Identify Road Clusters with High-Frequency Crashes Using Spatial Data Mining Approach

Zhang, Zhonggui; Ming, Yi; Song, Gangbing

doi:10.3390/app9245282

Cited by 6 publications

(2 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This literature review brings as a background, some of the spatial and statistical methods recently used to analyze traffic accidents. Local spatial statistics methods, the Moran's I and Getis-Ord statistics [10,11] have been popularly used while other studies have adopted methods such as Kernel Density Estimation (KDE) [12,13], Network KDE (NETKDE) [14,15], Quasi-Poisson model [8], Full Bayes hierarchical model [16 -18], Neuro-Fuzzy approach [19,20]and even Neural Networks [21 -25]. Moran's I (MI) is a local spatial statistical method that is used in traffic accident analysis to measure the spatial dependence of accident locations and can also be used to examine the density of their spatial pattern, how dispersed or randomly distributed the cluster patterns are.…”

Section: Literature Reviewmentioning

confidence: 99%

Computing traffic accident high-risk locations using graph analytics

et al. 2022

View full text Add to dashboard Cite

Analysis of the dynamic relationship between traffic accident events and road network topology based on connectivity and graph analytics offers a new approach to identifying, ranking and profiling traffic accident high risk-locations at different levels of space and time granularities. Previous studies on traffic accident hot spots have mostly adopted spatial statistics and Geographic Information Systems (GIS) where spatial point patterns are discovered based only on spatial dependence with no recognition of the temporal dependence of the events. A limitation arises from the fact that the results are either under or over-estimated because of the temporal aggregation of the events to an absolute time point. Furthermore, the existing methods apart from the Network Kernel Density Estimation (NETKDE), consider traffic accident events as events randomly on a 2-D geographic space. However, traffic accident events are network constrained events that happens majorly on the road network space. Therefore, in this paper, we adopt the connectivity of graph on a network space approach that identifies accident high risk-locations based on spacetime-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 for this study. Traffic accident high risk-locations are identified and ranked in space and time using time-dependent degree centrality and PageRank centrality graph metrics respectively through time-incremental graph queries. This study offers urban traffic accident analysts with a new and efficient approach to identify, rank and profile accident-prone areas in space and time at different scales.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Computing traffic accident high-risk locations using graph analytics

et al. 2022

View full text Add to dashboard Cite

show abstract

“…cross or t-intersection, freeways, highways) [9][10][11][12] . A good number of studies have explored and analyzed the spatial clusters of traffic crashes [13][14][15][16] . However, very few studies have analyzed the spatial concentration of crashes involving large-trucks.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Spatial Concentrations of Large-Truck Crashes Using Data Mining Methods

Tahfim¹,

Chen²

2022

IJSSE

View full text Add to dashboard Cite

In recent years, the number of studies on crashes involving large-trucks has increased due to its importance to the economy and the higher chance of fatalities. However, none of the previous studies has given attention to the spatial concentrations of large-truck crashes. Moreover, the literature lacks exploration of granular level land use and urban design factors. The current study used the DBSCAN (Density-Based Spatial Clustering of Application with Noise) method to identify the spatial concentrations of crashes involving large-trucks. Additionally, the study explored housing, population, employment, and road network density attributes along with the crash characteristics, roadway attributes, location type, traffic conditions, driver’s action and behavior, and environmental factors. The association rule analysis was employed to discover the contributory factors that lead to no injury, non-severe and severe injuries at the spatial concentrations of crashes involving large-trucks. The findings indicated that the rear-end collisions involving drunk drivers often lead to severe injuries in large-truck crashes. Non-interstate roads, speed limit from 40 to 80 kilometers per hour, high road network density, medium and high population density are frequent conditions of non-severe injuries. Lastly, collisions between large-trucks and fixed objects, sideswipe same direction collisions, snowy roads, clear weather, medium road network and employment density are likely to facilitate no injury crashes involving large-trucks. Road traffic authorities can use these insights to reduce the frequency and severity of crashes involving large-trucks at their spatial concentrations.

show abstract

Design of Portrait System for Road Safety Based on a Dynamic Density Clustering Algorithm

Cui

2022

Neural Computing for Advanced Applications

View full text Add to dashboard Cite

Identify Road Clusters with High-Frequency Crashes Using Spatial Data Mining Approach

Cited by 6 publications

References 56 publications

Computing traffic accident high-risk locations using graph analytics

Computing traffic accident high-risk locations using graph analytics

Analysis of Spatial Concentrations of Large-Truck Crashes Using Data Mining Methods

Design of Portrait System for Road Safety Based on a Dynamic Density Clustering Algorithm

Contact Info

Product

Resources

About