A combined method based on of Principal Component Analysis (PCA), Subtraction Fuzzy Clustering (SFC), and Non-Parametric Regression (NPR) is proposed in view of nonlinearity and uncertainty for short-term forecasting of real-time traffic flow. A highly efficient case database is created from the original traffic volumes after the operations of PCA and SFC. The data-driven method of K-nearest neighbours NPR is used to make the forecasting. An emulation experiment is designed to test the validity of the method. The example results show it is better than normal NPR and meets real-time requirement.
The Combined Short-Term Forecasting Approach to Traffic Flow Based on Non-Parametric Regression
First International Conference on Transportation Engineering ; 2007 ; Southwest Jiaotong University, Chengdu, China
2007-07-09
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
The Combined Short-Term Forecasting Approach to Traffic Flow Based on Non-Parametric Regression
British Library Conference Proceedings | 2007
|Short Term Traffic Flow Forecasting Based on Artificial Neural Network Combined Predictor
British Library Conference Proceedings | 2007
|Use of Local Linear Regression Model for Short-Term Traffic Forecasting
Transportation Research Record | 2003
|