Automatic recognition and analysis of emotional support in teacher-child interaction
Abstract
This project develops an AI-driven multimodal system to measure and analyze children's curiosity, Zone of Proximal Development (ZPD), and teacher scaffolding across rural and urban China. Using computer vision, natural language processing, and machine learning, the study will track 600 teacher-child pairs (ages 36-72 months) through synchronized cameras, sensors, and digital tools. The research aims to identify cultural differences in curiosity expression and develop culturally-responsive interventions to enhance teachers' ability to support curiosity-driven learning. The collaboration between CUHK (educational expertise) and Fudan University (AI technology) will deliver evidence-based strategies for improving early childhood education quality across diverse Chinese settings.
