Prediction of flight departure delays caused by weather conditions adopting data-driven approaches

Journal of Big Data

Table 15 Results of MDW airport (accuracy from 1 to 24 h)

Algorithm	1 h	2 h	3 h	4 h	5 h	6 h	7 h	8 h	9 h	10 h	11 h	12 h	13 h
DT	0.755	0.731	0.743	0.722	0.763	0.740	0.738	0.716	0.745	0.726	0.731	0.743	0.720
RF	0.811	0.762	0.787	0.766	0.791	0.783	0.803	0.767	0.789	0.767	0.779	0.783	0.783
SVM	0.735	0.587	0.716	0.600	0.690	0.651	0.664	0.613	0.627	0.616	0.642	0.630	0.606
KNN	0.728	0.642	0.727	0.646	0.710	0.675	0.698	0.643	0.661	0.692	0.688	0.674	0.667
LR	0.698	0.571	0.655	0.581	0.641	0.612	0.607	0.578	0.566	0.595	0.590	0.589	0.575
XGB	0.797	0.716	0.769	0.715	0.776	0.741	0.750	0.714	0.731	0.715	0.755	0.743	0.745
LSTM	0.785	0.785	0.769	0.756	0.773	0.746	0.744	0.741	0.750	0.749	0.737	0.727	0.734

Algorithm	14 h	15 h	16 h	17 h	18 h	19 h	20 h	21 h	22 h	23 h	24 h	Avg train (s)	Avg test (us)
DT	0.717	0.731	0.726	0.728	0.726	0.693	0.719	0.730	0.731	0.709	0.728	0.052	0.899
RF	0.777	0.776	0.774	0.773	0.765	0.756	0.766	0.790	0.782	0.756	0.773	1.243	23.820
SVM	0.584	0.615	0.615	0.596	0.608	0.572	0.600	0.622	0.563	0.580	0.600	4.500	884.045
KNN	0.661	0.663	0.667	0.659	0.665	0.653	0.646	0.658	0.670	0.652	0.663	0.008	48.090
LR	0.546	0.584	0.605	0.574	0.580	0.587	0.581	0.604	0.577	0.559	0.597	0.089	0.899
XGB	0.732	0.723	0.726	0.714	0.717	0.707	0.725	0.741	0.728	0.700	0.731	0.177	4.045
LSTM	0.716	0.728	0.718	0.724	0.713	0.729	0.723	0.715	0.709	0.727	0.714	391.6	3.596