Results of this run
OPTICS ordering of Customer Segments (minSamples 5, xi) cuts a reachability plot into 2 valleys. Reachability silhouette is 0.650, reachability Calinski-Harabasz is 347.6, reachability Davies-Bouldin is 0.399. Held-out segment versus valleys: adjusted Rand 0.216. 49 noise rows out of 200 stay on the noise floor. Valley 0 on Customer Segments holds 122 rows (annual_income 43229.4 / avg_spend 936.4). Valley 1 on Customer Segments holds 29 rows (annual_income 78173.2 / avg_spend 3119.0). Reachability sweep: min_samples 3 cuts 0.652 with 2 groups on Customer Segments; min_samples 5 cuts 0.650 with 2 groups on Customer Segments; min_samples 8 cuts 0.476 with 5 groups on Customer Segments. Ordering plot keeps 90.5%.

| min_samples | Silhouette | Clusters |
|---|---|---|
| 3 | 0.652 | 2 |
| 5 | 0.650 | 2 |
| 8 | 0.476 | 5 |
from sklearn.cluster import OPTICS
from sklearn.preprocessing import StandardScaler
Xs = StandardScaler().fit_transform(X)
labels = OPTICS(min_samples=5).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.
