d$.$ Remove the categorical attribute from the data.

In $[]$:

e$.$ Should the data be normalized or standardize for clustering? why?

In $[]$:

g$.$ Perform centroid analysis and give a name to each cluster.

In $[]$:

h$.$ Investigate the relationship between the clusters and the two categorical attributes that you removed. Which clu

ster has both hot and cold kinds of cereal? Which company only creates popular cereals that are not very nutritious?

In $[]$:

i$.$ The elementary public schools would like to choose a set of cereals to include in their daily cafeterias. Every

day a different cereal is offered, but all cereals should be healthy. The members of which cluster is better to be used?

Explain.

In $[]$

j$.$ Now we want to complement this analysis using PCA. Before applying PCA should we standardize or normalized the $d$

ataset?

In $[$

k$.$ Using the first few PCs$,$ come up with an annotated $3-$dimensional scatterplot that shows most of the variation in

the data. How much variation is shown? Make sure the figure has the element to guide the audience about the importance o

$f$ each $PC.$

In $[]$:

Looking at the $3-$dimensional scatterplot, would you say the choice of K$=7$ for K$-$means was good?

In $[]$