Skip to main content

Table 3 For a significance level of \(\epsilon = 0.05\) the Levene and Shapiro–Wilk tests show evidence that the metrics have homogeneous variances and that the residuals of the ANOVA model are normally distributed

From: Using amino acids co-occurrence matrices and explainability model to investigate patterns in dengue virus proteins

 

Leven’s test p value

ANOVA p value

Shapiro–Wilk p value

PR-AUC

0.83

\(4\times 10^{-5}\)

0.83

ROC-AUC

0.10

\(6\times 10^-5\)

0.09

Precision

0.07

\(4\times 10^{-4}\)

0.43

Recall

0.11

\(2\times 10^{-5}\)

0.53

F1-score

0.13

\(2\times 10^{-5}\)

0.63

  1. Finally, the null hypothesis of the ANOVA test is rejected, indicating that at least one of the metric means is different from the others