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
Which unsupervised learning technique reduces dimensionality by finding orthogonal axes that maximize variance in the data?
Answer: Principal Component Analysis (PCA)
PCA identifies principal components as orthogonal directions of maximum variance, enabling dimensionality reduction while preserving the most information.
In DBSCAN, what is a 'core point'?
Answer: A point with at least MinPts neighbors within radius epsilon
DBSCAN defines a core point as one that has at least MinPts data points within its epsilon-radius neighborhood, forming the dense regions of clusters.
What does the silhouette coefficient measure in clustering evaluation?
Answer: How similar a point is to its own cluster compared to the nearest neighboring cluster
The silhouette coefficient ranges from -1 to 1 and compares intra-cluster cohesion with nearest-cluster separation for each data point.
Which technique is most appropriate for discovering hierarchical groupings without specifying the number of clusters in advance?
Answer: Agglomerative hierarchical clustering
Agglomerative hierarchical clustering builds a dendrogram by iteratively merging the closest clusters, allowing the analyst to choose the number of clusters after the fact.
In a Gaussian Mixture Model (GMM), what algorithm is typically used to estimate the model parameters?
Answer: Expectation-Maximization (EM)
The EM algorithm iterates between computing soft cluster assignments (E-step) and updating the Gaussian parameters to maximize likelihood (M-step).
What is the primary purpose of t-SNE in unsupervised learning?
Answer: Visualizing high-dimensional data in two or three dimensions
t-SNE is a nonlinear dimensionality reduction technique designed specifically for visualizing high-dimensional datasets in low-dimensional space while preserving local structure.