面向智能交通的大数据实时分析技术

Big Data Real-time Analysis Technology for Intelligent Transportation

文中研究了基于改进马尔可夫链的交通流量预测方法,旨在利用大数据实时分析技术实现智能交通。首先,指出了交通流量预测在智能交通系统中的重要性和面临的挑战。随后,详细阐述了高阶多元马尔可夫链模型的理论基础,包括马尔可夫链的概念和高阶模型的扩展,从而提出了改进马尔可夫链的交通流量预测方法的总体架构。在此基础上,利用METR-LA数据集进行实验分析,对数据进行预处理和模型训练,并进行模型评估和数据分析。实验结果表明,改进后的马尔可夫链模型在交通流量预测方面具有一定的准确性和实用性。

This paper studies the traffic flow prediction method based on improved Markov chain,aiming to use big data real-time analysis technology implementation of intelligent transportation.First,the importance and challenges of traffic flow prediction in intelligent transportation system are pointed out.Then,the theoretical basis of the higher-order multivariate Markov chain model is elaborated,including the concept of Markov chain and the extension of the higher-order model,thus the overall architecture of the improved Markov chain traffic flow prediction method is proposed.On this basis,the METR-LA dataset is used for experimental analysis,the data is preprocessed and model trained,and model evaluation and data analytics are carried out.The experimental results show that the improved Markov chain model has certain accuracy and practicality in traffic flow prediction.