Table 1.

Measures for evaluating the performance of the DNN. N = number of examples, tp, tn, fp; fn, values of the confusion matrix; tp, true positive, tn, true negative, fp, false positive, fn, false negative.

Measure	Equation
Accuracy	$\frac{\mathit{tn}+\mathit{tp}}{N}$
Precision	$\frac{\mathit{tp}}{\mathit{tp}+\mathit{fp}}$
Recall	$\frac{\mathit{tp}}{\mathit{tp}+\mathit{fn}}$
F1 Score	$\frac{2\cdot \mathit{precision}\cdot \mathit{recall}}{\mathit{precision}+\mathit{recall}}$
Conditional Kappa	$\frac{\mathit{precision}-\frac{\mathit{tp}+\mathit{fn}}{N}}{1-\frac{\mathit{tp}+\mathit{fn}}{N}}$
True Skill Statistic	$\mathit{precision}+\frac{\mathit{tn}}{\mathit{tn}+\mathit{fn}}-1$
Gleichlaeufigkeit	$G_{\mathit{ix}}=\left\{\begin{array}{c}{x}_{i+1}-x_i>0:+\frac{1}{2}\\ x_{i+1}-x_i=0:0\\ x_{i+1}-x_i>0:-\frac{1}{2}\end{array}\right\}$ $G_{\mathit{xy}}=\frac{1}{n-1}{\displaystyle \sum _{i=1}^{n-1}\mathit{abs}(G_{\mathit{ix}}+G_{\mathit{iy}})}$