I have a dataset of about 53000 points. It has been clustered twice, based on two sets of unrelated attributes. For the first clustering (clustering 1) I used DBScan, and it ended up with about 700 clusters, with 30000 of the points being labeled as noise. Then, on another set of attributes, I clustered them using kmeans, which resulted in 5 different attributes (clustering 2).

However, I have a feeling the two sets of attributes may have some relation after all. Therefore, I want to measure how variable the points are in terms of cluster membership of clustering 2 within each cluster of clustering 1. I don't have a strong mathematical background, I've tried simply calculating the SD within each clustering 1 group, but this does not work since the cluster labels are of influence (a group with 50% in cluster 0 and 50% in cluster 1 will have different results than a group with 50% in cluster 0 and 50% in cluster 2)

Unfortunately I cannot share my data, but I have included a screenshot, I hope that clarifies what I mean.

enter image description here


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.