Using lecture slides to create an academic corpus

The idea of supporting language learning with the use of corpora has been around since at least 1997, when Tim Johns coined the term data-driven learning (DDL). This approach invites learners to tease out patterns from authentic text, and test their own linguistic hypotheses in the manner of a mini research project; it has an intuitive appeal to teachers who favour student-centred or inductive learning. In this study, students constructed and consulted their own web corpora, based on presentation slides from their subject lectures. This was intended to help them build up their subject vocabulary, as well as access authentic texts from their discipline. Although no firm conclusions about the success of the method were reached, and no formal evaluation was conducted, feedback gained from participants was positive. The corpus query tool Sketch Engine (http://sketchengine.co.uk/) was used for the study. It includes a web corpus generating module, WebBootCat. Readers may also be interested in the freely available version of BootCat, at http://bootcat.sslmit.unibo.it/.