Manolis Mavrikis

University College London, UK

Affect-aware support for exploratory learning environments

This iTalk2learn project developed and evaluated affect-aware intelligent support components that tailor feedback according to students’ affective state as deduced both from speech and interaction. The affect prediction is used to determine which type of feedback is provided and how that feedback is presented (interruptive or non-interruptive). The talk will first present the process by which we `trained’ the system from data gathered in a series of ecologically-valid Wizard-of-Oz studies. Additionally, there will be an emphasis on empirical results from realistic classroom settings that point to the potential and positive impact of affect-aware intelligent support in relation to students’ learning but also interaction with the system, particularly off-task behaviour.

Download Manolis Mavrikis’ presentation as a pdf file

Visit Manolis Mavrikis’ webpage.