A Multiple Kernel Learning Approach for Air Quality Prediction

Table 8

Performance comparison for predicting the future 3 hour’s AQHI in HK.

	Accuracy	Mse	wr	wf	wp
ARIMA	0.525	0.945	0.525	0.525	0.529
RF	0.782	0.275	0.782	0.778	0.781
MLP	0.938	0.09	0.938	0.936	0.935
SVC_linear	0.961	0.04	0.961	0.962	0.963
SVC_rbf	0.937	0.065	0.937	0.937	0.938
SVC_poly	0.954	0.047	0.954	0.955	0.956
SVC_sigmoid	0.267	4.994	0.267	0.113	0.071
LSTM	0.723	0.220	0.723	0.721	0.729
MKSVC	0.974	0.028	0.974	0.975	0.974