A traffic flow forecasting method based on the GA-SVR

This paper uses support vector regression (SVR) to predict short-term traffic flow, and studies the feasibility of SVR in flow prediction. The short-time has many influencing factors, which are characterized by nonlinearity, randomness periodicity. Therefore, algorithm advantages dealing with such problems. In order improve prediction accuracy SVR, this genetic (GA) optimize other parameters obtain global optimal solution. used construct model. selects data Jiangxi Provincial Transportation Department database verify effectiveness proposed