Providing drivers accurate and efficient traffic prediction is very important for their planning routes and reducing traffic congestion. For providing traffic prediction, monitors are deployed in roads to collect such traffic information as vehicle speed or the number of passing vehicles. Existing traffic prediction methods mainly use collected historical traffic information to make prediction. However, they have two drawbacks: (1) they didn’t pay much attention to data volume and consumed computation time; (2) they considered only the traffic information to make prediction. However, weather is a significant factor that affects the traffic condition. In this paper, we propose a weather-based traffic prediction system (WTPS) which uses both historical traffic information and weather information to make prediction. Experiments demonstrate the prediction accuracy of WTPS is superior to that of existing methods without considering weather. WTPS is also scalable for handling huge traffic volume because it is designed according to big data techniques.
A Weather-Based Traffic Prediction System Using Big Data Techniques
26.09.2022
957027 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Weather Adaptive Traffic Prediction Using Neurowavelet Models
Online Contents | 2013
|Traffic Volume Prediction Based on Weather Parameters
IEEE | 2024
|Calibrating a Real-Time Traffic Crash-Prediction Model Using Archived Weather and ITS Traffic Data
Online Contents | 2006
|