Results of this run
Bisecting K-Means repeatedly halves Customer Segments down to k=4. Split silhouette is 0.352, split Calinski-Harabasz is 346.0, split Davies-Bouldin is 1.171, split inertia is 158.84. Held-out segment versus the splits: adjusted Rand 0.768. 0 noise rows out of 200 after the bisections. Bisected group 0 on Customer Segments holds 53 rows (annual_income 77503.7 / avg_spend 2909.1). Bisected group 1 on Customer Segments holds 46 rows (annual_income 55553.5 / avg_spend 1618.1). Bisected group 2 on Customer Segments holds 58 rows (annual_income 30574.9 / avg_spend 717.2). Bisected group 3 on Customer Segments holds 43 rows (annual_income 47175.3 / avg_spend 606.5). Bisection path: k 2 halves to 0.602 with 2 groups on Customer Segments; k 3 halves to 0.469 with 3 groups on Customer Segments; k 4 halves to 0.352 with 4 groups on Customer Segments; k 5 halves to 0.254 with 5 groups on Customer Segments; k 6 halves to 0.238 with 6 groups on Customer Segments; k 7 halves to 0.187 with 7 groups on Customer Segments; k 8 halves to 0.163 with 8 groups on Customer Segments. Split scatter keeps 90.5%.

| k | Silhouette | Clusters |
|---|---|---|
| 2 | 0.602 | 2 |
| 3 | 0.469 | 3 |
| 4 | 0.352 | 4 |
| 5 | 0.254 | 5 |
| 6 | 0.238 | 6 |
| 7 | 0.187 | 7 |
| 8 | 0.163 | 8 |
from sklearn.cluster import BisectingKMeans
from sklearn.preprocessing import StandardScaler
Xs = StandardScaler().fit_transform(X)
labels = BisectingKMeans(n_clusters=4, 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: Synthetic customer records bundled with this site. License: Generated for demos; no external license. The segment 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.
