Research On The Spatial-temporal Characteristics Of Residents’ Travel And The Relationship Between Job And Housing Based On Multi-source Data Mining

The study of residents’ travel characteristics plays an important role in urban construction planning and urban transportation planning,and the rules of residents’ travel during morning and evening peak hours can reflect the spatial relationship between urban occupation and residence to a certain extent,so it is of great significance to excavate residents’ travel characteristics.Due to the diversity of travel modes of residents and the single survey data has the disadvantages of large workload,troublesome data processing,and strong subjectivity,the data obtained by traditional survey methods can no longer meet the needs of contemporary society.Today,with the widespread deployment of the global positioning system(GPS)and the rapid development of geographic information technology(GIS),various forms of traffic data are stored,which makes it possible to use intelligent big data to fully analyze the characteristics of residents’ travel.This paper takes taxi GPS trajectory data and subway data as the main,urban administrative boundary,road network and POI as the supplementary data,excavates the spatial and temporal characteristics of urban residents’ travel,and studies the urban occupationresidential relationship with the data of "suitable travel".First,analyze the data type and structure,and summarize the connections between the data.Secondly,the Map Reduce framework is built through python3.7 to process taxi GPS data,including data cleaning,coordinate system conversion and map matching,OD extraction and matching,etc.;using statistical methods to clean subway passenger flow data and OD extraction,And match the passenger flow with the subway POI station on the map to provide a data basis for the study of the article.Third,use the processed data to mine the spatial and temporal characteristics of residents’ travel,including overall travel characteristics,peak time identification,peak time travel characteristics,travel OD hotspot distribution,hotspot direction differentiation characteristics,and regional movement pattern characteristics,and make decision-making layers for them Fusion analysis.Among them,in order to better explore the hot spots of residents’ travel,the paper also proposes a hot spot detection model based on nuclear density analysis,and uses the natural fracture method to classify the hot spots.In addition,the standard deviation ellipse model established in ARCGIS detects the distribution direction of residents’ hot spots.Finally,determine the research scale of job-residential balance,and take taxi and subway commuter passenger flow as variable indicators,based on the Gaussian Mixture Model(GMM)to identify the functional location of the research station,and discuss the spatial distribution of urban residential and residential;on this basis,combined Based on the "adaptable travel" data,a generalized autoregressive conditional variance(GARCH)model is established to study the relationship between traffic commuting and job-housing balance around rail stations.Studies have shown that the nuclear density hotspot detection model can effectively detect the hotspots of residents’ travel,making up for the shortcomings of traditional clustering methods to determine the model parameters;compared with single data,multi-source data can more fully mine the spatial and temporal characteristics of urban residents’ travel;Xi’an The spatial distribution of occupational residences in the main urban area of the city is mainly centered on the southwestern part of the main urban area,and extends to the surroundings in the form of "employment-mixing-residential-mixing";the balance of occupations and residences around the rail station and the passenger flow in and out of the rail station Balance,cumulative outbound passenger flow,typical office,whether the rail station is an interchange station,the location of the station,whether there are large public facilities in the vicinity,is negatively related to the cumulative inbound passenger flow,typical residence,and whether the rail is the originating station Related.