Enhancing Free-text Interactions in a Communication Skills Learning Environment

Abstract

Learning environments frequently use gamification to enhance user interactions. Virtual characters with whom players engage in simulated conversations often employ pre- scripted dialogues; however, free user inputs enable deeper immersion and higher-order cognition. In our learning environment, experts developed a scripted scenario as a sequence of potential actions, and we explore possibilities for enhancing interactions by enabling users to type free inputs that are matched to the pre-scripted statements using Natural Language Processing techniques. In this paper, we introduce a clustering mechanism that provides recommendations for fine-tuning the pre-scripted answers in order to better match user inputs.