摘要
为更加准确地描述道路交通事故发生的规律,结合粒子群算法和支持向量机理论,提出一种粒子群算法(PSO)优化支持向量机(SVM)的道路交通事故预测模型,并用相应的数据进行仿真研究。结果表明,基于粒子群算法优化支持向量机模型(PSO-SVM)的预测精度更高,能较好地契合道路交通事故的变化趋势。
In order to describe the rule of traffic accident more accurately, it introduces a new method of linking up support vector machine and particle swarm algorithm（PSO-SVM）. And a satisfactory result is accomplished. The experimental results indicate that the PSO-SVM has greater accuracy than traditional SVM, and complies with the traffic accident trend.
出处
《交通科技与经济》
2013年第6期48-50,共3页
Technology & Economy in Areas of Communications
关键词
道路交通事故
支持向量机
粒子群算法
预测模型
traffic accident
support vector machine
particle swarm algorithm
forecasting model
