GBU_Quiz 3

Which of the following statements about unsupervised learning is correct?

• Unsupervised learning requires labeled data for training

• Unsupervised learning is used for classification tasks

• Unsupervised learning discovers patterns and structures in unlabeled data

• Unsupervised learning is only applicable to regression problems

In K-means clustering, what is the role of the centroid?

• Centroids represent the ground truth labels of the data.

• Centroids determine the number of clusters

• Centroids are the initial random points from which the algorithm starts.

• Centroids define the decision boundaries between clusters

Which of the following statements about hierarchical agglomerative clustering is correct?

• Hierarchical agglomerative clustering is a non-hierarchical clustering algorithm.

• Hierarchical agglomerative clustering creates a hierarchy of clusters.

• Hierarchical agglomerative clustering requires the pre-definition of the number of clusters.

• Hierarchical agglomerative clustering uses the k-nearest neighbors approach

What is the advantage of using the complete linkage method in hierarchical clustering?

• It tends to create compact clusters.

• It is computationally efficient for large datasets

• It is less sensitive to the initial choice of centroids.

• It can handle clusters with arbitrary shapes.

What does the silhouette coefficient measure in the context of clustering evaluation?

• The similarity between clusters.
  

• The similarity between clusters.
  

• The compactness and separation of clusters

• The time taken by the clustering algorithm to converge.

Which of the following scenarios is best suited for K-means clustering?

• Text document classification.

• Anomaly detection in network traffic

• Image recognition.

• Identifying customer segments based on purchasing behavior.

How does the choice of the number of clusters in K-means and hierarchical clustering impact the results?

• The number of clusters has no effect on the results.

• Choosing a higher number of clusters results in more fine-grained clusters

• Choosing a higher number of clusters results in more overlapping clusters

• Choosing a lower number of clusters results in higher accuracy

Which clustering algorithm is more sensitive to the initial choice of centroids?

• K-means clustering.

• Hierarchical agglomerative clustering

• Both are equally sensitive

• Neither of them is sensitive to the initial choice of centroids

Which of the following linkage methods is based on the shortest distance between two points in each cluster?

• Single Linkage

• Complete Linkage

• Average Linkage

• Centroid Method

What is unsupervised learning?

• Machine learning without any data

• Machine learning with labeled data

• Machine learning with unlabeled data

• Machine learning with feedback