Cluster Analysis. Read Section 10.3 of Introduction to Statistical Learning with Application in R, and answer the following questions

1) (True/False) If you run k-means cluster analysis multiple times, you can get different answers.

2) (True/False) k-means cluster analysis works with both numeric and categorical data (in other words, without converting the categorical data to numeric data before using the algorithm).

3) In hierarchical clustering, when computing dissimilarity between two clusters, which will give a greater value?

a) the complete linkage value will be at least as big as the average linkage value

b) the average linkage value will be at least as big as the complete linkage value

c) it just depends on the clusters

4) (True/False) In Fig. 10.9 of our text, whenever two leaf nodes have the same parent node, then they are more similar to each other than two leaf nodes that don't have the same parent node.

5) (True/False) In both k-means and hierarchical cluster analysis, the number of clusters must be specified in advance.

6) In Eq. 10.10 of our text, what does variable j represent?

a) a feature

b) a feature vector ("observation")

c) a cluster

7) In Fig. 10.7 of our text, what do the numbers above the figures represent?

a) whether the results are subjective or not

b) the time it took for the algorithm to converge

c) how good the clusters are

8) In Fig. 10.9 of our text, what do the values on the y axis represent?

a) cluster size

b) dissimilarity value

c) iteration count in the hierarchical clustering algorithm

9) In Fig. 10.10 (left) of our text, which are more similar:

a) observations 2 and 9

b) observations 6 and 4

10) Your data should be scaled before performing cluster analysis.

a) true

b) false

c) it depends on the situation