Deep learning for prediction of population health costs

Table 1 Performance assessment depending on the network depth

	r	\(\rho\)	MAE	\(r^2\)	CPM
Neural network (depth 2)	0.518	0.566	2170.83	0.265	0.27-
Neural network (depth 3)	0.525	0.622	2065.49	0.271	0.304981
Neural network (depth 4)	0.524	0.631	2013.35	0.264	0.323
Neural network (depth 5)	0.530	0.622	2066.74	0.270	0.304561
Neural network (depth 6)	0.526	0.617	2165.04	0.275031	0.271482
Neural network (depth 7)	0.525	0.616	2165.54	0.272261	0.271315
Neural network (depth 8)	0.497	0.601	2414.74	0.244568	0.187461

Evaluation of methods using: Pearson’s correlation (r), Spearman’s correlation (\(\rho\)), mean absolute error (MAE), R squared (\(r^2\)) and Cumming’ s Prediction Measure (CPM)

ISSN: 1472-6947