The PCA (ML Engine) function outputs a set of principal components, and each principal component is a linear combination of the set of original predictors.
In the output table, table pca_health_ev_scaled, (from PCA Example) the first-ranked principal component is:
-0.082 * age + 0.387 * bmi + (-0.0935) * bloodpressure + 0.042 * glucose …
Using a similar equation, the PCAScore function uses the coefficients from the output table of the PCA function to compute a principal component score for each observation.
A common use of PCAScore output is principal component regression (PCR).