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

FIGURE 5.

FIGURE 5

Classification results were based on four metrics including accuracy, precision, recall, and MCC (Matthews correlation coefficient) for selected features of FS1 and FS2. The classification models of MARS (Multi-adaptive regression splines), RF (Random Forest), and GPR (Gaussian process regression) were implemented. For accuracy, precision, and recall the values are categorized as follows: (—) for values < 0.3, (–) for 0.3 ≤ values < 0.6, (-) for 0.6 ≤ values < 0.7, (+) for 0.7 ≤ values < 0.8, (++) for 0.8 ≤ values < 0.9, and (+++) for value ≥ 0.9. Moreover, for the MCC metric, the categorization was done as: (—) for values < 0, (–) for 0 ≤ values < 0.1, (-) for 0.1 ≤ values < 0.3, (+) for 0.3 ≤ values < 0.7, (++) for 0.7 ≤ values < 0.9, and (+++) for value ≥ 0.9.