CONVERGING GENAI AND DATA-DRIVEN LEARNING FOR ESP MATERIALS DEVELOPMENT

Abstract

Data-driven learning (DDL) refers to an effective methodological approach to language teaching and learning that relies on using language corpus data and tools in combination with suitable pedagogical approaches (Crosthwaite & Boulton, in press). However, with recent developments in the field of Generative Artificial Intelligence (GenAI), there has been a need to merge the two approaches in order to bridge the existing gaps and transcend the limitations of DDL (Crosthwaite & Baisa, 2023; Mizumoto, 2023). Therefore, the current study exploits a corpus comprised of dental research abstracts generated by means of AntCorGen and analysed via Sketch Engine software tools in order to illustrate how DDL and GenAI can complement each other when it comes to assisting ESP materials design. For the purposes of designing an array of exercises and tasks intended for mixed-ability learners of English for Dental Medicine at a tertiary level, the inspiration has been drawn from Cambridge GenAI Idea Pack that offers teacher-friendly and easy-to-use prompt models applicable in various classroom settings. The assortment of classroom activities focused on field-specific collocation acquisition and tailored to our students’ specific learning context includes multiform vocabulary and listening exercises, as well as inductive learning activities. This approach to ESP materials design offers practical implications for converging corpus data and GenAI with the aim of facilitating student exposure to authentic language use.