Bogdan Babych
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Bogdan Babych has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Serge Sharoff
University of Leeds
4 shared publications Anthony Hartley
University of Leeds
3 shared publicationsOlga Mudraya
Lancaster University, United Kingdom
2 shared publications- Paul Rayson
School of Computing and Communications, Lancaster University, Lancaster, United Kingdom
2 shared publications - Scott Piao
Lancaster University
2 shared publications - Anthony Hartley
Centre for Translation Studies University of Leeds Woodhouse Lane LS16 5PT Leeds UK
1 shared publication
Latest Publications
- 1 June 2008
- Published by John Benjamins Publishing Company
- p. 293-309
18. Automatic extraction of translation equivalents of phrasal and light verbs in English and Russian
This chapter aims at bridging the functionalist theoretical perspective on word usage with corpus-based studies. We are dealing with the issue of construction of reliable lists of what is called ‘phraseological units’ in general linguistics literature or ‘multi-word expressions’ (MWEs) in literature on computational linguistics. The two groups of constructions under investigation in this chapter are phrasal verbs and light verb constructions. Another distinguishing feature of this study is its multilingual aspect. Previous computational approaches to MWEs have mainly focussed on English, and there has been little research on computational approaches to MWEs in other languages. In this chapter, we examine phrasal verbs in English and their translation equivalents in Russian, and compare English-Russian/Russian-English translation equivalents of selected light verb constructions in two case studies. Our study reveals some interesting cross-language structural divergences between the languages under consideration and shows that a phraseological expression in a language may have equivalent expressions in other languages with different morpho-syntactic structures and semantic properties. However, our investigation not only reveals marked differences between English and Russian, but also discovers some general corresponding structural patterns between them; for example, English phrasal verbs usually have single-word translation equivalents in Russian. Moreover, our study of phrasal and light verbs demonstrates that corpus-based resources can provide an invaluable help to a practising translator, as dictionaries do not cover a large variety of real-life language examples.
- 1 June 2008
- Published by John Benjamins Publishing Company
- p. 293-309
18. Automatic extraction of translation equivalents of phrasal and light verbs in English and Russian
This chapter aims at bridging the functionalist theoretical perspective on word usage with corpus-based studies. We are dealing with the issue of construction of reliable lists of what is called ‘phraseological units’ in general linguistics literature or ‘multi-word expressions’ (MWEs) in literature on computational linguistics. The two groups of constructions under investigation in this chapter are phrasal verbs and light verb constructions. Another distinguishing feature of this study is its multilingual aspect. Previous computational approaches to MWEs have mainly focussed on English, and there has been little research on computational approaches to MWEs in other languages. In this chapter, we examine phrasal verbs in English and their translation equivalents in Russian, and compare English-Russian/Russian-English translation equivalents of selected light verb constructions in two case studies. Our study reveals some interesting cross-language structural divergences between the languages under consideration and shows that a phraseological expression in a language may have equivalent expressions in other languages with different morpho-syntactic structures and semantic properties. However, our investigation not only reveals marked differences between English and Russian, but also discovers some general corresponding structural patterns between them; for example, English phrasal verbs usually have single-word translation equivalents in Russian. Moreover, our study of phrasal and light verbs demonstrates that corpus-based resources can provide an invaluable help to a practising translator, as dictionaries do not cover a large variety of real-life language examples.
- 12 December 2007
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 43 (1) , 15-25
‘Irrefragable answers’ using comparable corpora to retrieve translation equivalents
- 12 December 2007
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 43 (1) , 15-25
‘Irrefragable answers’ using comparable corpora to retrieve translation equivalents
- 1 January 2004
- Published by Association for Computational Linguistics (ACL)
- p. 621-628
Extending the BLEU MT evaluation method with frequency weightings
We present the results of an experiment on extending the automatic method of Machine Translation evaluation BLUE with statistical weights for lexical items, such as tf.idf scores. We show that this extension gives additional information about evaluated texts; in particular it allows us to measure translation Adequacy, which, for statistical MT systems, is often overestimated by the baseline BLEU method. The proposed model uses a single human reference translation, which increases the usability of the proposed method for practical purposes. The model suggests a linguistic interpretation which relates frequency weights and human intuition about translation Adequacy and Fluency.
- 1 January 2004
- Published by Association for Computational Linguistics (ACL)
- p. 621-628
Extending the BLEU MT evaluation method with frequency weightings
We present the results of an experiment on extending the automatic method of Machine Translation evaluation BLUE with statistical weights for lexical items, such as tf.idf scores. We show that this extension gives additional information about evaluated texts; in particular it allows us to measure translation Adequacy, which, for statistical MT systems, is often overestimated by the baseline BLEU method. The proposed model uses a single human reference translation, which increases the usability of the proposed method for practical purposes. The model suggests a linguistic interpretation which relates frequency weights and human intuition about translation Adequacy and Fluency.
- 1 January 2003
- Published by Association for Computational Linguistics (ACL)
- p. 1-8
Improving machine translation quality with automatic named entity recognition
- 1 January 2003
- Published by Association for Computational Linguistics (ACL)
- p. 1-8
Improving machine translation quality with automatic named entity recognition