Indian Institute of Technology Madras
BS Degree Programme · Examination & Study Paper
Machine Learning Practice
K-means clustering
Official Paper
Examination Instructions & Guidelines
- Solve on notebook/paper first before checking solutions.
- MCQ: Exactly 1 option is correct. Negative marking applies.
- MSQ: One or more options correct. Select all to score.
- Scientific calculator allowed for all calculations.
- Official answer key is sealed at the end of this sheet.
End Term · 31 Aug 2025 (FN)End Term
- Q1MCQ1 Mark·Choose one option

- A

- B

- C

- D

- E

- A
- Q2MSQ2 Marks·Choose all correct options

- AIt helps determine the optimal number of clusters by plotting the number of clusters vs. inertia (within-cluster sum of squares).
- BThe ”elbow point” is where adding more clusters yields a significantly larger reduction in inertia.
- CThe ”elbow point” is where adding more clusters yields diminishing returns in reducing inertia.
- DIt always produces a clear and unambiguous elbow point.
- EIt can be visualized using a plot of number of clusters vs. distortion score.
End Term · 31 Aug 2025 (AN)End Term
- Q3MCQ1 Mark·Choose one option

- A

- B

- C

- D

- E

- A
- Q4MSQ2 Marks·Choose all correct options

- AThe elbow method involves plotting the number of clusters against the inertia and identifying a point where the decrease in inertia slows down.
- BThe elbow point represents the cluster count beyond which additional clusters do not significantly improve the model fit.
- CThe metric used in the elbow method is the silhouette score.
- DThe elbow point is always the smallest number of clusters in the plot.
- EThe elbow method may sometimes be inconclusive if no clear bend is visible in the curve.
End Term · 22 Dec 2024 (AN)End Term
- Q5MCQ2 Marks·Choose one option

- AThe total amount spent by each customer.
- BThe cluster assignments indicating which group each customer belongs to.
- CThe number of items purchased by each customer.
- DThe centroid coordinates of the clusters formed.
- Q6MCQ3 Marks·Choose one option

- AOutputs the number of iterations and the final silhouette score of the clusters.
- BOutputs the cluster labels for each data point and the maximum distance between clusters.
- COutputs the coordinates of the cluster centroids and the number of points in each cluster.
- DOutputs the coordinates of the cluster centroids and the sum of squared distances of samples to their closest cluster center.
- Q7MCQ3 Marks·Choose one option

- A

- B

- C

- D

- A
- Q8MSQ3 Marks·Choose all correct options

- A

- B

- C

- D

- A