"PrincipalComponentsAnalysis" (Machine Learning Method)

Method for DimensionReduction, DimensionReduce, FeatureSpacePlot and FeatureSpacePlot3D.
Maps the data into a lower-dimensional space using the principal components analysis method.

Details & Suboptions

"PrincipalComponentsAnalysis" is a linear dimensionality reduction method. The method projects input data on a linear lower-dimensional space that preserves the maximum variance in the data.
The "PrincipalComponentsAnalysis" method works for datasets that have a large number of features and large number of examples; however, the learned manifold can only be linear.
The following plots show the results of the "PrincipalComponentsAnalysis" method applied to benchmark datasets including Fisher's Irises, MNIST and FashionMNIST:

"PrincipalComponentsAnalysis" is equivalent to the "Linear" and "LatentSemanticAnalysis" methods when the data is standardized.

open allclose all

Train a linear dimensionality reduction using the "PrincipalComponentsAnalysis" method from a list of vectors:

Use the trained reducer on new vectors:

Load the Fisher Iris dataset from ExampleData:

Generate a reducer function using "PrincipalComponentsAnalysis" with the features of each example:

Group the examples by their species:

Reduce the dimension of the features:

Visualize the reduced dataset:

Top