Table 4.

Performance comparison of ML models on eight real datasets described in Table 1

Dataset	SVM	RF	GB	MNB	LR1	LR2
F1-macro
CBH	0.78(0.03)	0.73(0.03)	0.74(0.04)	0.66(0.03)	0.41(0.04)	0.17(0.01)
CSS	0.63(0.07)	0.58(0.08)	0.48(0.05)	0.49(0.03)	0.26(0.03)	0.24(0.02)
HMP	0.97(0.01)	0.97(0.01)	0.95(0.01)	0.95(0.01)	0.94(0.01)	0.93(0.01)
CS	0.88(0.05)	0.87(0.05)	0.74(0.06)	0.76(0.04)	0.16(0.04)	0.19(0.06)
FS	0.94(0.03)	1.00(0.01)	0.91(0.06)	0.98(0.01)	0.60(0.05)	0.58(0.04)
FSH	0.68(0.04)	0.63(0.08)	0.55(0.06)	0.50(0.04)	0.17(0.01)	0.17(0.00)
IBD	0.68(0.04)	0.57(0.02)	0.65(0.02)	0.43(0.01)	0.47(0.02)	0.43(0.01)
PDX	0.29(0.13)	0.28(0.09)	0.35(0.05)	0.18(0.03)	0.15(0.01)	0.15(0.01)
F1-micro
CBH	0.93(0.02)	0.91(0.02)	0.89(0.02)	0.88(0.02)	0.76(0.02)	0.68(0.00)
CSS	0.71(0.03)	0.67(0.03)	0.57(0.04)	0.58(0.03)	0.48(0.03)	0.48(0.03)
HMP	0.97(0.01)	0.97(0.01)	0.95(0.01)	0.95(0.01)	0.94(0.01)	0.93(0.01)
CS	0.88(0.06)	0.88(0.04)	0.75(0.05)	0.75(0.05)	0.23(0.05)	0.28(0.07)
FS	0.94(0.03)	1.00(0.01)	0.91(0.06)	0.98(0.01)	0.68(0.03)	0.67(0.03)
FSH	0.70(0.08)	0.69(0.05)	0.58(0.06)	0.62(0.03)	0.33(0.01)	0.33(0.01)
IBD	0.79(0.02)	0.78(0.02)	0.77(0.02)	0.76(0.02)	0.76(0.02)	0.76(0.02)
PDX	0.44(0.07)	0.43(0.07)	0.40(0.05)	0.42(0.04)	0.42(0.04)	0.42(0.04)

We consider several existing supervised ML methods. For each experiment, we consider 10-fold cross-validation and use F1-macro and F1-micro scores to quantify performance as defined in Classification performance metrics. For each fold, we perform five simulation runs with standard deviations shown between round brackets