Q. What is a common initialization method for K-means clustering?
A.
Randomly selecting data points as initial centroids
B.
Using the mean of the dataset as the centroid
C.
Hierarchical clustering to determine initial centroids
D.
Using the median of the dataset as the centroid
Show solution
Solution
A common initialization method for K-means is to randomly select data points from the dataset to serve as the initial centroids.
Correct Answer:
A
— Randomly selecting data points as initial centroids
Learn More →
Q. What is the role of the 'k' parameter in K-means clustering?
A.
It determines the maximum number of iterations
B.
It specifies the number of clusters to form
C.
It sets the learning rate for the algorithm
D.
It defines the distance metric used
Show solution
Solution
The 'k' parameter in K-means specifies the number of clusters that the algorithm will attempt to form from the data.
Correct Answer:
B
— It specifies the number of clusters to form
Learn More →
Q. Which of the following is NOT a characteristic of K-means clustering?
A.
It can converge to local minima
B.
It can handle non-spherical clusters well
C.
It is sensitive to the initial placement of centroids
D.
It requires numerical input data
Show solution
Solution
K-means clustering struggles with non-spherical clusters, as it assumes clusters are spherical and equally sized.
Correct Answer:
B
— It can handle non-spherical clusters well
Learn More →
Q. Which of the following is NOT a common application of clustering methods?
A.
Market segmentation
B.
Image compression
C.
Spam detection
D.
Predictive modeling
Show solution
Solution
Predictive modeling is not a common application of clustering methods, as it typically involves supervised learning rather than unsupervised clustering.
Correct Answer:
D
— Predictive modeling
Learn More →
Showing 1 to 4 of 4 (1 Pages)
