摘要
道路交通事故的发生原因是多方面的,实际的交通事故存在潜在的规律性,挖掘这种潜在的规律可以更好地改善交通事故现状。提出一种基于智能微粒群的交通事故成因关联挖掘的方法,通过实际的交通事故发生时现场数据对粒子进行编码,使用后代繁殖变异的方法对粒子进行更新,设置支持度与置信度构造优化的目标函数寻找符合的粒子做为挖掘的规则。实验仿真结果证明,本文的算法能够对交通事故的数据进行合理建模,对实际的人工调控交通有很好的指导意义。
Road traffic accident reason is various, the actual traffic accident potentially regularity, and mining the poten- tial law can better improve traffic accident situation. Is proposed based on a particle swarm intelligence of the traffic ac- cident causes associated mining method, through the actual traffic accident happened to particles on data coding, use the method of offspring breeding variation of particle to update and set up support for the confidence and the optimization of the structure of the objective function looking for particle as a mining rules. The experiment the simulation results show that this algorithm can for the traffic accident on the reasonable data modeling, artificial regulation to the actual traffic has a good guide.
出处
《科技通报》
北大核心
2012年第12期149-151,共3页
Bulletin of Science and Technology
基金
河南省科技厅项目(122102210450)
关键词
智能微粒群
关联规则
粒子变异
交通事故
数据挖掘
particle swarm intelligence
association rules
particle variation
traffic accident
data mining
