Research Article
Analyzing Accident Injury Severity via an Extreme Gradient Boosting (XGBoost) Model
Table 1
Data list of main variables related to crashes.
| | Variable | Categories | Value | Fatal (n = 169) | Injury (n = 385) |
| | Time of day | Day (7:00–19:00) | 1 | 92 (54.4%) | 263 (68.3%) | | Night (19:00–7:00) | 0 | 77 (45.6%) | 122 (31.7%) | | Day of week | Weekday | 1 | 118 (69.8%) | 294 (76.4%) | | Weekend/holiday | 0 | 51 (30.2%) | 91 (23.6%) | | Rush hour | Rush hour (7:00–9:00, 17:00–20:00) | 1 | 49 (29.0%) | 126 (32.7%) | | Off-peak hour | 0 | 120 (71.0%) | 259 (67.3%) | | Weather | Good (sunny, cloudy) | 1 | 138 (81.7%) | 320 (83.1%) | | Adverse (rainy, snowy, etc.) | 0 | 31 (18.3%) | 65 (16.9%) | | Crash position | Road section | 1 | 110 (65.1%) | 266 (69.1%) | | Intersection | 0 | 59 (34.9%) | 119 (30.9%) | | Crash involvements | Vehicle-bicycle | 1 | 53 (31.4%) | 179 (46.5%) | | Vehicle-pedestrian | 0 | 116 (68.6%) | 206 (53.5%) |
|
|
