Identifying the most important variables in data with PCA

We've seen PCA in use in the Learning groups in data without prior information recipe as a dimensionality reduction technique—a method for reducing the size of our dataset whilst retaining the important information. As you might imagine, that means that we can get an idea of which of the original variables are contributing most to our reduced representation and we can, therefore, work out which are the most important. We'll see how that works in this recipe.

Getting ready

For this recipe, we'll use the factoextra package and the built-in iris dataset.

...