Investigating Visitor Engagement in Interactive Science Museum Exhibits with Multimodal Bayesian Hierarchical Models

摘要: Engagement plays a critical role in visitor learning museums. Devising computational models of engagement shows significant promise for enabling adaptive support to enhance visitors’ experiences and providing analytic tools museum educators. A salient feature science museums is their capacity attract diverse populations that range broadly age, interest, prior knowledge, socio-cultural background, which can significantly affect how visitors interact with exhibits. In this paper, we introduce Bayesian hierarchical modeling framework predicting learner Future Worlds, tabletop exhibit environmental sustainability. We utilize multi-channel data (e.g., eye tracking, facial expression, posture, interaction logs) captured from interactions fully-instrumented version Worlds model dwell time the museum. demonstrate proposed approach outperforms competitive baseline techniques. These findings point toward opportunities enriching our understanding multimodal analytics.