14 September 2020 7 8K Report

I am using PCA to extract major meaningful data from many different parameters that are embedded in my remote sensing satellite imagery. I know that PCA is a dimension reduction model and the final components are perpendicular to each other and the very first PCAs are the most important ones and can explain the major part of my data's variance.

Since I am using R studio to extract my PCs, I first calculated the normalized matric and then proceed with the eigenvalues and eigenvectors. my first PC has around 80 percent of the total variance and a higher association with all parameters.

The only question here is that how can we physically explain the PC1 or PC2? what are their numbers tell us? imagine we have a PC1 range from -10 to +25, so how can I define the exact meanings behind these numbers?

More Nima Karimi's questions See All
Similar questions and discussions