Skip to main content

Table 4 Statistics of performances among different methods (95% CIs were generated by Wilson Score interval)

From: Federated learning algorithms for generalized mixed-effects model (GLMM) on horizontally partitioned data from distributed sources

  

Precision

Recall

F1-score

AUC

threshold

Centralized method with LA

Value

0.1507

0.6204

0.2425

0.6789

0.0900

Lower bound (0.95)

0.1474

0.6160

0.2386

0.6700

Upper bound (0.95)

0.1539

0.6248

0.2464

0.6878

GH with regularization

Value

0.1705

0.6546

0.2705

0.7178

0.0108

Lower bound (0.95)

0.1670

0.6503

0.2664

0.7091

Upper bound (0.95)

0.1739

0.6589

0.2745

0.7265