"Collaborative Temperature"

This project Constructed a low-cost, low-effort, ethical method to extract “collaboration temperature” from social structures in a large-scale co-located museum environment (Connected Worlds), captured through video data. For each frame of the video, a network arrangement joining people on the exhibit floor was constructed. Using the principles of proxemics, social network analysis different patterns of collaborative arrangements and features of collaboration were extracted. Upon clustering using KMeans algorithm this work defined a measure to gauge degree of collaboration within the social structures formed during large group social interaction. We term the combinations of these social structures as “collaborative temperature” as these can be used to understand the collaborative state of the visitors’ interactions in the exhibit. The “collaborative temperature” was used to understand the impact of an educational intervention, the use of learning analytics dashboard.