Table 5 Evaluation metrics of ensemble machine learning algorithms using ROS data under default and tuned hyperparameter settings.

From: Comparative performance of bagging and boosting ensemble models for predicting lumpy skin disease with multiclass-imbalanced data

Ensemble model	Stage	Class	Accuracy		Precision		Recall		F1-score		AUC
Ensemble model	Stage	Class	ROS	ROS + Tuning	ROS	ROS + Tuning	ROS	ROS + Tuning	ROS	ROS + Tuning	ROS	ROS + Tuning
DT	Training	Healthy	70.29%	70.29%	0.64	0.64	0.87	0.87	0.74	0.74	0.83	0.83
		Diseased			0.64	0.64	0.63	0.63	0.64	0.64
		Dead			0.82	0.82	0.66	0.66	0.74	0.74
	Test	Healthy	66.35%	66.35%	0.66	0.66	0.95	0.95	0.78	0.78	0.82	0.82
		Diseased			0.70	0.70	0.52	0.52	0.60	0.60
		Dead			0.44	0.44	0.12	0.12	0.19	0.19
RF	Training	Healthy	87.75%	88.8%	0.94	0.95	0.82	0.84	0.87	0.89	0.95	0.95
		Diseased			0.86	0.84	0.81	0.85	0.84	0.85
		Dead			0.85	0.88	1.00	0.97	0.92	0.92
	Test	Healthy	80.29%	82%	0.95	0.96	0.80	0.80	0.87	0.87	0.90	0.93
		Diseased			0.64	0.65	0.79	0.86	0.70	0.74
		Dead			0.44	0.57	0.89	0.89	0.59	0.70
AdaBoost	Training	Healthy	83%	81%	0.92	0.81	0.76	0.79	0.83	0.80	0.90	0.97
		Diseased			0.77	0.80	0.78	0.71	0.78	0.75
		Dead			0.83	0.82	0.96	0.93	0.89	0.87
	Test	Healthy	73.56%	74.5%	0.92	0.91	0.73	0.78	0.81	0.84	0.84	0.92
		Diseased			0.55	0.57	0.75	0.68	0.63	0.62
		Dead			0.39	0.28	0.78	0.56	0.52	0.37
GBoost	Training	Healthy	64.6%	88%	0.82	0.92	0.67	0.84	0.74	0.88	0.96	0.97
		Diseased			0.55	0.88	0.66	0.80	0.60	0.84
		Dead			0.61	0.85	0.61	1.00	0.61	0.92
	Test	Healthy	67%	82%	0.89	0.93	0.71	0.83	0.79	0.87	0.92	0.97
		Diseased			0.57	0.65	0.64	0.73	0.61	0.69
		Dead			0.19	0.44	0.33	0.89	0.15	0.59
XGBoost	Training	Healthy	88%	88.8%	0.93	0.94	0.82	0.84	0.87	0.89	0.97	0.97
		Diseased			0.85	0.85	0.84	0.85	0.84	0.85
		Dead			0.87	0.88	0.97	0.97	0.92	0.92
	Test	Healthy	72.28%	81.25%	0.88	0.93	0.77	0.82	0.82	0.87	0.93	0.93
		Diseased			0.63	0.65	0.64	0.79	0.64	0.71
		Dead			0.26	0.57	0.65	0.89	0.38	0.70

Back to article page

Quick links

Search

Quick links