Practice Cluster Evaluation Metrics (6.1.3) - Unsupervised Learning – Clustering & Dimensionality Reduction

Practice: Cluster Evaluation Metrics

Lesson 6.1.3Unsupervised Learning – Clustering & Dimensionality Reduction

Practise this section

9 questions

Flashcard drill
3 cards from this section. Quick, good the night before a test.
Exercise
6 questions · about 4 min · pass mark 70%. Hints available.
Sign up to start the exercise
Quiz
3 questions · about 2 min · pass mark 70%. Hints available.
Sign up to start the quiz

6 sample questions

Question 1.
What does the Silhouette Score measure?
Show hintHide hint
Think about cluster similarity.
Question 2.
What does a lower Davies-Bouldin Index imply?
Show hintHide hint
Remember, lower is better!
3Question 3.
What does the Silhouette Score range from?
- 0 to 1
- -1 to 1
- 1 to 10
Show hintHide hint
Think about the lowest and highest scores.
4Question 4.
True or False: A lower Davies-Bouldin Index signifies better clustering.
- True
- False
Show hintHide hint
How do we want our index to behave?
Question 5.
You have a dataset with several distinct natural clusters. Upon using K-Means, you find a Davies-Bouldin Index of 1.5 and a Silhouette Score of 0.3. What actions could you take to improve your clustering?
Show hintHide hint
Think about what these indices tell you about the current clustering quality.
Question 6.
In an analysis, you're applying the Elbow Method and notice the elbow point is at K=4. However, you notice better cluster separation visually at K=6. How do you decide which K to use?
Show hintHide hint
Think of balancing visual observations with quantitative evaluations.

5 more questions in this section. Enroll free to practise them all.