Nonparametric regression is an important method for short-term traffic flow forecasting, but the traditional nonparametric regression method needs a large storage space and slow query speed when the data are large and the dimension is high. In this paper, an improved nonparametric regression traffic flow forecasting algorithm is proposed. Subtraction fuzzy clustering method is used to cluster historical data to reduce the amount of data in the pattern database. Principal component analysis (PCA) is used to reduce the dimension of the pattern to overcome the problems of slow matching speed and interference of irrelevant dimension caused by the high dimension of the pattern. The support vector machine method is used to estimate the value of the final predicted variables by searching the patterns. The operation efficiency and prediction accuracy of the algorithm are improved. An online simulation-based test shows that the algorithm exhibits better efficiency and accuracy compared with traditional methods.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Nonparametric Regression Algorithm for Short-Term Traffic Flow Forecasting Based on Data Reduction and Support Vector Machine


    Contributors:
    Wu, Jin-wu (author) / Zhang, Hai-feng (author) / Ran, Xu-dong (author)


    Publication date :

    2020-09-01


    Size :

    82020-01-01 pages




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    Unknown





    Short-Term Traffic Flow Prediction Based on Lagrange Support Vector Regression

    Liu, Yanzhong / Li, Xuhong / Shao, Xiaojian | ASCE | 2007


    Short-Term Traffic Flow Prediction Based on Lagrange Support Vector Regression

    Liu, Y. / Li, X. / Shao, X. et al. | British Library Conference Proceedings | 2007


    Short-Term Traffic Flow Prediction Algorithm for Expressway Based on Long Short-Term Memory and Support Vector Regression

    Guo, Lan Ying / Chang, Hui / Cheng, Xin et al. | British Library Conference Proceedings | 2020