3 Easy Steps to Perform Dimensionality Reduction Using Principal Component Analysis (PCA)

Running the PCA algorithm twice is the most effective way of performing PCA

Photo by Ashley Jurius on Unsplash

What is the dimensionality of a dataset?

In the context of both statistics and machine learning, the dimensionality of a dataset refers to the number of input variables (features) in the dataset.

If the dataset contains only two input variables as in the following image, it is called a two-dimensional dataset. In this case, the observations (data points) can be plotted in a 2D scatterplot.

Two-dimensional data (Image by author)

If we add another variable called Age to the same dataset, we implicitly add another dimension to the dataset. Now, the dataset becomes three-dimensional and the observations (data points) can be plotted in a 3D scatterplot.

Three-dimensional data (Image by author)

Likewise, the dataset is very high-dimensional when there are many variables in the dataset. It is…