Table 3 ROC-AUC: Plots the true positive rate against the false positive rate. AUC represents the area under this curve.

From: Identifying top ten predictors of type 2 diabetes through machine learning analysis of UK Biobank data

Model performance

Metric

Main model

Reduced model

ROC-AUC

0.903 (95% CI 0.900–0.909)

0.881 (95% CI 0.875–0.888)

Accuracy

0.924 (95% CI 0.922–0.925)

0.921 (95% CI 0.920–0.923)

Sensitivity

0.623 (95% CI 0.603–0.641)

0.569 (95% CI 0.549–0.587)

Specificity

0.932 (95% CI 0.930–0.934)

0.931 (95% CI 0.930–0.933)

F1-measure

0.311 (95% CI 0.300–0.323)

0.287 (95% CI 0.275–0.300)

Precision

0.207 (95% CI 0.198–0.217)

0.192 (95% CI 0.183–0.201)

PR-AUC

0.291 (95% CI 0.275–0.309)

0.255(95% CI 0.239–0.272)

  1. Accuracy: (TP + TN)/(TP + TN + FP + FN).
  2. Sensitivity (or Recall): TP/(TP + FN).
  3. Specificity: TN/(TN + FP).
  4. F1-measure: Harmonic mean of precision and sensitivity.
  5. Precision: TP/(TP + FP).
  6. PR-AUC: Area under the precision-recall curve, plotting precision against recall.
  7. TP = True Positive; TN = True Negative; FP = False Positive; FN = False Negative.
Back to article page