Recent Contributions of Data Mining to Language Learning Research

This paper will review the role of data mining in research on second language learning. Following a general introduction to the topic, three areas of data mining research will be summarized - clustering techniques, text-mining, and social network analysis - with examples from both the broader field and studies conducted by the authors. The application of data mining in second language learning research is relatively new, and more theoretical and empirical support is needed in the appropriate collection, use, and interpretation of data for specific research and pedagogical objectives. The three examples that we introduce illustrate how new data sources accessible in online environments can be analyzed to better understand the optimal instructional context for corpus-based vocabulary learning (clustering technique), characteristics and patterns of collaborative written interaction using Google Docs (text mining and visualizations), and issues of access and community in computer-mediated discussion (social network analysis). Implications of these new techniques for L2 research will be discussed.