-ngram LMs (which use following words as context) and dependency LMs (which exploit dependency structure of a sentence and can use e.g. the governing word as context). Dependency LMs could be useful whenever a topology of a dependency tree is available, but its lexical labels are unknown, e.g. in tree-to-tree machine translation. In comparison with baseline interpolated trigram LM both of the approaches achieve significantly lower perplexity for all seven tested languages (Arabic, Catalan, Czech, English, Hungarian, Italian, Turkish).

CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies

The Conference on Computational Natural Language Learning (CoNLL) features a shared task, in which participants train and test their learning systems on the same data sets. In 2017, the task was devoted to learning dependency parsers for a large number of languages, in a real-world setting without any gold-standard annotation on input. All test sets followed a unified annotation scheme, namely that of Universal Dependencies. In this paper, we define the task and evaluation methodology, describe how the data sets were prepared, report and analyze the main results, and provide a brief categorization of the different approaches of the participating systems.

New Language Pairs in TectoMT

The TectoMT tree-to-tree machine translation system has been updated this year to support easier retraining for more translation directions. We use multilingual standards for morphology and syntax annotation and language-independent base rules. We include a simple, non-parametric way of combining TectoMT’s transfer model outputs. We submitted translations by the Englishto-Czech and Czech-to-English TectoMT pipelines to the WMT shared task. While the former offers a stable performance, the latter is completely new and will require more tuning and debugging.

MT-ComparEval: Graphical evaluation interface for Machine Translation development

Abstract The tool described in this article has been designed to help MT developers by implementing a web-based graphical user interface that allows to systematically compare and evaluate various MT engines/experiments using comparative analysis via automatic measures and statistics. The evaluation panel provides graphs, tests for statistical significance and n-gram statistics. We also present a demo server http://wmt.ufal.cz with WMT14 and WMT15 translations.