Dimensionality Reduction
There are two ways to reduce the dimensions of any data set. Either you select some features and discard the others or you create some new features that allow you concisely represent the same amount of information as your existing features.
| Feature Selection | Feature Extraction |
|---|---|
| Backward Elimination | PCA |
| Forward Selection | LDA |
| Bidirectional Elimination | Kernel PCA |
| Score Comparison |
The things that we will cover in this section are as follows: