Figure 6From: Automated vocabulary discovery for geo-parsing online epidemic intelligenceEvaluation with respect to MetaCarta tags. Precision, recall and F1-score with respect to MetaCarta labels for increasing dictionary cut-offs according to the λ threshold. Performances of models trained on T0 (1,000 alerts), T1 (2,500 alerts), T2 (5,000 alerts) and T1 with location and disease targets.Back to article page