Search

This paper summarizes findings of a three-year study on verb synonymy in translation based on both syntactic and semantic criteria and reports on recent results extending this work. Primary language resources used are existing Czech and English lexical and corpus resources, namely the Prague Dependency Treebank-style valency lexicons, FrameNet, VerbNet, PropBank, WordNet and the parallel Prague Czech-English Dependency Treebank, which contains deep syntactic and partially semantic annotation of running texts. The resulting lexicon (called formerly CzEngClass, now SynSemClass) and all associated resources linked to the existing lexicons and corpora following from this project are publicly and freely available. While the project proper assumes manual annotation work, we expect to use the resulting resource (together with the existing ones) as a necessary resource for developing automatic methods for extending such a lexicon, or creating similar lexicons for other languages.

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.

This paper presents the participation of the Charles University team in the WMT 2014 Medical Translation Task. Our systems are developed within the Khresmoi project, a large integrated project aiming to deliver a multi-lingual multi-modal search and access system for biomedical information and documents. Being involved in the organization of the Medical Translation Task, our primary goal is to set up a baseline for both its subtasks (summary translation and query translation) and for all translation directions. Our systems are based on the phrasebased Moses system and standard methods for domain adaptation. The constrained/unconstrained systems differ in the training data only.

Phraseological components of valency dictionaries for two West Slavic languages are presented, namely, of the PDT-Vallex dictionary for Czech and of the Walenty dictionary for Polish. Both dictionaries are corpus-based, albeit in different ways. Both are machinereadable and employable by syntactic parsers and generators. The paper compares the expressive power of the phraseological subformalisms of these dictionaries, discusses their limitations and makes recommendations for their possible extensions, which can be possibly applied also to other valency dictionaries with rich phraseological components.

Under the term grammaticalized alternations, we understand changes in valency frames of verbs corresponding to different surface syntactic structures of the same lexical unit of a verb. Czech grammaticalized alternations are expressed either (i) by morphological means (diatheses), or (ii) by syntactic means (reciprocity). These changes are limited to changes in morphemic form(s) of valency complementations; moreover, they are regular enough to be captured by formal syntactic rules.

Corpus annotation plays an important role in linguistic analysis and computational processing of both written and spoken language. Syntactic annotation of spoken texts becomes clearly a topic of considerable interest nowadays, driven by the desire to improve automatic speech recognition systems by incorporating syntax in the language models, or to build language understanding applications. Syntactic annotation of both written and spoken texts in the Czech Academic Corpus was created thirty years ago when no other (even annotated) corpus of spoken texts has existed. We will discuss how much relevant and inspiring this annotation is to the current frameworks of spoken text annotation.

We present a description of a new resource (Prague Dependency Treebank of Spoken Language) being created for English and Czech to be used for the task of speech understanding, broad natural language analysis for dialog systems and other speech-related tasks, including speech editing. The resources we have created so far contain audio and a standard transcription of spontaneous speech, but as a novel layer, we add an edited (ldquoreconstructedrdquo) version of the spoken utterances. These edits go beyond the scope of current speech reconstruction efforts in that we allow, on top of the usual deletions of speech artifacts, fillers, etc. also for word modifications, insertions and word order changes. We have used both monologue and dialogue recordings in English and Czech to verify the feasibility of such transcription. We have also assessed the quality of the resulting annotation since the relative freedom of the editing raises an issue of what a ldquocorrectrdquo annotation is.