Short-term Traffic Flow Forecasting based on SVR with Improved Artificial Fish Swarm Algorithm

摘要

Real-time and accurate traffic flow forecasting is one of the key contents of Intelligent Transportation System. For the disadvantage of parameter selection of Support Vector Regression (SVR), an improved artificial fish swarm(IAFS) algorithm using the adaptive search mechanism was applied to optimize SVR. This method aimed at improving the prediction accuracy and extensibility of short-term traffic flow forecasting. Then a short-term traffic flow forecasting model based on IAFS-SVR was proposed. The results show that the proposed method has better prediction performance, and is suitable for short-term traffic flow forecasting.