Urban Traffic Travel Time Short-Term Prediction Model Based on Spatio-Temporal Feature Extraction

Table 4

XGBoost model input and output.

Symbol	The values of features									No.
Minute	0	2	4	6	8	10	12	14	…	f0
Hour	8	8	8	8	8	8	8	8	…	f1
Dayofweek	2	2	2	2	2	2	2	2	…	f2
Vacation	0	0	0	0	0	0	0	0	…	f3
in_degree	1	1	1	1	1	1	1	1	…	f4
out_degree	1	1	1	1	1	1	1	1	…	f5
Length	48	48	48	48	48	48	48	48	…	f6
Width	3	3	3	3	3	3	3	3	…	f7
Area	144	144	144	144	144	144	144	144	…	f8
Link_ID	43	43	43	43	43	43	43	43	…	f9
T1	1.931	1.931	1.931	1.931	1.931	1.931	1.931	1.931	…	f10
closeness_centrality	0.053	0.053	0.053	0.053	0.053	0.053	0.053	0.053	…	f11
PageRank	0.004	0.004	0.004	0.004	0.004	0.004	0.004	0.004	…	f12
infomap	5	5	5	5	5	5	5	5	…	f13
TTt−1	2.197	1.903	1.911	1.918	1.921	1.918	1.906	1.899	…	f14
TTt−2	1.995	2.197	1.903	1.911	1.918	1.921	1.918	1.906	…	f15
TTt−3	2.197	1.995	2.197	1.903	1.911	1.918	1.921	1.918	…	f16
TTt−4	2.197	2.197	1.995	2.197	1.903	1.911	1.918	1.921	…	f17
TTt−5	2.140	2.197	2.197	1.995	2.197	1.903	1.911	1.918	…	f18
Average_speed	6.000	8.403	8.329	8.262	8.231	8.264	8.378	8.447	…	f19
out_link1_lag1	2.360	2.379	2.360	2.370	2.476	2.595	2.803	2.687	…	f20
out_link2_lag1	−999	−999	−999	−999	−999	−999	−999	−999	…	f21
TTt	2.042	1.932	1.949	1.948	1.948	1.950	1.945	1.937	…	…