Skip to main content

Journal of Population Sciences

Genus Cover Image

Table 2 Model accuracy metrics for all models as evaluated on the test data

From: Machine learning approach for predicting under-five mortality determinants in Ethiopia: evidence from the 2016 Ethiopian Demographic and Health Survey

Confusion matrix   Random Forest Logistic regression KNN model
   Predicted Predicted Predicted
   Alive Dead Alive Dead Alive Dead
Observed Alive 698 343 632 418 475 566
  Dead 12 28 16 24 14 26
    %   %   %
Accuracy    67.2   59.9   46.3
Sensitivity    98.3   97.5   97.1
Specificity    7.5   5.34   4.4
Positive predictive value    70.0   59.9   45.6
Negative predictive value    67.0   60.0   65.0
AUC    72.0   66.1   55.5