The invention provides a method for predicting traffic flow by using dynamic multi-graph fusion, and belongs to the technical field of spatio-temporal data mining. The method comprises the following steps: firstly, constructing three different types of graphs, including a distance graph, a function similar graph and a time pattern similar graph, for representing a context relationship of nodes, constructing a fusion graph by adopting a weight fusion mode by utilizing the three constructed graphs, and adjusting the fusion graph by utilizing a designed dynamic graph convolutional network so as to obtain a fusion graph; a self-adaptive adjacency matrix is further introduced into the dynamic graph convolutional network to deduce hidden dependence between nodes, and the dynamic graph convolutional network is embedded into a recurrent neural network to achieve simultaneous capture of spatial-temporal correlation. According to the method, the defect of predicting by using a single graph is overcome, and the dynamic relationship between the nodes is better modeled.

    本发明提供了一种使用动态多图融合进行交通流量预测的方法,属于时空数据挖掘技术领域。本发明首先构建了三种不同类型的图,包括距离图、功能相似图、以及时间模式相似图,用来表示节点的上下文关系,利用构建的三种图采取权重融合的方式构建融合图,利用设计的动态图卷积网络对融合图进行调整,在动态图卷积网络中还引入了一个自适应的邻接矩阵来推断节点之间的隐藏依赖,将动态图卷积网络嵌入到循环神经网络中实现同时捕获时空相关性。弥补了利用单一图进行预测的不足,并且更好地建模了节点之间的动态关系。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Method for predicting traffic flow by using dynamic multi-graph fusion


    Weitere Titelangaben:

    一种使用动态多图融合进行交通流量预测的方法


    Beteiligte:
    ZHANG QIANG (Autor:in) / WANG PENGFEI (Autor:in) / SHEN BO (Autor:in)

    Erscheinungsdatum :

    2023-10-31


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Traffic flow prediction method based on static and dynamic multi-graph fusion

    YU RUIYUN / WANG CHEN | Europäisches Patentamt | 2023

    Freier Zugriff

    Multi-view fusion space-time dynamic graph convolutional network urban traffic flow prediction method

    YUAN GUAN / ZHAO WENZHU / ZHANG YANMEI et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Dynamic multi-graph convolutional network traffic flow prediction method based on graph transformation

    HU YONGLI / PENG TING / GUO KAN et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    System and method for predicting traffic flow by using graph neural network

    LI SHUANG / QI YUNSONG | Europäisches Patentamt | 2024

    Freier Zugriff

    Traffic flow prediction method based on fusion of space-time adaptive graph learning and dynamic graph convolution

    ZHANG HONG / CHEN LINBIAO / CHEN LINLONG et al. | Europäisches Patentamt | 2024

    Freier Zugriff