I'm working with a dataset with latitude, longitude and date-time. I have millions of points per day and I want to group them using a clustering algorithm.

I read this question Clustering geographic points but in my case I don't have an expected number of clusters. I explored HDBSCAN but it only accepts 2 features (lat lon in radians).

The idea is to get clusters of my datapoints regrouping nearest points in location and time together. So depending on the day, the output should change.

Is there a way to do a spatiotemporal clustering that includes the 3 features?

  • 2
    Is this a global data set or is it on a small enough region that it can be considered flat?
    – Spacedman
    May 22, 2020 at 6:25
  • thanks for the question. The region can be considered flat (geodetic point of view) May 22, 2020 at 13:02


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.