Hybrid GA Based Online Support Vector Machine Model for Short-Term Traffic Flow Forecasting
In this paper, a hybrid genetic algorithm (GA) based online support vector machine (OSVM) prediction model for short-term traffic flow forecasting is proposed, according to the data collected sequentially by the probe vehicle or the loop detectors, which can update the forecasting function in real time via online learning way, and the parameters used in the OSVM were optimized by GA. As a result, it is fitter for the real engineering application. The GA based OSVM model was tested by using the I-880 database, the result shows that this model is superior to the back- propagation neural network (BPNN) model.
Haowei Su Shu Yu
College of Computer Science and Engineering, South China University of Technology, Guangzhou 510640, P. R. China
国际会议
7th International Symposium,APPT 2007(第7届高级并行处理技术大会)
广州
英文
743-752
2007-11-22（万方平台首次上网日期，不代表论文的发表时间）