Broad intuition · Study map 10 of 85
Clustering, compression, and anomalies
Suggested preparation: Linear models, trees, neighbors, and ensembles
Picture it
Clustering draws neighborhoods; dimensionality reduction finds a smaller set of coordinates; anomaly detection asks what does not fit. None of these automatically discovers the categories a human cares about.
What to understand
- k-means, density clustering, PCA, and autoencoders
- Distances, scaling, outliers, and the limits of visualization
Follow the references
1 · Documentation
scikit-learn · Unsupervised learning ↗Use clustering, decomposition, and novelty detection. Compare their assumptions on the same small dataset.
You understand it when…
Why can changing feature units completely change a clustering result?
This is a study map. Follow the reference for the full explanation; use the question to test your mental model. Browse the reference shelf.