In the era of big data analysis, it is common to deal with datasets with a large number of features or dimensions. Visualization of data with high dimensionality is extremely challenging, as we will show later in this chapter, because we need to project all these dimensions to two-dimensional space (for example, a screen or paper).
In general, there are two types of dimensionality reduction approaches: linear and non-linear. Here are a few examples of each category for your information:
| Method | Type |
| Principal component analysis (PCA) | Linear |
| Linear discriminant analysis (LDA) | Linear |
| Generalized discriminant analysis (GDA) | Linear |
| t-distributed stochastic neighbor embedding (t-SNE) | Non-linear |
| Isomap ... |
