Clustering and Decision Trees Quiz

Clustering is a type of:

• Supervised learning

• Unsupervised learning

• Reinforcement learning

• Semi-supervised learning

K-means clustering partitions data into:

• Hierarchical groups

• K non-overlapping clusters

• Random subsets

• Tree-based structures

The “K” in K-means represents:

• Kernel function

• Number of clusters

• Number of features

• Sample size

K-means uses which metric for clustering?

• Euclidean distance

• Cosine similarity

• Manhattan distance only

• Hamming distance

A common method to determine the number of clusters is:

• Confusion matrix

• Elbow method

• ROC curve

• Decision tree

A drawback of K-means is:

• It requires predefining the number of clusters

• It is too slow for large datasets

• It only works with categorical data

• It guarantees global optimum

K-means clustering is best suited for:

• Small, categorical datasets

• Large, continuous numerical datasets

• Only time-series data

• Binary classification

A diagnostic technique for evaluating clustering quality is:

• Silhouette score

• ROC curve

• Accuracy percentage

• Gradient checking

A caution when using K-means is:

• It is insensitive to scaling

• It is sensitive to initial centroids

• It works only for time series data

• It doesn’t require iteration

Classification is a type of:

• Supervised learning

• Unsupervised learning

• Clustering method

• Reinforcement learning