Lecture 25 – Clustering, Part 2
Presented by Isaac Schmidt
Content by Isaac Schmidt
A reminder – the right column of the table below contains Quick Checks. These are not required but suggested to help you check your understanding.
Graphs. Vertices and edges. Types of data that are well-represented by graphs. Differences between graphs and tabular data.
The adjacency matrix, a mathematical representation of graphs. Properties of adjacency matrix.
The Laplacian matrix. Eigenvalues and eigenvectors of the Laplacian matrix.
A spectral clustering algorithm. Eigenvectors as features. Example of algorithm on toy graph.
Constructing a graph from point data. The distance matrix and nearest neighbors. Affinity Matrices.
Case study of spectral clustering on college basketball data.