Research on Methods of Short-Term Traffic Forecasting Based on Support Vector Regression
Gao Yong-liang · Journal of Beijing Jiaotong University · 2006
The paper proposes short-term traffic forecasting model based on support vector regression.First,the traffic volumes,occupancy-rate,average velocity at several preceding periods of time and upstream and downstream collected by RTMS are considered as input,traffic volumes at current period of time are considered as output.Second,the support vector regression is trained after selecting a kernel function.Finally,the traffic volumes being forecasted at several periods of time in the future are available by inputting the traffic volumes,occupancy-rate and average velocity necessary to the trained support vector regression.The paper also uses the real time data of certain urban road to test the efficiency of the proposed model and the result is satisfied.