Modelling on Car-Sharing Serial Prediction Based on Machine Learning and Deep Learning

Table 10

Machine learning for historical data with weather data evaluation results.

	MAE	MSE	RMSE	MAPE	RMSLE
XGBoost	0.07077	0.00944	0.09843	0.10385	0.07077
VAR	1.05296	1.192	1.480543	0.43748	1.12674
SVM	1.08711	1.60693	1.68354	1.79101	1.1896
KNN	1.19562	2.14944	1.7146	1.90031	1.44379
(a) Class A
	MAE	MSE	RMSE	MAPE	RMSLE
XGBoost	0.09003	0.02233	0.1498	0.14143	0.05296
VAR	0.44379	0.36095	0.60079	0.13328	0.09003
SVM	0.72164	1.03892	1.02093	1.14312	0.69041
KNN	0.76347	1.4106	1.18769	1.23748	0.72168
(b) Class B
	MAE	MSE	RMSE	MAPE	RMSLE
XGBoost	0.11705	0.03418	0.18679	0.27058	0.11705
VAR	0.43726	0.34063	0.58364	0.41371	0.43726
SVM	0.80968	0.98371	0.99181	1.12363	0.54982
KNN	0.83909	1.86221	1.36451	1.22429	0.83915
(c) Class C
	MAE	MSE	RMSE	MAPE	RMSLE
XGBoost	0.02945	0.00307	0.05401	0.03076	0.02951
VAR	0.43389	0.34523	0.58756	0.1595	0.43389
SVM	0.54583	0.69474	0.83312	1.16663	0.54594
KNN	0.61329	1.14941	1.07194	1.18616	0.81517
(d) Class D
	MAE	MSE	RMSE	MAPE	RMSLE
XGBoost	0.01711	0.00215	0.04635	0.00462	0.01711
VAR	0.08469	0.08491	0.29139	0.1393	0.08519
SVM	0.13407	0.11862	0.34422	0.52147	0.08471
KNN	0.2732	0.13283	0.36446	1.94885	0.37281
(e) Class E