BMC Medical Informatics and Decision Making

Table 3 Predictive performance of models using pre-assigned (P) or learned (L) weights in the weighted sampling (WS) strategy

From: Learning temporal weights of clinical events using variable importance

	Accuracy			AUC			AUPRC
ADE	PWS	LWS		PWS	LWS		PWS	LWS
D611	76.46	90.38		0.887	0.946		0.949	0.981
D642	97.79	99.47		0.979	0.983		0.814	0.936
D695	71.10	87.74		0.895	0.923		0.806	0.881
E273	81.78	79.92		0.615	0.650		0.264	0.272
G620	91.64	92.22		0.815	0.837		0.334	0.441
I952	91.71	91.64		0.557	0.554		0.116	0.118
L270	67.14	73.07		0.843	0.828		0.791	0.795
L271	84.18	85.22		0.753	0.759		0.356	0.406
M804	95.17	95.12		0.571	0.598		0.072	0.079
M814	98.71	98.69		0.702	0.722		0.039	0.052
O355	82.13	96.49		0.980	0.989		0.948	0.977
R502	99.16	99.22		0.840	0.774		0.148	0.280
T782	94.81	94.86		0.741	0.754		0.188	0.255
T783	83.33	86.16		0.771	0.788		0.393	0.524
T784	66.86	75.46		0.762	0.761		0.830	0.829
T801	90.35	93.55		0.868	0.880		0.533	0.635
T808	77.93	84.38		0.869	0.879		0.954	0.954
T886	94.95	94.86		0.741	0.738		0.159	0.190
T887	60.19	69.72		0.762	0.771		0.765	0.802
Average	84.49	88.85		0.787	0.797		0.498	0.548
p-value	0.002838			0.01597			0.00001907

Bold indicates winning

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com