ML Case Labs — Validation, Boosting, Segmentation, Pipelines
Back to M3 — Clustering + Pipelines/SHAP

Segmentation Lab + Business Readout

Clusters + readout memo. 100-pt rubric inside.

71 minutesProject submission
🎯 Free Guest Mode: You are learning for free. Sign in to save your completion progress and quiz answers.

Assignment Instructions

Brief

Submit clusters on the unlabeled segmentation set plus a business readout memo (segment profiles + one targeting action each).

Dataset

Unlabeled segmentation set (no holdout labels; stability re-run is the check).

Starter guardrails (88)

  • Slots: k-selection evidence, profiles table, readout memo.
  • Profiles in business columns only (no scaled-axis jargon in the memo).

Deterministic checks

k-selection artifacts present (elbow + silhouette); seed-stability re-run reported; profiles recompute from assignments.

100-pt rubric

  • k justification (elbow + silhouette): 20
  • Profile quality (business columns): 25
  • Action-implying names + owners: 20
  • Stability evidence: 15
  • Readout persuasiveness: 20
Checking your access…