Assessment of Artificial Intelligence Models for Developing Single-Value and Loop Rating Curves

Table 5

Comparing SSE of different methods for developing loop rating curves.

Method	Data part	Dataset 5	Dataset 6	Dataset 7	Dataset 8
Conventional	Training	68050636919	6057	476798368	10082970
	Testing	21822676000	1814	24906002	3018398
EA	Training	67935976928	6006	87692181	10047888
	Testing	21780067017	1916	24133003	2992628
RF	Training	26498285944	2462	61379579	1172853
	Testing	12548719606	3574	15944107	8379269
MHBMO	Training	46177882487	6006	89711266	10047482
	Testing	13839023896	1916	24907408	2988805
ANN	Training	35089435935	1681.726874	17520075.67	9352598.961
	Testing	22189984412	2651.39974	6499608.451	3376646.962
MGGP	Training	19204199283	2107	30070788	8838013
	Testing	11708806606	2500	6572739	2889826
Hybrid MGGP-GRG	Training	18655186649	2082	17645820	8822776
	Testing	11337677713	2496	6492404	2881054