The invention discloses a road traffic flow prediction method based on stacked regression, and the method comprises the steps: obtaining a public data set, carrying out the preprocessing and feature extraction, dividing a feature sequence into a training set and a test set, and enabling the training set and the test set to serve as the input of a subsequent model; a Bi-GRU model and a GBR model are constructed, model parameters corresponding to the two models are set, and construction of the two models is completed; a Bi-GRU model and a GBR model are used as basic model units, a ridge algorithm is used as a meta-model, a stack regression algorithm is used for building a hybrid model, and then the model is subjected to centralized right processing; carrying out model training on the hybrid model and realizing prediction of model data through weight division; and checking a model prediction result and detecting the accuracy of the model prediction result. According to the method, the GBR model and the Bi-GRU model are stacked by using the stacked regression algorithm and using the ridge algorithm as a meta-model, and the mixed model is trained and predicted, so that the accuracy and accuracy of traffic flow prediction can be improved.

    本发明公开了一种基于堆叠回归的道路交通流量预测方法,包括获取公共数据集,并进行预处理以及特征提取,将特征序列划分为训练集以及测试集,作为后续模型的输入;构建Bi‑GRU模型以及GBR模型,设置两个模型对应的模型参数,完成两个模型的构建;将Bi‑GRU模型和GBR模型作为基础模型单元,通过将ridge算法作为元模型,利用堆叠回归算法搭建出混合模型,然后对模型进行集权处理;对混合模型进行模型的训练以及通过权重划分,实现模型数据的预测;将模型预测的结果进行校验并且检测其准确性。本发明利用堆叠回归算法将GBR与Bi‑GRU模型以ridge算法作为元模型堆叠起来,对混合模型进行训练与预测,可以提高了交通流量预测的精确性与准确性。


    Access

    Download


    Export, share and cite



    Title :

    Road traffic flow prediction method based on stacked regression


    Additional title:

    一种基于堆叠回归的道路交通流量预测方法


    Contributors:
    HU YONG (author) / JIN YING (author) / KANG YAO (author) / LI LEYI (author) / XU YIRU (author) / ZHANG SHUIXING (author)

    Publication date :

    2023-11-24


    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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Road traffic flow prediction method

    HOU JIA / CHENG XIAOMING / GUO LICHAO et al. | European Patent Office | 2024

    Free access

    Urban road traffic flow prediction method

    ZHANG HUIZHEN / YAO SHUILIN / FENG WENJUAN et al. | European Patent Office | 2024

    Free access

    Urban Short-Term Traffic Flow Prediction Based on Stacked Autoencoder

    Zhao, Xinran / Gu, Yuanli / Chen, Lun et al. | ASCE | 2019


    Road traffic greenhouse gas emission prediction method based on traffic flow

    DENG TIANMIN / CHEN YUETIAN / DENG TIANQING et al. | European Patent Office | 2024

    Free access

    Road network dynamic traffic distribution method based on traffic flow prediction

    LI XU / ZUO JINPENG / YANG SHIHUAI et al. | European Patent Office | 2024

    Free access