The invention provides a lifelong learning-based congestion duration optimization method, which comprises the steps of collecting data, extracting influence factors, combining a MapReduce model with a multivariate logarithmic linear regression method to construct a traffic congestion prediction model, and optimizing the traffic congestion prediction model by using a survival analysis method. A MapReduce model and a multivariable linear regression method are combined to construct a prediction model of the traffic congestion duration to improve the prediction precision and efficiency of the traffic congestion duration, and on this basis, the traffic congestion duration prediction model of survival analysis is used to further research and analyze the congestion duration in a traffic network. And a better traffic information service is provided for travelers.
本发明提供一种基于终身学习的拥堵持续时间优化方法,包括采集数据,提取影响因子,将MapReduce模型与多元对数线性回归方法相结合构建交通阻塞预测模型,并运用生存分析方法对交通阻塞预测模型进行了优化,把MapReduce模型与多变量线性回归方法进行结合构建交通拥堵持续时间的预测模型提高交通拥堵持续时间的预测的精度与效率,在此基础上运用生存分析的交通拥堵持续时间预测模型对交通网络中的拥堵持续时间进行更近一步的研究分析,为出行人员提供了更好的交通信息服务。
Congestion duration prediction method based on lifelong learning
基于终身学习的拥堵持续时间预测方法
2023-09-05
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
CONGESTION PREDICTION METHOD AND CONGESTION PREDICTION DEVICE
European Patent Office | 2021
|CONGESTION PREDICTION DEVICE, CONGESTION PREDICTION SYSTEM, AND CONGESTION PREDICTION METHOD
European Patent Office | 2020
|CONGESTION PREDICTION PROGRAM, CONGESTION PREDICTION METHOD AND CONGESTION PREDICTION DEVICE
European Patent Office | 2020
|European Patent Office | 2018
|CONGESTION PREDICTION PROGRAM, CONGESTION PREDICTION APPARATUS, AND CONGESTION PREDICTION METHOD
European Patent Office | 2018
|