The invention discloses a traffic flow prediction method and system under data missing based on tensor decomposition, and belongs to the technical field of intelligent traffic, and the method comprises the steps: obtaining a spatial adjacency matrix of a traffic road network and traffic flow containing missing values, and inputting the spatial adjacency matrix and the traffic flow into a pre-trained tensor graph convolutional network model for prediction; comprising the following steps: constructing a time adjacency matrix according to the space adjacency matrix and traffic flow containing missing values; the traffic flow containing the missing value is converted into a tensor, tensor decomposition is carried out, and a feature tensor is obtained through decomposition; inputting the spatial adjacency matrix, the temporal adjacency matrix and the feature tensor into an image convolution layer, and outputting a prediction feature; and performing linear conversion on the prediction features, and outputting a model prediction result to obtain predicted traffic flow. The Tucker decomposition and the graph convolutional neural network are combined to construct the traffic flow prediction model, so that the model can predict the traffic flow more accurately under the condition of data missing.

    本发明公开了一种基于张量分解的数据缺失下交通流量预测方法及系统,属于智能交通技术领域,获取交通路网的空间邻接矩阵和包含缺失值的交通流量输入至预先训练好的张量图卷积网络模型进行预测,包括:根据所述空间邻接矩阵和包含缺失值的交通流量构建时间邻接矩阵;将包含缺失值的交通流量转化为张量并进行张量分解,分解得到特征张量;将所述空间邻接矩阵、时间邻接矩阵和特征张量输入至图卷积层中,输出预测特征;将所述预测特征进行线性转换,输出模型预测结果,得到预测的交通流量。将Tucker分解与图卷积神经网络结合构建交通流量预测模型,在数据缺失情况下,使得模型可以更准确地预测交通流量。


    Access

    Download


    Export, share and cite



    Title :

    Traffic flow prediction method and system under data loss based on tensor decomposition


    Additional title:

    一种基于张量分解的数据缺失下交通流量预测方法及系统


    Contributors:
    WANG CHAO (author) / ZHANG YUNFAN (author) / WU WEILING (author) / WANG YITONG (author) / WANG CHENGCHENG (author) / XU ZHAO (author) / XU RUN (author) / TANG JINJUN (author) / DUAN YIXIN (author)

    Publication date :

    2024-08-30


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Traffic flow data completion method and system based on tensor chain decomposition

    ZHU CHENLU / YIN PU / YANG ZECAN et al. | European Patent Office | 2023

    Free access

    Traffic flow prediction method based on tensor decomposition and reconstruction fusion image

    LI QIN / YANG XUAN / ZHENG ZUOCAI et al. | European Patent Office | 2024

    Free access

    Robust Tensor Ring Decomposition for Urban Traffic Data Imputation

    Yu, Linfang / Guan, Chenyu / Wang, Hao et al. | IEEE | 2025


    Traffic speed data filling method based on space-time regularization tensor decomposition

    DONG XIANGJUN / XIE HAOJIE | European Patent Office | 2024

    Free access

    A tensor-based K-nearest neighbors method for traffic speed prediction under data missing

    Zheng, Liang / Huang, Huimin / Zhu, Chuang et al. | Taylor & Francis Verlag | 2020