Skip to main content
[Preprint]. 2023 Dec 4:2023.08.25.23294636. Originally published 2023 Aug 28. [Version 2] doi: 10.1101/2023.08.25.23294636

Table 7:

AD Phenotyping Performance on Balanced Testing Set in Experiment 1 BioClinical BERT)

Model Accuracy Precision Recall F1 NPV Specificity
Logistic Regression 0.7321 0.7241 0.7500 0.7368 0.7407 0.7500
SVM 0.7321 0.7826 0.6429 0.7059 0.6970 0.7857
Decision Tree 0.5893 0.6316 0.4286 0.5106 0.5676 0.7500
Random Forest 0.6964 0.7037 0.6786 0.6909 0.6897 0.8214
KNN 0.6786 0.7273 0.5714 0.6400 0.6471 0.7857
XGBoost 0.6071 0.6154 0.5714 0.5926 0.6000 0.8571
AdaBoost 0.6429 0.6538 0.6071 0.6296 0.6333 0.7857
Stacking Classifier 0.6964 0.7391 0.6071 0.6667 0.6667 0.7500