Research On Urban Traffic Trip Portrait Based On Frequent Pattern Mining

As an important part of urban system,urban transportation system is indispensable in the research of urban risk.The prevention and monitoring of urban risks need to consider the characteristics of urban transportation.Mining traffic trajectory data,we can find urban residents’ movement laws and the interactive characteristics of urban system space,and then portray the traffic portrait of the entire city.This article takes the portrayal of urban traffic as a starting point,studies how to efficiently mine travel trajectory data,explores the reasons for the formation of these driving trajectories,and carries out the following work:Firstly,the urban traffic travel portrait method based on frequent pattern mining is proposed.This method is based on the existing label propagation algorithm and frequent pattern mining technology.It takes traffic trajectory data as input and uses subdivided plots as nodes to construct a directed network.It simplifies the processing of trajectory data and effectively mines frequent moving paths and designs support for nodes to identify the best parking location.Secondly,the spatial trajectory networks of online car-hailing and shared bicycles are constructed and analyzed.This article cleans and processes the real GPS trajectory data of online car-hailing and shared bicycles in Tianjin,and builds dynamic trajectory networks of online car-hailing and shared bicycles that change with time according to the geographical area,and analyzes the basic characteristics of the network to explore the trajectory features of the network.Finally,using the constructed urban traffic GPS trajectory networks,the above model is empirically analyzed.Taking Tianjin as a research background,based on the traffic trajectory network,this article excavates the movement paths and frequent trajectory patterns of two types of vehicles,summarizes the periodic laws of path changes,residents’ movement patterns and travel patterns,and gives corresponding explanations.This article also identifies the best parking location points,in order to portray the portrait of Tianjin’s traffic travel,to verify the effectiveness and practicability of the model.In summary,this paper completes the study of urban traffic travel portrait methods based on frequent pattern mining.Using online car-hailing and shared bicycle trajectory data in Tianjin,the spatial trajectory networks are constructed and empirical analysis is conducted.The experimental results show that the above model can effectively mine the motion patterns in the trajectory data and identify the best parking location points.The characteristics of the traffic travel portrait extracted from the results can provide certain suggestions for urban traffic management and have important value in pratice.