Corruption red flags in public procurement: new evidence from Italian calls for tenders

EPJ Data Science

Table 8 Predictive accuracy across samples and models

Note: The measures presented in the table are calculated as follows: Precision = Tp/(Tp + Fp), Recall = Tp/(Tp + Fn), F = 2⋅Precision⋅Recall/(Precision + Recall). Where Tp indicates the true positives, Fn indicates the false negatives, and Fp indicates the false positives.