Highway Traffic Accident Prediction Based on SVR Trained by Genetic Algorithm

Abstract:

The multi researches and experiments show that the future highway traffic accident situation is shown by the highway traffic accident prediction. In the paper, support vector regression trained by genetic algorithm is presented in highway traffic accident prediction. In the method, genetic algorithm is used to train the parameters of support vector regression. Firstly, the regression function of support vector regression algorithm is introduced, and the parameters of support vector regression are optimized by genetic algorithm. The computation results between G-SVR and SVR can indicate that the prediction ability for highway traffic accidents of G-SVR is better than that of SVR absolutely.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 5886-5889
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.5886
• Online since: January 2012
• Authors: Zhen Qi Yang
• Keywords: Genetic Algorithm (GA), Highway Traffic Accident, Prediction Model, Support Vector Regression (SVR)

References

[1] LI Gang, HUANG Tong-yuan, YAN He, HUANG Xi-yu, Grey residual error model of highway traffic accident forecast, Journal of Traffic and Transportation Engineering, 2009, vol. 9, no. 5, pp.88-93.

[2] Chinmaya S. Hardas, Toni L. Doolen, Dean H. Jensen, Development of a genetic algorithm for component placement sequence optimization in printed circuit board assembly, Computers & Industrial Engineering, 2008, vol. 55, no. 1, pp.165-182.

DOI: 10.1016/j.cie.2007.11.016

[3] Pei-Chann Chang, Shih-Hsin Chen, Chin-Yuan Fan, Chien-Lung Chan, Genetic algorithm integrated with artificial chromosomes for multi-objective flowshop scheduling problems, Applied Mathematics and Computation, 2008, vol. 205, no. 2, pp.550-561.

DOI: 10.1016/j.amc.2008.05.027

[4] Haldun Aytug, Cem Saydam, Solving large-scale maximum expected covering location problems by genetic algorithms: A comparative study, European Journal of Operational Research, 2002, vol. 141, no. 3, pp.480-494.

DOI: 10.1016/s0377-2217(01)00260-0

[5] Janne Koljonen, Markus Mannila, Merja Wanne, Testing the performance of a 2D nearest point algorithm with genetic algorithm generated Gaussian distributions, Expert Systems with Applications, 2007, vol. 32, no. 3, pp.879-889.

DOI: 10.1016/j.eswa.2006.01.030

[6] Athanassia Chalimourda, Bernhard Schölkopf, Alex J. Smola, Experimentally optimal ν in support vector regression for different noise models and parameter settings, Neural Networks, 2004, vol. 17, no. 1, pp.127-141.

DOI: 10.1016/s0893-6080(03)00209-0