Self-Tuning Inference Model for Settlement in Shield Tunneling: A Case Study of the Taipei Mass Rapid Transit System’s Songshan Line

Table 10

Average FPR, TPR, and AUC values.

Dataset	Training			Testing
	FPR (X)	TPR (Y)	AUC	FPR (X)	TPR (Y)	AUC
Original	0.002172	0.921618	0.959722	0.006503	0.864318	0.928907
PDDBS oversampling	0.008211	0.987896	0.989842	0.009798	0.982104	0.986153
PDDBS median sampling	0.010442	0.988281	0.988919	0.011312	0.982462	0.985574
SMOTE oversampling	0.034273	0.990761	0.978243	0.014878	0.984997	0.985060
SMOTE median sampling	0.015613	0.98888	0.986633	0.015781	0.985788	0.985003