Interview: 'when should you NOT cluster?'
mediumAnswer
- (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
- Interview: 'you're asked to segment 10M customers — how?'
- Interview: cluster 5M e-commerce customers for a marketing campaign.
- Interview: how would you group log messages from a large distributed system?
- Interview: business asks 'why is customer X in Cluster 3?' — how do you explain?
- Interview: 'you have to pick k for the executive team — walk me through it.'
- Interview: after segmentation, one segment has 60% women but each region shows 40%. Why?