A systematic comparison of methods for low-resource dependency parsing on genuinely low-resource languages

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 1105–1116, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 1105 A systematic comparison of methods for low-resource dependency parsing on genuinely low-resource languages Clara Vania1 Yova Kementchedjhieva2 Anders Søgaard2 Adam Lopez1 1School of Informatics, University of Edinburgh, UK 2University of Copenhagen, Copenhagen, Denmark c.vania@ed.ac.uk, {yova|soegaard}@di.ku.dk, alopez@inf.ed.ac.uk Abstract Parsers are available for only a handful of the world’s languages, since they require lots of training data. How far can we get with just a small amount of training data? We sys- tematically compare a set of simple strate- gies for improving low-resource parsers: data augmentation, which has not been tested be- fore; cross-lingual training; and translitera- tion. Experimenting on three typologically diverse low-resource languages—North Sámi, Galician, and Kazah—We find that (1) when only the low-resource treebank is available, data augmentation is very helpful; (2) when a related high-resource treebank is available, cross-lingual training is helpful and comple- ments data augmentation; and (3) when the high-resource treebank uses a different writ- ing system, transliteration into a shared ortho- graphic spaces is also very helpful. 1 Introduction Large annotated treebanks are available for only a tiny fraction of the world’s languages, and there is a wealth of literature on strategies for parsing with few resources (Hwa et al., 2005; Zeman and Resnik, 2008; McDonald et al., 2011; Søgaard, 2011). A popular approach is to train a parser on a related high-resource language and adapt it to the low-resource language. This approach benefits from the availability of Universal Dependencies (UD; Nivre et al., 2016), prompting substantial re- search (Tiedemann and Agic, 2016; Agi´c, 2017; Rosa and Mareˇcek, 2018), along with the VarDial and the CoNLL UD shared tasks (Zampieri et al., 2017; Zeman et al., 2017, 2018). But low-resource parsing is still difficult. The organizers of the CoNLL 2018 UD shared task (Zeman et al., 2018) report that, in general, results on the task’s nine low-resource treebanks “are ex- tremely low and the outputs are hardly useful for downstream applications.” So if we want to build a parser in a language with few resources, what can we do? To answer this question, we systemat- ically compare several practical strategies for low- resource parsing, asking: 1. What can we do with only a very small target treebank for a low-resource language? 2. What can we do if we also have a source tree- bank for a related high-resource language? 3. What if the source and target treebanks do not share a writing system? Each of these scenarios requires different ap- proaches. Data augmentation is applicable in all scenarios, and has proven useful for low-resource NLP in general (Fadaee et al., 2017; Bergmanis et al., 2017; Sahin and Steedman, 2018). Transfer learning via cross-lingual training is applicable in scenarios 2 and 3. Finally, transliteration may be useful in scenario 3. To keep our scenarios as realistic as possible, we assume …

Content truncated to 500 words