Hierarchical Clustering QUIZ (MCQ QUESTIONS AND ANSWERS)

Hierarchical clustering is more sensitive to outliers

Hierarchical clustering requires specifying the number of clusters in advance

Hierarchical clustering is less interpretable

Hierarchical clustering is more computationally expensive

Agglomerative clustering

Divisive clustering

K-means clustering

Ward's method

By applying the clustering algorithm to the features instead of the data points

By applying the clustering algorithm to the data points and using the cluster centroids as a reduced feature space

By applying the clustering algorithm to the data points and using the cluster labels as a reduced feature space

By applying the clustering algorithm to the data points and using the dendrogram to guide the selection of a reduced feature space

Hierarchical clustering does not require specifying the number of clusters in advance, while partitional clustering does

Hierarchical clustering uses a tree structure to represent the relationships between clusters, while partitional clustering does not

Hierarchical clustering can handle missing data, while partitional clustering cannot

Hierarchical clustering is deterministic, while partitional clustering is stochastic

Yes, by converting text data into numerical representations, such as term frequency-inverse document frequency (TF-IDF) vectors

Yes, by using cosine similarity as the distance metric

No, hierarchical clustering is not suitable for text data

No, hierarchical clustering can only be applied to continuous data

A measure of the agreement between the original distances between data points and the distances represented in the dendrogram

A measure of the compactness of clusters

A measure of the separation between clusters

A measure of both the compactness and separation of clusters

To determine the optimal number of clusters

To assess the quality and stability of the clustering solution

To identify the best linkage function for the dataset

To determine the most appropriate distance metric for the dataset

By using the silhouette score to rank the importance of features

By examining the dendrogram to identify redundant features

By selecting features that minimize the total within-cluster variance

By clustering features and selecting representative features from each cluster

DTW distance is invariant to time shifts and scaling

DTW distance is sensitive to noise

DTW distance is only suitable for continuous data

DTW distance is a similarity measure rather than a distance metric

It can handle missing data

It does not require specifying the number of clusters in advance

It can detect seasonality and trends in the data

It can capture the temporal structure of the data

Yes, by using imputation methods to fill in missing values

Yes, by using distance metrics that can handle missing data

No, hierarchical clustering requires complete data

No, hierarchical clustering is only suitable for categorical data

Euclidean distance

Manhattan distance

Cosine similarity

Pearson correlation coefficient

Bottom-up starts with each data point in a separate cluster, while top-down starts with all data points in a single cluster

Bottom-up is a deterministic approach, while top-down is a stochastic approach

Bottom-up is used for continuous data, while top-down is used for categorical data

Bottom-up is a density-based approach, while top-down is a distance-based approach

A predetermined number of clusters is reached

The minimum inter-cluster distance exceeds a threshold

The maximum intra-cluster distance falls below a threshold

The total within-cluster sum of squares is minimized

To measure the agreement between the original distances between data points and the distances represented in the dendrogram

To measure the compactness of clusters

To measure the separation between clusters

To determine the optimal number of clusters

K-means and K-medoids

Agglomerative and divisive

Density-based and grid-based

Model-based and graph-based

Yes, by identifying small clusters or isolated data points in the dendrogram

Yes, by identifying data points with a high silhouette score

No, hierarchical clustering is not suitable for outlier detection

No, hierarchical clustering can only be used for dimensionality reduction

A linkage method that minimizes the total within-cluster variance

A linkage method that maximizes the total between-cluster variance

A linkage method that minimizes the average distance between clusters

A linkage method that maximizes the centroid distance between clusters

Gower distance

One-hot encoding

Standardization

All of the above

Euclidean distance

Manhattan distance

Cosine similarity

Pearson correlation coefficient

Sensitivity to the choice of linkage function

Inability to handle large datasets

Inability to undo previous steps

All of the above

It does not require specifying the number of clusters in advance

It is more computationally efficient

It is less sensitive to the initial placement of centroids

It can handle categorical data

By maximizing the within-cluster sum of squares

By minimizing the between-cluster sum of squares

By examining the dendrogram and selecting an appropriate cut-off point

By using the elbow method on the resulting tree structure

The minimum distance between data points in two clusters

The maximum distance between data points in two clusters

The average distance between data points in two clusters

The centroid distance between two clusters

The minimum distance between data points in two clusters

The maximum distance between data points in two clusters

The average distance between data points in two clusters

The centroid distance between two clusters

The minimum distance between data points in two clusters

The maximum distance between data points in two clusters

The average distance between data points in two clusters

The centroid distance between two clusters

To determine the distance between data points

To determine the distance between clusters

To determine the optimal number of clusters

To determine the clustering algorithm to use

A diagram that represents the tree structure of hierarchical clustering

A diagram that represents the optimal number of clusters

A diagram that represents the density of clusters

A diagram that represents the distance between clusters

Each data point in a separate cluster

All data points in one cluster

A predefined number of clusters

The optimal number of clusters

Each data point in a separate cluster

All data points in one cluster

A predefined number of clusters

The optimal number of clusters