Comprehensive Visualization Analysis Of Large Data Of Multi-source Public Travel

With the advent of the era of big data,people's cognition of data has also been subverted.They no longer think that data is static and historical,but regard it as dynamic and fresh.Public travel is a major component of urban traffic behavior.The amount of data generated by public travel every day is very large.However,due to the diversity of public travel methods,the big data has heterogeneous heterogeneity,but the spatio-temporal information contained therein.It is also a good bond that connects them together.Based on the spatio-temporal information in the public travel big data,using the map as the carrier,spatial visualization can effectively depict the spatio-temporal portraits of public travel.Web GIS has become an important carrier of spatial visualization with its unique flexibility.This paper takes Beijing bus IC card data,subway card data,taxi track data as the data foundation,and uses the spatio-temporal information contained in the data as a link to establish the relationship between multi-source data and construct a public travel flow.Models,passenger travel chain models,and vehicle trajectory models provide a visual analysis of public travel flow and passenger behavior patterns.The research content and methods of this paper are:(1)Building a Hadoop cluster.The distributed file system HDFS is used to realize the storage of large amounts of data,and the efficient query and processing of massive data is realized under the parallel computing architecture of MapReduce,thus establishing the background support of visual research.(2)Data noise reduction and statistics.Noise reduction and statistics of the research data are achieved by writing filters.(3)Correlation and modeling of data.The spatio-temporal attribute carried by public travel data is an important link for establishing multi-source data association.Through the spatial correlation and time series relationship between various types of data,the association and fusion of multi-source data are realized,and the establishment of time and space is established.The total passenger flow distribution model of various modes of travel;according to the association of passenger IC card number,and carry out transfer analysis to obtain the passenger's travel chain;through the bus line and road network matching,get the bus line running route,and according to the adjacent The time difference between the passengers in the two stations is obtained,and the running speed of the bus and the road section is obtained.Similarly,the running speed between the adjacent track points of the taxi is calculated,thereby establishing the vehicle running track model.(4)Visual analysis of data.After the data is retrieved according to the established model,the passenger time and space distribution,passenger travel OD matrix,passenger travel chain and vehicle trajectory will be obtained,and the analysis results of the data will be combined through dynamic thematic map and non-geographic chart.Visual analysis.(5)Linkage analysis of multi-source data.Based on the visualization,this paper takes the passenger flow of the subway as the core,and analyzes the data of the bus card and the taxi trajectory data to verify the passenger flow reduction of the subway station and the surrounding bus stations,and the density of the taxi unloading point.Positive correlation between them.