AI support for language learning

Roman YANGARBER
Department of Digital Humanities, University of Helsinki (UH)

We discuss how recent advances in language technology and educational data science can be leveraged to create machines that can help people learn effectively. We present our approach – Revita – which tries to simulate a good teacher’s assistant, by continually assessing the learner’s state and progress. An important aspect of our approach is letting the user learn from any content – arbitrary authentic texts, chosen by the learners themselves. The system tries to collect data about the learning process – about typical patterns of mistakes, paths of progress, etc. – to continually improve the teaching ability of the system.