When is mini-batch k-means preferred?
mediumAnswer
- Large datasets (millions of points).
- Sample small random mini-batches (~1000 points), update centroids incrementally toward batch means with a learning rate.
- Slightly worse SSE than full k-means, but 10-100x faster and streaming-friendly.
- Standard for big-data clustering (Spark, scikit-learn's MiniBatchKMeans).
- For tabular data > 100k rows, prefer it over standard k-means.
Check yourself — multiple choice
- Same as regular
- Sample mini-batches (~1000 pts) → incremental centroid updates; 10-100x faster; standard for large / streaming data
- Only tiny data
- Not real
Mini-batch k-means: streaming-friendly, big-data alternative to full k-means.
#clustering
Practise Unsupervised Learning
214 interview questions in this topic.