SyncFlow: A Scalable Platform for Multimodal Learning Analytics

The new wave of educational technologies (EdTech) is revolutionizing digital education but faces challenges with the complexities of multimodal human interactions in computer-based learning environments (CBLEs). Researchers are investigating multimodal learning analytics (MMLA) as a comprehensive approach to analyzing and supporting students. However, the integration of MMLA into scalable and automated learning environments is difficult because of the absence of standardized solutions for reliable multimodal data collection and analysis. Current MMLA systems are limited in their compatibility with modern web technologies and infrastructure for browser and Internet-of-Things (IoT) integration. To address these challenges, we introduce SyncFlow, an open-source platform offering scalable, robust cloud infrastructure for automated MMLA deployments. This paper presents an end-to-end application of SyncFlow, demonstrating its integration with AI-powered CBLEs and illustrating its capabilities. SyncFlow bridges critical gaps in MMLA data collection and processing, supporting scalable and impactful CBLEs in real-world settings.