Mean shift clustering — mechanism and use case.
hardAnswer
- Non-parametric mode-seeking.
- For each point, repeatedly shift it toward the local mean within a bandwidth-radius kernel until convergence.
- Points converging to the same mode form a cluster.
- No k needed — clusters = number of density modes.
- Bandwidth is the key parameter.
- Uses: image segmentation, mode tracking.
- Slow but robust to cluster shape.
Check yourself — multiple choice
- Random
- Iteratively shift each point toward local density mean (kernel bandwidth); auto-selects k = number of modes; used in image segmentation
- Same as k-means
- Not real
Mean shift: mode-seeking with kernel bandwidth; no k needed.
#clustering#density
Practise Unsupervised Learning
214 interview questions in this topic.