基于变分贝叶斯的城市交通拥堵预测模型  

Urban Traffic Congestion Prediction Model Based on Variational Bayesian Model

作　　者：范东霞 FAN Dongxia(Jinzhong College of Information,Jinzhong,Shanxi 030800,China)

机构地区：[1]晋中信息学院,山西晋中030800

出　　处：《自动化应用》2025年第6期88-90,共3页Automation Application

基　　金：山西省2024年度教育改革创新项目“数字化背景下基于APOS理论的教学实践研究——以概率论与数理统计为例”(J20241800)。

摘　　要：旨在研究变分贝叶斯方法的原理以及该方法在交通拥堵预测中的应用效果。通过构建一种基于变分贝叶斯的交通拥堵预测模型,并将其与传统的贝叶斯模型进行对比,发现变分贝叶斯模型在预测精度方面表现出显著的优势。研究结果表明,变分贝叶斯方法能够更有效地捕捉交通数据的复杂性和不确定性,从而生成更为准确的预测结果。此外,还分析了影响模型性能的因素,并探讨了未来的研究方向。This paper aims to explore the application effect of variational Bayesian method in traffic congestion prediction.By constructing a traffic congestion prediction model based on variational Bayes and comparing it with the traditional Bayesian model,we found that the variational Bayesian model shows significant advantages in prediction accuracy.The research results show that variational Bayesian methods can more effectively capture the complexity and uncertainty of traffic data,thereby generating more accurate prediction results.In addition,this paper also analyzes the factors affecting model performance and discusses the directions for future research.

关 键 词：变分贝叶斯模型 交通数据 交通拥堵 预测模型 变分分布 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]