Table of Contents Author Guidelines Submit a Manuscript
Mathematical Problems in Engineering
Volume 2013 (2013), Article ID 302627, 8 pages
Research Article

A Hybrid Algorithm of Traffic Accident Data Mining on Cause Analysis

1State Key Laboratory of Automobile Dynamic Simulation, Jilin University, Changchun 130022, China
2College of Traffic, Jilin University, Changchun 130022, China
3School of Automobile, Chang’an University, Xi’an 710064, China
4School of Automotive Engineering, Tongji University, Shanghai 201804, China

Received 6 October 2012; Revised 30 December 2012; Accepted 31 December 2012

Academic Editor: Baozhen Yao

Copyright © 2013 Jianfeng Xi et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Road traffic accident databases provide the basis for road traffic accident analysis, the data inside which usually has a radial, multidimensional, and multilayered structure. Traditional data mining algorithms such as association rules, when applied alone, often yield uncertain and unreliable results. An improved association rule algorithm based on Particle Swarm Optimization (PSO) put forward by this paper can be used to analyze the correlation between accident attributes and causes. The new algorithm focuses on characteristics of the hyperstereo structure of road traffic accident data, and the association rules of accident causes can be calculated more accurately and in higher rates. A new concept of Association Entropy is also defined to help compare the importance between different accident attributes. T-test model and Delphi method were deployed to test and verify the accuracy of the improved algorithm, the result of which was a ten times faster speed for random traffic accident data sampling analyses on average. In the paper, the algorithms were tested on a sample database of more than twenty thousand items, each with 56 accident attributes. And the final result proves that the improved algorithm was accurate and stable.