Abstract
Road Traffic Accidents (RTAs) are currently the leading causes of traffic congestion, human death, health problems, environmental pollution, and economic losses. Investigation of the characteristics and patterns of RTAs is one of the high-priority issues in traffic safety analysis. This paper presents our work on mining RTAs using association rule based methods. A case study is conducted using UK traffic accident data from 2005 to 2017. We performed Apriori algorithm on the data set and then explored the rules with high lift and high support respectively. The results show that RTAs have strong correlation with environmental characteristics, speed limit, and location. With the network visualization, we can explain in details the association rules and obtain more understandable insights into the results. The promising outcomes will undoubtedly reduce traffic accident effectively and assist traffic safety department for decision making.
| Original language | English |
|---|---|
| Title of host publication | International Conference on Brain Inspired Cognitive Systems |
| Subtitle of host publication | BICS 2019 - Advances in Brain Inspired Cognitive Systems |
| Place of Publication | Cham, Switzerland |
| Publisher | Springer |
| Pages | 520-529 |
| Number of pages | 10 |
| ISBN (Electronic) | 9783030394318 |
| ISBN (Print) | 9783030394301 |
| DOIs | |
| Publication status | Published - 1 Feb 2020 |
Keywords
- association rules
- data mining
- data visualization
- traffic accident analysis
- patterns
- crashes