cluster analysis FADS

Cluster analysis can be seen as a data reduction technique.

• True

• False

The objective of clustering analysis is to find clusters that are:

• Homogeneous within and heterogeneous between

• Homogeneous between and heterogeneous within

• Big, no matter the number of them

Hierarchical clustering techniques (e.g. ward's, centroid's, etc) differ in...

• the way "distance" between 2 clusters is computed

• the criteria to choose the number of clusters to retain

• none of the options

Which of the following is an advantage of Hierarchical Cluster Analysis?

• No reallocation of observations

• Greedy algorithm

• No need to define number of clusters a priori

Which of the following is NOT a characteristic of k-means clustering:

• Observations can be reallocated after being assigned to a cluster

• The number of clusters needs to be defined a priori

• Resistant to outliers

Hierarchical clustering and K-means can be viewed as complementary techniques:

• True

• False