MS-DS Master of Data science Master of Data science Unsupervised Learning Techniques 1

Question 1

Which unsupervised learning algorithm assigns each data point to the nearest cluster centroid and iteratively updates centroids until convergence?

Accepted Answer

K-Means

Answer

OPTICS

Answer

Gaussian Mixture Models

Answer

Autoencoders

Question 2

In a Gaussian Mixture Model (GMM), the Expectation-Maximization (EM) algorithm alternates between which two steps?

Accepted Answer

Computing posterior probabilities and updating distribution parameters

Answer

Encoding and decoding latent variables

Answer

Building a dendrogram and cutting it at a threshold

Answer

Projecting data onto principal components and reconstructing the input

Question 3

What is the primary purpose of t-SNE (t-Distributed Stochastic Neighbor Embedding)?

Accepted Answer

Visualizing high-dimensional data in two or three dimensions while preserving local structure

Answer

Clustering high-dimensional data into a fixed number of groups

Answer

Removing correlated features before supervised learning

Answer

Detecting anomalies by measuring reconstruction error

Question 4

In hierarchical agglomerative clustering, which linkage criterion defines the distance between two clusters as the maximum pairwise distance between their members?

Accepted Answer

Complete linkage

Answer

Single linkage

Answer

Average linkage

Answer

Ward linkage

Question 5

An autoencoder trained without labels learns a compressed representation of input data in its bottleneck layer. What is this bottleneck representation called?

Accepted Answer

Latent code (or latent space)

Answer

Principal component

Answer

Cluster centroid

Answer

Eigenvector

Question 6

Which metric is commonly used to determine the optimal number of clusters in K-Means by plotting within-cluster sum of squares against K and looking for a sharp bend?

Accepted Answer

Elbow method

Answer

Gap statistic

Answer

Rand index

Answer

Davies-Bouldin index