Key Node Identification And Defect Area Mining In Urban Area Network Based On Trajectory

At present,the diseases of major cities in many cities are very serious.These problems include population explosion,traffic congestion,environmental pollution,and resource shortages.These problems have seriously affected the quality of life of urban residents.In many places,there has been a wave of counter-urbanization.With the development of science and technology and big data,the concept of urban computing and smart transportation came into being.The so-called urban computing is to use all sensors in the city to feel the dynamics of the city,and use the data to analyze and calculate in order to solve the problems in modern cities and serve the city and residents.We can use these existing urban multi-source data for research and analysis to solve problems in these cities and improve the quality of life of urban residents.This paper studies some problems of urban computing.The main contents are as follows: 1)the storage and processing of urban data.We mainly use Hadoop and Neo4 j graph databases to store urban data,and then use the Hadoop platform's computational framework to process the data.2)Spatial-temporal characteristics of urban trajectory data are analyzed.Through analysis,we found that the urban trajectory data shows a strong periodicity in time,the regularity of morning and evening peaks is also very obvious.However,the spatial regularity is closely related to its POI characteristics.3)Identify key nodes in the urban road network.We first need to build a regional network,and we use fishing nets to divide urban areas.Then,we use OD-based entropy regional importance assessment methods and some network centrality indictors to make importance assessments for each region,and finally to find road network node in some regions with the highest ranking.4)Discover potential defects in urban planning and mine causal relationships between them.We first need to construct region transition matrices,then use the improved Skyline algorithm to find the skyline points(region pairs)in every region transition matrix.Finally,we use this to mine frequent subgraphs to determine the really defective region pairs and analyze the causal relationship between defective region pairs.The above research content can provide urban planners with some suggestions for reference.This paper uses the data of two parts.The first part is the GPS data of more than 30,000 taxis in Shanghai in April 2015,and Shanghai Road Network data,map data,etc.The second part is unicom mobile phone signalling data for the first week of each month from July to December of 2017,and road network and map data for Changchun City.We need to preprocess these trajectory data and map it to the corresponding region to continue the subsequent algorithm.In the experiment,we used different methods to identify the key intersections of the two cities,compared the advantages of different methods,and compared the results with the actual situation to verify the validity of the method;Meanwhile,we use the improved Skyline algorithm to mine the potential planning defects in the two cities and analyze the causal relationships between these defects using the frequent subgraph mining method.Then we analyze and compare with the actual situation to verify the validity of the method.