0
$\begingroup$

I wanna perform Cluster Analysis and I have Mixed data (a couple of dummy variables). I've found some information about Gower's distance, but when I tried to use it, R broke down (i guess) because I have a large sample size (almost 1,000,000). Any advice?

$\endgroup$
0
$\begingroup$

Begin with a subset first. To learn what works and what doesn't.

Don't immediately try to scale.

Also you don't need to use all your data. If you find patterns on a subset, these supposedly are still patterns on the entire data. The purpose of clustering is not to label each point a, b, c. But rather to find actionable patterns and regularities.

| cite | improve this answer | |
$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.