Spectral Clustering on Wine Quality

See Spectral Clustering applied to the Wine Quality dataset (178 samples, 13 features). Interactive visualization, metrics, and analysis.

How Spectral Clustering Works

Spectral clustering uses eigenvalues of a similarity graph to find non-convex cluster structures like rings, spirals, and manifolds.

spectral clusteringgraph clusteringeigenvaluesLaplaciannon-convexmanifold

About the Wine Quality Dataset

178 wine samples with 13 chemical properties. Ideal for discovering natural groupings or predicting wine class.

Samples
178
Features
13
Type
Numeric
Category
Graph-based

Key Metrics to Watch

Silhouette Score

Measures how similar a point is to its own cluster vs. other clusters. Ranges from −1 to +1; higher is better.

Calinski-Harabasz Index

Ratio of between-cluster to within-cluster variance. Higher values indicate denser, well-separated clusters.

Davies-Bouldin Index

Average similarity between each cluster and its most similar cluster. Lower is better.

Inertia (Within-Cluster SSE)

Sum of squared distances from each point to its assigned centroid. Lower indicates tighter clusters.

When to Use Spectral Clustering

Spectral Clustering belongs to the Graph-based family of clustering algorithms. These methods model data as a graph and find clusters using spectral properties of the similarity matrix.

Related Examples