MS-DS Master of Data science Unsupervised Learning Techniques Questions and Answers Flashcards
6 cards from real MS-DS Master of Data science practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 6 MS-DS Master of Data science Unsupervised Learning Techniques Questions and Answers flashcards as text
What advantage does DBSCAN have over K-Means for spatial data with irregular cluster shapes?
Answer: DBSCAN can discover clusters of arbitrary shape based on density
DBSCAN groups together densely connected points regardless of cluster geometry, making it effective for non-convex and irregularly shaped clusters.
In K-Means clustering, what problem can arise from poor centroid initialization?
Answer: Convergence to a suboptimal local minimum
Poor initialization can cause K-Means to converge to a local minimum rather than the global optimum, resulting in suboptimal cluster assignments.
Which linkage criterion in hierarchical clustering measures the distance between the closest members of two clusters?
Answer: Single linkage
Single linkage defines inter-cluster distance as the minimum distance between any pair of points from the two clusters, which can produce elongated chain-like clusters.
What is the role of the covariance matrix in a Gaussian Mixture Model?
Answer: It captures the shape, orientation, and spread of each Gaussian component
Each component's covariance matrix defines the ellipsoidal shape, orientation, and spread of that Gaussian distribution in the feature space.
What does the reconstruction error of an autoencoder indicate about the input data?
Answer: How well the compressed representation captures the essential features of the input
Reconstruction error measures the difference between the original input and the autoencoder's output, indicating how effectively the bottleneck layer preserves important information.
Which method helps determine the optimal number of clusters for K-Means by plotting within-cluster sum of squares against cluster count?
Answer: The Elbow Method
The Elbow Method plots WCSS for different values of K and identifies the point where adding more clusters yields diminishing returns, forming an elbow shape.