Data Science Data Science Unsupervised Learning Techniques Questions and Answers 3

Question 1

What problem does the 'elbow method' help solve when using K-Means clustering?

Accepted Answer

Determining the optimal number of clusters

Answer

The elbow method plots within-cluster sum of squares against the number of clusters and identifies the point where adding more clusters yields diminishing returns.

Question 2

Which unsupervised learning algorithm groups data points by modeling them as generated from a mixture of probability distributions?

Accepted Answer

Gaussian Mixture Models (GMM)

Answer

GMM assumes data is generated from a mixture of several Gaussian distributions with unknown parameters, using expectation-maximization to estimate them.

Question 3

What is a key advantage of DBSCAN over K-Means clustering?

Accepted Answer

It can discover clusters of arbitrary shape without specifying the number of clusters

Answer

DBSCAN identifies clusters based on density, allowing it to find arbitrarily shaped clusters and automatically determine the number of clusters.

Question 4

In association rule mining, what does the 'lift' metric indicate when its value is greater than 1?

Accepted Answer

The items appear together more often than expected by chance

Answer

A lift value greater than 1 indicates a positive association, meaning the items co-occur more frequently than would be expected if they were independent.

Question 5

Which technique uses an autoencoder neural network for unsupervised feature learning?

Accepted Answer

Training a network to reconstruct its input through a compressed hidden layer

Answer

Autoencoders learn compressed representations by training a neural network to reconstruct its own input through a bottleneck hidden layer.

Data Science Practice Test

Data Science Practice Test

Data Science Data Science Unsupervised Learning Techniques Questions and Answers 3