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

Table 2 Disintegration time 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	85.5:10	5.54	10.65	12.34	23.96	0.92	0.77	0.63	0.44	0.83	0.60	0.92	0.73
ANN	90.5:5	11.89	12.00	34.64	23.52	0.43	0.52	0.62	0.47	0.70	0.54	0.77	0.67
GRNN	90:10	0.47	12.88	5.25	31.90	0.98	0.35	0.98	0.34	0.98	0.54	0.99	0.64
DT	80:20	0.009	16.38	0.26	34.72	0.99	0.36	0.99	0.29	0.99	0.40	1.0	0.50
MLR	80:20	22.18	23.27	37.19	36.27	0.31	0.41	0.18	0.24	0.24	0.24	0.31	0.30
SVR	80:20	20.29	19.68	40.32	34.50	0.25	0.24	0.24	0.22	0.32	0.29	0.39	0.35
RF	80:20	4.72	11.99	10.15	23.33	0.95	0.70	0.68	0.34	0.80	0.46	0.86	0.56

ISSN: 1472-6947