09. Quiz: How Does PCA Work?

Before getting into the code, let's check some of the main ideas of PCA.

Which of the following best describes the goal of PCA?

To reduce the dimensionality of an existing dataset to a smaller number of features.

To create a new dataset.

To maximize the variability in a dataset.

None of the above are related to the goal of PCA.

SOLUTION:

To reduce the dimensionality of an existing dataset to a smaller number of features.

You can also think of a principal component as a… (select all that apply)

linear combination of the original features in a dataset.

latent variable.

a new feature that can be used in a future analysis.

SOLUTION:

The amount of information lost when performing PCA can be thought of as

the variability in the original data.

the difference between the value we predict and the value that is actually calculated.

the distance from the original points to the created component.

SOLUTION:

the distance from the original points to the created component.