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

Journal of Big Data

Table 13 Results of ICN 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	0.688	0.688	0.687	0.681	0.686	0.678	0.674	0.678	0.677	0.675	0.680	0.675	0.661
RF	0.750	0.749	0.748	0.735	0.746	0.738	0.741	0.744	0.751	0.743	0.743	0.751	0.739
SVM	0.697	0.651	0.686	0.646	0.653	0.654	0.652	0.641	0.660	0.645	0.641	0.638	0.640
KNN	0.659	0.641	0.654	0.652	0.652	0.646	0.646	0.662	0.651	0.641	0.644	0.635	0.638
LR	0.669	0.595	0.659	0.583	0.589	0.592	0.582	0.598	0.571	0.573	0.547	0.537	0.548
XGB	0.734	0.721	0.733	0.707	0.715	0.726	0.718	0.727	0.717	0.714	0.712	0.722	0.712
LSTM	0.622	0.644	0.609	0.609	0.624	0.602	0.579	0.587	0.573	0.580	0.571	0.562	0.539

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	0.686	0.677	0.687	0.675	0.680	0.689	0.684	0.682	0.685	0.680	0.687	0.011	0.318
RF	0.737	0.743	0.745	0.747	0.739	0.739	0.756	0.745	0.738	0.746	0.743	3.613	20.701
SVM	0.643	0.628	0.641	0.636	0.651	0.651	0.628	0.640	0.641	0.640	0.647	6.295	970.382
KNN	0.643	0.630	0.649	0.653	0.629	0.629	0.641	0.655	0.640	0.657	0.641	0.004	47.134
LR	0.521	0.568	0.525	0.554	0.527	0.527	0.528	0.546	0.554	0.544	0.547	0.074	0.637
XGB	0.715	0.710	0.714	0.714	0.715	0.715	0.716	0.713	0.706	0.716	0.705	0.211	1.592
LSTM	0.558	0.563	0.540	0.564	0.561	0.561	0.571	0.571	0.548	0.545	0.580	468.7	184.713