Table 3.

The optimized hyperparameter values and the corresponding ranges for an optimal parameter grid search.

Model	Hyperparameters	Range	Optimal Parameter Value
CBR	depth	(10, 18)	16
	learning_rate	(0.01, 0.1)	0.1
	iterations	(30, 50, 100)	100
RFR	n_estimators	(30, 50, 100, 150)	150
	min_samples_split	(0.6, 2, 3)	2
	max_features	(‘sqrt’, ‘log2’, 0.6,1.0)	log2
XgBR	learning_rate	(0.05, 0.20)	0.20
	n_estimators	(500, 1500, 2000)	2000
KNNR	n_neighbors	(3, 10, 1)	6
	weights	(‘uniform’, ‘distance’)	‘distance’
MLPR	max_iter	(1000, 2000)	1000
	hidden_layer_sizes	((150,100,50),(200,150,100))	(200,150,100)
	alpha	(0.05, 0.001)	0.05
DTR	criterion	('squared_error', 'friedman_mse', 'absolute_error', 'poisson')	'squared_error'
	max_depth	(5, 20)	17