基于支持向量回归机的道路交通事故预测模型研究

针对道路交通事故预测具有随机波动性较大、信息量较少和非线性数据序列预测的特点，引入支持向量回归机（SVR），建立基于SVR的道路交通事故预测模型。通过实例计算，证明基于SVR的道路交通事故预测模型具备非线性、所需数据资料较少、建模简单和计算快捷等优点，同时与RBF神经网络预测模型相比，该模型的预测精度高、泛化能力强，更适用于道路交通事故预测。

Prediction Model for Road Traffic Accident Based on Support Vector Regression

In view of road accident prediction has such characteristics as great stochastic fluctuation, little information and nonlinear data series prediction, the Support presented and prediction model for road traffic accidents based on SVR instance proves that the road traffic accident prediction model based on Vector Regression （SVR） is established. The practical SVR has such advantages as nonlinear data, less information, simple modeling and quick computation. Compared with RBF neural network prediction model, SVR prediction model has higher precision and better generalization, therefore it is much well applied to road traffic accident prediction.