AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Graph Theory - Beyond Just Time Series
When we look at basic clustering, correlation type clustering models, where you're looking to infer relationships, the techniques that I'm most familiar with are time, features, and topology. Do you ever see data sets moving from just X and Y into X, Y, and Z? Yes. What does something like that look like? I don't know if it's necessarily the right answer, but what immediately popped into my mind was graph theory. It's more than just sort of X and Y, you've got additional relationships as well. That's another area of interest.