Research On Plot Information Calibration And Influencing Factors Of Flow Of Taxi Passenger Base On Multi-source Data

With the outbreak of COVID-19 around the world at the end of 2019,the life and travel patterns of residents in various countries have undergone great changes,and the information of plots inside cities has undergone great changes.As an important part of the urban transportation system,the size and distribution of passenger flows of taxis have a significant impact on urban transportation.However,due to the lack of relevant studies on influencing factors of taxi travel volume under the influence of the epidemic,it is often difficult for the government to predict the real travel volume of each region,and it is difficult for the government to formulate reasonable taxi control measures,resulting in frequent traffic problems within cities.In view of the above problems,this paper first uses taxi OD data and social media information to recalibrate the plot information,and on this basis,establishes a geographically weighted regression model for regression analysis to explore the changes of influencing factors and causes of taxi trips before and after the outbreak of the epidemic.The research contents of this paper mainly include: 1)On the basis of cluster analysis,the traffic district is redivided: Since the division of traffic communities is usually manually divided and the division cycle is long,it is difficult to reflect the changes of residents’ travel characteristics.This paper uses the OD point information of taxi trips and the improved Kmeans clustering algorithm on the basis of road network information to achieve the re-division of traffic communities.2)based on social media data calibration of traffic plot attribute: in this paper,and took up with POI label and the location information of social media data,through the establishment of membership function of different attributes,fitting to evaluate a single social media data contains the fuzzy information,finally through the calculation of fuzzy set with the standard set of target block the fitting degree of final calibration plot attribute.3)Comparative analysis of influencing factors of taxi trips at different stages of the epidemic according to vaccine coverage rate: Based on the geographical information data of New York City and taking taxi trips as the research object,a geographically weighted regression model and a global regression model were established for comparative analysis.The results show that the variable interpretation ability of the geographically weighted regression model is better than that of the general global regression model in describing the spatial change,and the spatial structure of a city has a significant impact on the taxi passenger flow.In addition,the proportion of positive patients,inpatients,vaccination rate,population density,proportion of elderly population,proportion of middle income,road network density,proportion of business district,subway accessibility,POI number have significant differences in the numerical and geographical distribution of the impact on the number of taxi trips at different stages of the epidemic in New York.This provides a scientific reference for the subsequent formulation of taxi pricing,control and other policies under the impact of the epidemic.