Predicting Freeway Traffic Crash Severity Using XGBoost-Bayesian Network Model with Consideration of Features Interaction

Table 1

Summary of the attribute variables of the raw dataset.

Order	Attribute name	Order	Attribute name
1	Administrative division of province	19	Scene form
2	Prefectural administrative divisions	20	Weather
3	Road no.	21	Visibility
4	Road name	22	Pavement condition
5	Milepost	23	Surface condition
6	Meters	24	Traffic control mode
7	Week	25	Light condition
8	Month	26	Roadway type
9	Year	27	Administrative class
10	Occurred date	28	Geography
11	Occurred hour	29	Road alignment
12	Total deaths in 24 H	30	Road section type
13	Total injuries in 24 H	31	Physical road isolation
14	Location	32	Pavement structure
15	Direct property loss	33	Type of central isolation facility
16	Crash type	34	Type of roadside protection facilities
17	Crash cause	35	Number of involved
18	Crash form