Exploring Problem Framing activity using BERTopic

Problem framing is a foundational aspect of the engineering design process, shaping how designers perceive challenges and potential solutions. Qualitiative methods, such as protocol analysis, have provided valuable insights about problem framing but are labor-intensive and time consuming. This study explores the use of a NLP technique BERTopic, to analyze framing in design conversations. BERTopic retains contextual nuances, offering a tool for uncovering the diversity and uniqueness of concepts explored by design teams while also making the analysis process more efficient. The results provide one representation of eight design group’s processes, highlighting the different and changing topic representations that emerge throughout a design session. The findings highlight the potential of NLP tools for enhancing our understanding of framing in design cognition and team dynamics.