Skip to main content
. 2020 Jul 23;8:851. doi: 10.3389/fbioe.2020.00851

FIGURE 4.

FIGURE 4

Feature selection results for features selected from (A) feature set 1 and (B) feature set 2. MARS (Multi-adaptive regression splines), RF (Random Forest), GPR (Gaussian process regression) were the surrogate models implemented for the selection of the features that have a significant impact on classification. The other feature selection methods were PCA (Principal component analysis) and correlations between the variables.