Explain 'Spectral Clustering'.

Explain 'Spectral Clustering'.

Approach: Describe clustering based on the eigenvalues of the similarity matrix.

A technique that makes use of the spectrum (eigenvalues) of the similarity matrix of the data to perform dimensionality reduction before clustering in fewer dimensions. Excellent for non-convex clusters.

Follow-up: What is the Graph Laplacian?

Key concepts: laplacian matrix, eigenvectors, graph partitioning.