SciPy - Clustering and Distance
Why does scikit-learn's KMeans use the 'k-means++' initialization by default instead of random initialization like scipy's kmeans2?
Why does scikit-learn's KMeans use the 'k-means++' initialization by default instead of random initialization like scipy's kmeans2?
hard📝 Conceptual Q10 of 15
Step-by-Step Solution
Solution:
Step 1: Understand initialization methods
'k-means++' selects initial centers to spread out clusters, improving convergence and results.Step 2: Compare with random initialization
Random initialization can lead to poor convergence and local minima; 'k-means++' reduces this risk.Final Answer:
Because 'k-means++' improves convergence speed and clustering quality. -> Option BQuick Check:
Initialization method benefit = Because 'k-means++' improves convergence speed and clustering quality. [OK]
Quick Trick: 'k-means++' helps clusters converge better [OK]
Common Mistakes:
- Thinking random init is unsupported in scikit-learn
- Believing 'k-means++' uses labels
- Assuming scipy requires random init
Master "Clustering and Distance" in SciPy
Want More Practice?
15+ quiz questions · All difficulty levels · Free
Free Signup - Practice All QuestionsMore SciPy Quizzes