Results of this run
Bisecting K-Means repeatedly halves Wine (UCI wine recognition) down to k=3. Split silhouette is 0.252, split Calinski-Harabasz is 62.5, split Davies-Bouldin is 1.452, split inertia is 1350.21. Held-out wine_class versus the splits: adjusted Rand 0.643. 0 noise rows out of 178 after the bisections. Bisected group 0 on Wine (UCI wine recognition) holds 45 rows (proline 500.9 / magnesium 91.56). Bisected group 1 on Wine (UCI wine recognition) holds 62 rows (proline 1100.2 / magnesium 108.0). Bisected group 2 on Wine (UCI wine recognition) holds 71 rows (proline 594.3 / magnesium 97.75). Bisection path: k 2 halves to 0.265 with 2 groups on Wine (UCI wine recognition); k 3 halves to 0.252 with 3 groups on Wine (UCI wine recognition); k 4 halves to 0.230 with 4 groups on Wine (UCI wine recognition); k 5 halves to 0.170 with 5 groups on Wine (UCI wine recognition); k 6 halves to 0.176 with 6 groups on Wine (UCI wine recognition); k 7 halves to 0.153 with 7 groups on Wine (UCI wine recognition); k 8 halves to 0.124 with 8 groups on Wine (UCI wine recognition). Split scatter keeps 55.4%.

| k | Silhouette | Clusters |
|---|---|---|
| 2 | 0.265 | 2 |
| 3 | 0.252 | 3 |
| 4 | 0.230 | 4 |
| 5 | 0.170 | 5 |
| 6 | 0.176 | 6 |
| 7 | 0.153 | 7 |
| 8 | 0.124 | 8 |
from sklearn.cluster import BisectingKMeans
from sklearn.preprocessing import StandardScaler
Xs = StandardScaler().fit_transform(X)
labels = BisectingKMeans(n_clusters=3, random_state=42).fit_predict(Xs)
Who, how, and why
Method: StandardScaler, drop constant columns. Code version 3e4afc4b4352 (scripts/seo/build_explore_artifacts.py). scikit-learn 1.4.2. Seed 42. Dataset source: UCI Machine Learning Repository, wine recognition. License: CC BY 4.0. The wine_class column was held out of the fit and used only for the adjusted Rand score.
Results computed on 2026-11-02.
The results paragraph is filled from these numbers by a fixed template in the site code, so the scores can be checked against the same dataset in the tool.
Maintainer: nikolaou1982.
Contact: nick1982100@gmail.com.
