How useful is it to model conversations as mathematical graphs?
Can we learn about how groups communicate by looking at how frequently they change topics, how many people contribute to a given topic, the content of a topic, and the segues (edges) between topics?
This repo currently contains a graph of the first 30 minutes of a conversation I had with some friends. It's a side project I find interesting but haven't had much time for. I haven't dug into the data yet.
I plan to turn the graph into a hashmap (dictionary) with topics as nodes/keys and segues to other topics as edges/values. If anyone wants to do it for me, I'd be forever thankful.