Combined Liver Stiffness and Α-fetoprotein Further beyond the Sustained Virologic Response Visit as Predictors of Long-Term Liver-Related Events in Patients with Chronic Hepatitis C

Table 3

Performance of decision tree and random forest classifiers in predicting LREs.

	AUC	Accuracy	F1	Precision	Recall
Average
Decision tree	0.790	0.967	0.960	0.968	0.967
Random forest	0.911	0.952	0.933	0.954	0.952
LRE: yes
Decision tree	0.790	0.967	0.564	1.000	0.393
Random forest	0.911	0.952	0.194	1.000	0.107
LRE: no
Decision tree	0.790	0.967	0.983	0.967	1.000
Random forest	0.911	0.952	0.975	0.952	1.000

Performances of the trained models were evaluated using AUC, accuracy ([true positives + true negatives]/total predictions), precision (true positives/[true positives + false positives]), recall (true positives/[true positives + false negatives]), and F1 (2/[(1/precision) + (1/recall)]). AUC, area under the curve; LRE, liver-related events.