Corpus-based learning of Cantonese for Mandarin speakers

Abstract

This article presents the first study on using a parallel corpus to teach Cantonese, the variety of Chinese spoken in Hong Kong. We evaluated this approach with Mandarin-speaking undergraduate students at the beginner level. Exploiting their knowledge of Mandarin, a closely related language, the students studied Cantonese with authentic material in a Cantonese-Mandarin parallel corpus, transcribed from television programs. They were given a list of Mandarin words that yield a range of possible Cantonese translations, depending on the linguistic context. Leveraging sentence and word alignments in the parallel corpus, the students independently searched for example sentences to discover these translation equivalents. Experimental results showed that, in both the short- and long-term, this data-driven learning approach helped students improve their knowledge of Cantonese vocabulary. These results suggest the potential of applying parallel corpora at even the beginners’ level for other L1-L2 pairs of closely related languages.

Efficacy of ChatGPT in Cantonese Sentiment Analysis: A Comparative Study (Preprint)

“…The third linguistic peculiarity is that, compared with Standard Chinese, which is mainly used in formal writing, Cantonese is mostly informal and colloquial. In regions such as Hong Kong, Cantonese, when used in online communication channels, including social media platforms, forums, and online chats, is more informal and colloquial compared with its use in formal writing or speech, which often contains more typos or grammatical peculiarities [15]. The informal and colloquial nature of Cantonese poses significant challenges for NLP to capture the intended sentiments.…”

Corpora as open educational resources for language teaching

“…Finally, and importantly, for the readership of this journal, there has been growing empirical evidence that DDL works not only for English but also for LOTEs: Catalan (Marco & van Lawick, 2015), Chinese (e.g., Chen, Wu, Yang, & Pan, 2016;Wong & Lee, 2016); Czech (Osolsobě & Vališová, 2012), French (e.g., Chambers & O'Sullivan, 2004), German (e.g., Ortner & Weber, 2018;Rets, 2017;Vyatkina, 2016aVyatkina, , 2016b, Italian (e.g., Kennedy & Miceli, 2001, and Spanish (e.g., Benavides, 2015). See the appendix for resources for working with DDL in these languages.…”

EuroCALL 2023. CALL for all Languages - Short Papers

“…Previous studies have shown that MT helps reduce the cognitive burden of translations (Baraniello et al, 2016) and encourages self-directed learning (Godwin- Jones, 2015;Wong & Lee, 2016). At the same time, there are still ongoing debates among educators as to whether MT use by learners should constitute cheating (Jolly & Luiane, 2022), or if it turns learners into mere passive recipients of information (Innes, 2019).…”