Table 4 Prediction Model Performance Evaluation Indicators.

Mean Absolute Error (MAE)

\(\frac{1}{n}\mathop \sum \limits_{i = 1}^{n} \parallel y_{i} - \widehat{{y_{i} }}\parallel\)

Progress prediction, continuous quality metrics

Root Mean Square Error (RMSE)

\(\sqrt {\frac{1}{n}\mathop \sum \limits_{i = 1}^{n} \left( {y_{i} - \widehat{{y_{i} }}} \right)^{2} }\)

High-precision numerical prediction tasks

F1-Score

\(\frac{{2 \cdot {\text{Precision}} \cdot {\text{Recall}}}}{{{\text{Precision}} + {\text{Recall}}}}\)

Risk classification, quality categorization

Area Under Curve (AUC)

\(\mathop \smallint \limits_{0}^{1} {\text{TPR}}\left( t \right)\,d{\text{FPR}}\left( t \right)\)

Binary risk assessment, anomaly detection

Mean Absolute Percentage Error (MAPE)

\(\frac{{100{\text{\% }}}}{n}\mathop \sum \limits_{i = 1}^{n} \frac{{y_{i} - \widehat{{y_{i} }}}}{{y_{i} }}\)

Relative error assessment, percentage-based metrics