Choice of refractive surgery types for myopia assisted by machine learning based on doctors’ surgical selection data

Table 4 Accuracy and Macro_F1 of machine learning models on right eye data with different hyperparameters max_depth and with/without feature selection

	Model	Max_depth
		9		10		11
		ACC	Macro_F1	ACC	Macro_F1	ACC	Macro_F1
Feature selection (Select the top 12 features with importance greater than 1.4%.)	NBM	0.7435	0.4633	0.7435	0.4633	0.7435	0.4633
	DBN	0.7487	0.4824	0.7487	0.4824	0.7487	0.4824
	RF	0.8125	0.7694	0.8229	0.8080	0.8073	0.7527
	AdaBoost	0.7644	0.6475	0.7382	0.6047	0.7330	0.5931
	XGBoost	0.8020	0.7373	0.7917	0.7102	0.7917	0.7084
	BP Neural Network	0.7708	0.6621	0.7708	0.6621	0.7708	0.6621
No feature selection was performed (There are 20 features in total.)	NBM	0.7487	0.4657	0.7487	0.4657	0.7487	0.4657
	DBN	0.7435	0.4262	0.7435	0.4262	0.7435	0.4262
	RF	0.8177	0.7998	0.8021	0.7427	0.8073	0.7527
	AdaBoost	0.7435	0.6411	0.7592	0.6428	0.7487	0.6337
	XGBoost	0.7969	0.7182	0.8073	0.7426	0.7969	0.7203
	BP Neural Network	0.7760	0.6818	0.7760	0.6818	0.7760	0.6818

ISSN: 1472-6947