Datasets used in the published runs

Iris, wine, Wholesale Customers, and Online Retail from the UCI Machine Learning Repository, the 200-row Mall Customers tutorial table, the IBM Telco Customer Churn sample, the credit-card fraud table from Worldline and the ULB Machine Learning Group, and synthetic survival records.

These tables are what Copilot clusters when you ask in English.

Iris Flowers dataset

Classic 150-sample dataset with 4 petal/sepal measurements across 3 species. The gold standard for clustering & classification demos.

150 rows, 4 numeric columns. License: CC BY 4.0.

Wine (UCI wine recognition) dataset

178 samples and 13 chemical measurements from three cultivars. The column held out of clustering is the cultivar, not a quality score.

178 rows, 13 numeric columns. License: CC BY 4.0.

Wholesale Customers dataset

440 wholesale customers and 6 annual spending columns. Channel and Region are not in this table.

440 rows, 6 numeric columns. License: CC BY 4.0.

Mall Customers dataset

200 shoppers with age, annual income, and spending score. A small tutorial table for segmentation, not a scale benchmark.

200 rows, 3 numeric columns. License: Tutorial table hosted on this site.

Telco Customer Churn dataset

7,043 telecom customers and whether they left. The full IBM sample table for churn classification.

7,043 rows, 21 columns. License: Apache 2.0.

Online Retail dataset

541,909 invoice lines from a UK online retailer, 1 December 2010 through 9 December 2011. The segmentation blueprint uses a customer table rolled up from these lines.

541,909 rows, 8 columns. License: CC BY 4.0.

Titanic Survival dataset

400 synthetic passenger records with class, age, fare, and family size. Binary classification demo for who survived.

400 rows, 7 numeric columns. License: Generated for demos; no external license.

Credit Card Fraud dataset

284,807 card transactions. Class is 1 for the 492 fraud rows. The published run uses a 100,000-row sample that keeps every fraud row.

284,807 rows, 31 columns. License: ODbL and DbCL.