Because of the complicated nonlinear relation between road traffic accident and its influence factors, how to forecast road traffic accident is a problem. Support vector machine(SVM) is the new machine learning tool which is accord with statistic learning theory and overcoming the extra-learning of ANN. In order to improve the forecasting performance, the paper proposes the support vector machine forecasting model of road traffic accident based on statistical data. In the modeling process, the penalty factor and kernel parameter of SVM affects its predict accuracy, therefore the particle swarm optimization arithmetic is utilized to automatically search above parameters. The application of the true example shows the PSO-SVM traffic forecasting model is feasible and precise.
The Evolutionary Support Vector Machine Forecasting Model of Road Traffic Accident
First International Conference on Transportation Engineering ; 2007 ; Southwest Jiaotong University, Chengdu, China
2007-07-09
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
The Evolutionary Support Vector Machine Forecasting Model of Road Traffic Accident
British Library Conference Proceedings | 2007
|An Information Renewal GNN Model for Road Traffic Accident Forecasting
British Library Conference Proceedings | 2009
|British Library Conference Proceedings | 1998
|British Library Online Contents | 1995
|