A prediction model based on artificial intelligence techniques for disintegration time and hardness of fast disintegrating tablets in pre-formulation tests

Table 3 Hardness tablet response variable

Method	Split Train . Validation :test	MAE		RMSE		R²		Accuracy tolerance = 10%		Accuracy tolerance = 15%		Accuracy tolerance = 20%
Method	Split Train . Validation :test	Train	Test	Train	Test	Train	Test	Train	Test	Train	Test	Train	Test
DNN	90.5:5	0.13	0.20	0.35	0.28	0.92	0.91	0.92	0.85	0.96	0.98	0.98	0.99
ANN	90.5:5	0.20	0.24	0.65	0.38	0.74	0.82	0.87	0.79	0.93	0.89	0.95	0.95
GRNN	90:10	0.009	0.35	0.05	0.85	0.99	0.20	0.99	0.74	0.99	0.83	0.99	0.89
DT	80:20	0.0005	0.31	0.01	0.52	0.99	0.87	1.0	0.67	1.0	0.80	1.0	0.88
MLR	80:20	0.65	1.72	1.07	19.47	0.27	0.23	0.39	0.36	0.53	0.50	0.65	0.62
SVR	80:20	0.60	0.65	1.08	1.46	0.20	0.10	0.47	0.45	0.60	0.57	0.70	0.68
RF	80:20	0.11	0.32	0.23	0.85	0.96	0.68	0.96	0.73	0.98	0.87	0.99	0.92

ISSN: 1472-6947