Barium Titanate Semiconductor Band Gap Characterization through Gravitationally Optimized Support Vector Regression and Extreme Learning Machine Computational Methods

Table 3

Performance evaluation for the developed models.

Developed model		CC	MAE (ev)	RMSE (ev)
GSVR	Training set	1.0000	0.0001	0.0001
	Testing set	0.9266	0.2737	0.1817
	Entire dataset	0.9413	0.0364	0.1224
ELM-Sine	Training set	0.3166	0.2845	0.3498
	Testing set	0.8886	0.2199	0.2420
	Entire dataset	0.4267	0.2716	0.3310
ELM-sig	Training set	0.3167	0.2845	0.3498
	Testing set	0.8888	0.2200	0.2420
	Entire dataset	0.4267	0.2716	0.3310