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

Selecting the best distance metric

Answer

Choosing initial centroid positions

Answer

Deciding between hard and soft clustering

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

K-Means Clustering

Answer

DBSCAN

Answer

Self-Organizing Maps

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

It always converges faster than K-Means

Answer

It requires fewer hyperparameters

Answer

It works only with numerical features

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

The items are negatively correlated

Answer

The rule has no practical significance

Answer

The confidence of the rule is below threshold

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

Backpropagation with labeled training data

Answer

Gradient boosting on raw features

Answer

Dropout regularization during supervised training

Question 6

What does the term 'inertia' refer to in K-Means clustering?

Accepted Answer

The sum of squared distances between each point and its assigned cluster centroid

Answer

The number of iterations needed for convergence

Answer

The ratio of between-cluster to within-cluster variance

Answer

The probability of a point belonging to a specific cluster