Corpus-based tasks for learning Chinese: a data-driven approach

Over the last couple of decades, there have been many studies on the utility of data-driven learning (DDL) approaches to the acquisition of English and other Indo-European languages. Very little research has touched on DDL for Chinese, or indeed any corpus-based approaches to learning it. This is surprising, given the otherwise large choice of IT applications, including flashcards, online dictionaries, and stroke order practice software: certainly, it seems, people do wish to learn Chinese with computers. Certain features of the Chinese language make it especially suited to a data-driven approach. In DDL, learners typically explore collocational and colligational patterns among words, but would not expect to be able to look at the internal structure of words using a corpus tool. The logographic Chinese writing system, however, allows the learner to investigate the ways that characters/morphemes pattern to form words. We offered several corpus-based tasks to intermediate-level Mandarin learners, alongside traditional-communicative conversation classes. We describe these tasks, as well as some of the corpora and corpus interfaces used in our approach