Short-Time Prediction of Traffic Flow Based on PSO Optimized SVM

To solve the parameter selection problem of the support vector machine(SVM) prediction model, the particle swarm optimization(PSO) algorithm is introduced. The swarm is used to select corresponding learning parameters to achieve optimal PSO-SVM prediction model. Through examples of simulation experiment, the results show PSO-SVM based prediction is superior to prediction with neural network. It overcomes overstudy by neural network training and avoid local optimum solution, to provide better generalization ability. Thus, it is also better than SVM prediction model and offers solution for the parameter selection problem for effective traffic flow prediction.