Unsupervised Learning — ML Technical Interview Questions & Answers

Finding structure in data without labels.

Unsupervised Learning

  • Simple meaning: No label column. The sheet is only features. You look for structure: groups (clustering) or a simpler picture (PCA).
  • Picture: 200 students, no pass/fail. K-means might find “high hours / high marks” vs “low hours”. You still must check if the groups mean something — the algorithm does not know “truth”.
  • You cannot compute accuracy against y because there is no y. Use silhouette, plots, or a human looking at cluster profiles.
  • Trap: treating cluster IDs as official categories in a report without a check. Trap: calling PCA a classifier.

Technical Interview Tip: No labels. Groups or a simpler view. Sanity-check the groups.

Code Examples for Unsupervised Learning

Unsupervised Learning — working example

# Tiny 1D k-means-style update (k=2)
points = [1.0, 1.5, 2.0, 8.0, 8.5, 9.0]
c1, c2 = 1.0, 9.0
for _ in range(3):
    g1 = [p for p in points if abs(p - c1) <= abs(p - c2)]
    g2 = [p for p in points if abs(p - c1) > abs(p - c2)]
    c1 = sum(g1) / len(g1)
    c2 = sum(g2) / len(g2)
print(round(c1, 2), round(c2, 2), len(g1), len(g2))

Unsupervised Learning — clustering groups unlabeled points; k-means alternates assignment and centroid updates.

How to Explain Unsupervised Learning to an Interviewer

Unsupervised vs supervised?

No y vs has y

Two unsupervised jobs?

Clustering Dimensionality reduction