Table 3 Values of Area Under Curve scores for ROC and Precision-Recall curves. Numbers in brackets represent the score achieved by a model which randomly predicts 1 or 0 in proportion to the dataset prior, averaged over 100 trials
From: Evaluating structural edge importance in temporal networks
Dataset | Balanced accuracy | ROC AUC | Precision-Recall AUC |
|---|---|---|---|
Bilateral trade | 0.542 (0.5 ± 0.0011) | 0.554 (0.5 ± 0.0013) | 0.628 (0.595 ± 0.0021) |
College messaging | 0.623 (0.5 ± 0.0027) | 0.678 (0.5 ± 0.0029) | 0.017 (0.005 ± 0.0045) |
Equity-1 | 0.568 (0.5 ± 0.0039) | 0.576 (0.5 ± 0.0046) | 0.365 (0.313 ± 0.0062) |
Equity-2 | 0.566 (0.5 ± 0.0061) | 0.579 (0.5 ± 0.0073) | 0.430 (0.351 ± 0.010) |
Equity-3 | 0.527 (0.5 ± 0.0025) | 0.542 (0.5 ± 0.0045) | 0.424 (0.381 ± 0.0030) |