Skip to content

🧪 06 — Exercises: Clustering

Dataset

import pandas as pd

customers = pd.DataFrame({
    "spend": [500, 700, 800, 2500, 2800, 3200, 10000],
    "visits": [2, 3, 4, 10, 12, 11, 1],
    "tenure": [3, 4, 5, 24, 30, 28, 1]
})

Tasks

  • scale features
  • run K-Means with 2 and 3 clusters
  • calculate silhouette score
  • profile clusters
  • try DBSCAN
  • explain which clustering result is most useful

Next

➡️ Feature Engineering