Ecosense: a revolution in urban air quality forecasting for smart cities

Chatterjee, Kalyan; Thara, Machakanti Navya; Reddy, Mandadi Sriya; Selvamuthukumaran, N.; Priyadharshini, M.; Reddy, Tummala Abhinav Vardhan; Chakraborty, Somenath; Mallik, Saurav; Shah, Mohd Asif; Li, Aimin

doi:10.1186/s13104-025-07099-1

BMC Research Notes

Table 5 Comparison of Recall between SVR [31], MLP [32], RAQP [34], Vlachogianni (Vlacho) [4], LSTM [35], BLSTM [36], SLSTM [37], and the proposed BlaSt Models

From: Ecosense: a revolution in urban air quality forecasting for smart cities

APCs	Method	1hr	2hr	3hr	4hr	5hr	6hr	7hr	8hr	9hr	10hr	11hr	12hr
\(PM_{2.5}\)	SVR	0.942	0.888	0.849	0.788	0.759	0.712	0.710	0.700	0.692	0.680	0.678	0.662
	MLP	0.958	0.911	0.899	0.856	0.815	0.790	0.738	0.713	0.701	0.700	0.695	0.650
	RAQP	0.923	0.915	0.899	0.895	0.875	0.858	0.820	0.799	0.765	0.725	0.710	0.700
	Vlacho	0.865	0.840	0.825	0.788	0.760	0.749	0.715	0.706	0.681	0.655	0.615	0.610
	LSTM	0.951	0.950	0.930	0.889	0.855	0.832	0.820	0.799	0.778	0.740	0.720	0.711
	BLSTM	0.969	0.943	0.911	0.880	0.851	0.833	0.815	0.789	0.760	0.732	0.723	0.712
	SLSTM	0.975	0.954	0.943	0.922	0.890	0.870	0.850	0.812	0.783	0.775	0.765	0.758
	BlaSt	0.995	0.980	0.930	0.910	0.910	0.900	0.890	0.880	0.865	0.849	0.835	0.829
\(NO_{2}\)	SVR	0.925	0.913	0.905	0.892	0.890	0.885	0.870	0.860	0.850	0.805	0.799	0.785
	MLP	0.955	0.875	0.853	0.832	0.819	0.795	0.779	0.769	0.739	0.720	0.710	0.701
	RAQP	0.969	0.960	0.930	0.920	0.895	0.870	0.860	0.848	0.795	0.785	0.768	0.720
	Vlacho	0.895	0.868	0.840	0.835	0.823	0.810	0.805	0.797	0.769	0.740	0.739	0.723
	LSTM	0.982	0.970	0.940	0.910	0.880	0.860	0.840	0.795	0.775	0.765	0.745	0.725
	BLSTM	0.973	0.960	0.945	0.910	0.890	0.860	0.840	0.815	0.790	0.775	0.750	0.730
	SLSTM	0.991	0.970	0.965	0.930	0.890	0.840	0.835	0.825	0.790	0.775	0.750	0.735
	BlaSt	0.993	0.991	0.895	0.890	0.892	0.867	0.865	0.840	0.842	0.830	0.829	0.825
CO	SVR	0.930	0.895	0.885	0.865	0.840	0.810	0.780	0.770	0.765	0.695	0.689	0.681
	MLP	0.960	0.945	0.942	0.880	0.860	0.830	0.801	0.780	0.750	0.740	0.730	0.722
	RAQP	0.960	0.940	0.911	0.890	0.873	0.853	0.843	0.832	0.810	0.780	0.765	0.740
	Vlacho	0.872	0.859	0.858	0.849	0.843	0.840	0.832	0.811	0.805	0.801	0.790	0.779
	LSTM	0.972	0.940	0.915	0.880	0.851	0.835	0.790	0.765	0.740	0.735	0.720	0.710
	BLSTM	0.973	0.960	0.945	0.900	0.890	0.860	0.820	0.811	0.795	0.775	0.740	0.708
	SLSTM	0.980	0.950	0.934	0.900	0.860	0.840	0.820	0.790	0.760	0.730	0.720	0.707
	BlaSt	0.990	0.960	0.920	0.895	0.890	0.880	0.870	0.860	0.850	0.840	0.830	0.820

Bold values indicate the efficiency as well as the superior performance of our proposed model

Back to article page

ISSN: 1756-0500

Contact us

General enquiries: journalsubmissions@springernature.com