Skip to main content

Table 7 Model performance in FAERS_SI dataset

From: Crossfeat: a transformer-based cross-feature learning model for predicting drug side effect frequency

Method

Binary classification

Regression

AUROC

AUPRC

RMSE

MAE

SDPred

0.83 ± 0.03

0.84 ± 0.02

0.81 ± 0.03

0.68 ± 0.02

SDPred_loss

0.85 ± 0.02

0.86 ± 0.01

0.77 ± 0.03

0.63 ± 0.04

Ridge regression

0.86 ± 0.01

0.85 ± 0.02

1.79 ± 0.04

1.49 ± 0.04

XGBoost

0.77 ± 0.01

0.84 ± 0.02

1.76 ± 0.04

1.42 ± 0.04

MLP

0.87 ± 0.01

0.87 ± 0.01

0.85 ± 0.05

0.71 ± 0.04

\(\hbox {CrossFeat}_{\textrm{MLP}}\)

0.85 ± 0.01

0.86 ± 0.01

0.93 ± 0.06

0.86 ± 0.03

CrossFeat

0.86 ± 0.01

0.87 ± 0.01

0.72 ± 0.04

0.57 ± 0.02

  1. Bold indicates the best result among all models listed in each metric