PCA (Principal Component Analysis) is a dimensionality reduction technique that is particularly well-suited for the visualization of high-dimensional datasets. It is based on linear algebra and preserves the most important variations in the data.
The PCA algorithm works by finding the directions of maximum variance in the data and projecting the data onto these directions.
The example below shows the 3D Swiss Roll data set and the corresponding2D UMAP layout. Both visualizations are interactive and allow you to explore the data by picking and dragging. The 3D visualization was created using Three.js. The 2D visualization was created using D3.js.