EasyDeepLearn

Interview: 'when should you NOT cluster?'

medium

Answer

  • (1) Data is truly unimodal / uniform — no structure to find.
  • (2) Very high dimensionality without reduction → distance concentration ruins similarity.
  • (3) The business question is regression / classification, not grouping.
  • (4) You need explanations, not groups (use decision trees or PDP).
  • (5) Sample size too small (< 100).
  • (6) You need causal answers → clustering can produce spurious segments.
  • Always sanity-check whether clustering is the right tool before running one.
Check yourself — multiple choice
  • Always cluster
  • Unimodal data / high-dim no DR / need classification-regression / need explanations / tiny sample / need causal — clustering isn't always the right tool
  • Random
  • Never

Don't cluster: unimodal / high-dim without DR / task is prediction / tiny data.

#interview#clustering

Practise Unsupervised Learning

214 interview questions in this topic.

Related questions