TY  - JOUR
PY  - 2022//
TI  - Computing traffic accident high-risk locations using graph analytics
JO  - Spatial information research
A1  - Maduako, Iyke
A1  - Ebinne, Elijah
A1  - Uzodinma, Victus
A1  - Okolie, Chukwuma
A1  - Chiemelu, Emmanuel
SP  - ePub
EP  - ePub
VL  - ePub
IS  - ePub
N2  - Several studies on traffic accident hot spots have adopted spatial statistics and Geographic Information Systems where spatial point patterns are modelled solely on spatial dependence without consideration of the temporal dependence of the events. This could lead to under-estimation or over-estimation of results because of the temporal aggregation of the events to an absolute time point. Furthermore, traffic accidents are usually considered as events occurring randomly in two-dimensional geographic space. However, traffic accidents are network-constrained events. In this study, we adopt the connectivity of graph on a network space approach that identifies accident high risk-locations based on space-time-varying connectivity between traffic accident events and the road network geometry. A simple but extensible traffic accident space time-varying graph (STVG) model is developed and implemented using traffic accident data from 2010 to 2015 for Brevard County in Florida, United States. Traffic accident high risk-locations were identified and ranked in space and time using time-dependent degree centrality and PageRank centrality graph metrics respectively through time-incremental graph queries. In the analysis, traffic accident patterns were discovered based on network graph analytics. Our findings offer a new and efficient approach for identifying, ranking and profiling accident-prone areas in space and time at different scales.<p />  <p>Language: en</p>
LA  - en
SN  - 2366-3286
UR  - http://dx.doi.org/10.1007/s41324-022-00448-3
ID  - ref1
ER  -