Abstract: Human-human interactions are fascinating and complex phenomena. However, it is notoriously difficult to study, given the dynamic and multimodal (e.g., audio, verbal, and visual) nature of the interactions and sensing data characteristics. In this talk, I will share two studies that characterize and model parent-child interactions using audio/video data recordings, one in math education and another in early childhood parenting coach contexts, aiming for models that are both explainable and actionable. I will discuss the implications of those studies in designing AI-supported coaching systems for realizing societal impacts on a large scale. Time permitting, I will also share some of my other recent projects aiming to develop problem-solving and communication skills in data science education and the CAREER project, all in higher education settings.

Short Bio: Karen Chen is an Assistant Professor of Information Systems at the University of Maryland, Baltimore County (UMBC). Her research interests are applied machine learning and data science with applications in a variety of domains, in particular, those with complex, fine-grained multimodal data collected from humans. She directs the Informatics for Human Flourishing Lab (iFlourishing) lab at UMBC, aiming to design and develop data-driven human-centered systems to support individuals thriving in all aspects of their lives. She received a Ph.D. in Information Systems (2020) and master’s degrees in Information Systems (2004) and Statistics (2012), all from Carnegie Mellon University. At CMU, she was a fellow of the IES-funded Fellowship Program in Interdisciplinary Educational Research (PIER). In 2023, she led the Caselet project to become one of the winning teams in the DARPA AI Tools for Adult Learning competition. In 2024, she received the NSF CAREER Award for a project integrating learning analytics research with data science and data literacy education.