K-Means Clustering Check for Understanding

What is the main purpose of clustering in unsupervised machine learning, such as K-Means clustering?

• To group similar data based on different features

• To predict the outcome of new data points

• To classify data into predefined categories

• To find the average value of data points

What does the 'K' in K-means clustering represent?

• The type of data being clustered

• The distance between data points

• The maximum number of iterations

• The number of clusters

What method is used to find the optimized 'K' value in K-means clustering?

• The centroid method

• The elbow method

• The silhouette score

• The maximum distance method

When using the elbow method what would be on the axis of the graph?

• Average Distance Between Clusters vs. K

• Number of Clusters vs. K

• Variation from Centroid vs. K

• Average Number of Data Points Within a Cluster vs. K

Based on this graph, what value of k is optimal?

• 2

• 3

• 4

• 5

What is the effect of increasing the number of clusters (k) in K-Means?

• Decreases the similarities of the data within a cluster

• Increases the similarities of the data within a cluster

• Increases the distance between centroids

• Increases the distance between clusters

What is the purpose of calculating Euclidean distance in K-means clustering?

• To calculate the total number of clusters

• To find the nearest centroid for each data point

• To measure the variability within a cluster

• To determine the shape of clusters

What do we use to find the Euclidean distance?

• SOCAHTOA

• Diameter of the cluster

• Distance formula

• Average distance between points

How does the K-Means algorithm initially determine cluster centroids?

• Randomly selecting data points as centroids

• Setting centroids to the mean of all data points

• Uses hierarchical clustering to determine centroids

• Selecting points with the minimum variation from the mean of all data points

Centroid for cluster 1 is (-6, 3), centroid for cluster 2 is (2, -5). The point P(1, 1) would be assigned to which cluster?

• Cluster 1

• Cluster 2

• Could be in either cluster