Journal of Big Data

Table 12 Results of MDW airport

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

Algorithm		Time difference: 2 h						Time difference: 4 h
Algorithm		Accuracy	Precision	Recall	F1-score	Train (s)	Test (us)	Accuracy	Precision	Recall	F1-score	Train (s)	Test (us)
DT	Normal	0.731	0.741	0.702	0.721	0.073	0.955	0.722	0.750	0.659	0.702	0.045	0.955
	Delayed		0.721	0.759	0.740				0.701	0.784	0.740
RF	Normal	0.762	0.748	0.784	0.766	0.997	17.834	0.766	0.750	0.792	0.771	1.011	17.197
	Delayed		0.777	0.741	0.759				0.784	0.741	0.762
SVM	Normal	0.587	0.588	0.558	0.573	3.716	542.994	0.600	0.596	0.599	0.598	3.715	564.968
	Delayed		0.586	0.615	0.600				0.604	0.601	0.602
KNN	Normal	0.642	0.645	0.620	0.632	0.007	73.248	0.646	0.649	0.623	0.636	0.009	29.936
	Delayed		0.640	0.664	0.652				0.643	0.668	0.656
LR	Normal	0.571	0.570	0.548	0.558	0.079	0.637	0.581	0.579	0.567	0.573	0.076	0.955
	Delayed	0.594	0.582				0.583	0.595	0.589				0.578
XGB	Normal	0.716	0.718	0.703	0.710	0.178	2.866	0.715	0.717	0.702	0.709	0.142	3.185
	Delayed		0.714	0.728	0.721				0.713	0.728	0.720
LSTM	Normal	0.785	0.755	0.849	0.799	404.2	2.548	0.756	0.744	0.786	0.765	411.7	2.548
	Delayed		0.824	0.719	0.768				0.769	0.725	0.746

Algorithm		Time difference: 8 h						Time difference: 16 h
Algorithm		Accuracy	Precision	Recall	F1-score	Train (s)	Test (us)	Accuracy	Precision	Recall	F1-score	Train (s)	Test (us)
DT	Normal	0.716	0.731	0.677	0.703	0.042	0.955	0.726	0.752	0.667	0.707	0.046	1.274
	Delayed		0.704	0.755	0.729				0.705	0.783	0.742
RF	Normal	0.767	0.75	0.775	0.767	1.070	17.834	0.774	0.752	0.812	0.781	1.331	17.516
	Delayed		0.774	0.758	0.766				0.800	0.737	0.767
SVM	Normal	0.613	0.619	0.574	0.595	4.172	612.102	0.615	0.619	0.584	0.601	3.862	637.898
	Delayed		0.609	0.652	0.630				0.612	0.646	0.629
KNN	Normal	0.643	0.657	0.585	0.619	0.006	27.389	0.667	0.677	0.629	0.652	0.006	29.618
	Delayed		0.632	0.701	0.664				0.659	0.705	0.681
LR	Normal	0.578	0.577	0.554	0.565	0.071	1.911	0.605	0.605	0.587	0.596	0.080	0.637
	Delayed		0.572	0.601	0.589				0.606	0.624	0.615
XGB	Normal	0.714	0.723	0.687	0.704	0.140	3.185	0.726	0.731	0.706	0.719	0.150	3.185
	Delayed		0.707	0.741	0.723				0.721	0.745	0.733
LSTM	Normal	0.741	0.736	0.759	0.747	409.3	2.548	0.718	0.701	0.772	0.735	410.8	2.548
	Delayed		0.746	0.723	0.735				0.741	0.664	0.700

Algorithm		Time difference: 24 h						Time difference: 48 h
Algorithm		Accuracy	Precision	Recall	F1-score	Train (s)	Test (us)	Accuracy	Precision	Recall	F1-score	Train (s)	Test (us)
DT	Normal	0.703	0.747	0.683	0.713	0.045	0.955	0.727	0.744	0.684	0.713	0.049	0.955
	Delayed		0.712	0.773	0.741				0.712	0.769	0.740
RF	Normal	0.773	0.762	0.787	0.774	1.081	18.153	0.772	0.759	0.790	0.774	1.128	16.879
	Delayed		0.784	0.758	0.771				0.785	0.754	0.769
SVM	Normal	0.600	0.611	0.528	0.566	3.974	595.223	0.614	0.619	0.577	0.597	4.377	630.573
	Delayed		0.591	0.670	0.628				0.610	0.652	0.630
KNN	Normal	0.663	0.667	0.641	0.654	0.006	27.389	0.664	0.679	0.609	0.642	0.008	26.433
	Delayed		0.660	0.685	0.672				0.651	0.717	0.683
LR	Normal	0.597	0.601	0.558	0.579	0.084	0.955	0.596	0.598	0.563	0.580	0.096	0.637
	Delayed		0.594	0.635	0.614				0.594	0.628	0.611
XGB	Normal	0.731	0.737	0.710	0.723	0.134	3.503	0.725	0.732	0.703	0.717	0.152	3.503
	Delayed		0.725	0.751	0.738				0.719	0.747	0.733
LSTM	Normal	0.714	0.697	0.768	0.731	415.2	2.548	0.712	0.719	0.702	0.710	416.1	2.548
	Delayed		0.736	0.659	0.696				0.705	0.722	0.713

Bold values indicate the greatest results

Back to article page