Clustering and Association Rule Mining Quiz

What is the primary objective of clustering in data mining?

• Classification of data

• Grouping similar data points together

• Detecting anomalies in the data

• Predicting future trends

Which of the following is NOT a similarity measure used in clustering?

• Cosine similarity

• Euclidean distance

• Manhattan distance

• Linear regression

In hierarchical clustering, the process of merging clusters is referred to as:

• Agglomerative clustering

• Divisive clustering

• Partitional clustering

• K-means clustering

What is the main difference between hierarchical and partitional clustering?

• Hierarchical clustering produces non-overlapping clusters

• Partitional clustering uses a bottom-up approach

• Partitional clustering produces a single division of data

• Hierarchical clustering requires the number of clusters to be pre-defined

Which distance measure is most commonly used for continuous data in clustering?

• Hamming distance

• Euclidean distance

• Jaccard index

• Minkowski distance

What is the primary purpose of outlier detection in clustering?

• To improve the clustering process

• To remove data points that are far from the rest of the dataset

• To add more clusters

• To increase the computational efficiency

Which clustering algorithm works best for large datasets with a predefined number of clusters?

• K-means clustering

• DBSCAN

• Hierarchical clustering

• Agglomerative clustering

Which of the following is a limitation of partitional clustering algorithms like K-means?

• They cannot handle numeric data

• They are sensitive to the initial placement of centroids

• They work only with small datasets

• They are always slower than hierarchical methods

What is the primary goal of the DBSCAN clustering algorithm?

• To create a fixed number of clusters

• To identify clusters based on density

• To minimize the variance within clusters

• To find outliers

What is a major advantage of hierarchical clustering over K-means clustering?

• It does not require the number of clusters to be specified in advance

• It is faster

• It performs better on large datasets

• It is less computationally expensive