What’s after?

We used PCA and received modified data, with a smaller dimension. What will be our next step?

As mentioned earlier, the data obtained is used in machine learning models, i.e. PCA basically acts only as a data processing step.

Once we have our data processed by PCA, we can use it in any machine learning model. It can be a model that solves the problem of classification, regression, clustering, etc. As an example, the use of PCA when working with images, because datasets are often large and with a lack of capacity, a dataset with more than 500,000 images can already become difficult to process. A few examples of how PCA can be used on its own:

• Visualization of multidimensional data

• Information compression

And examples of when PCA is used as a data process:

• Data dimension reduction

• Noise reduction in data

In the following chapters, we will explore in detail some of the most common uses for PCA.