Method and system for analyzing interactive passenger flow volume between subway stations

Abstract

The invention discloses an inter-subway station interactive passenger flow analysis method and system, which divides subway station domain units through a Thiessen polygon method, can precisely define a space range, captures land use functions and population distribution characteristics around subway stations, combines building census data and POI data, divides the land use functions into three types of living, working and business, and improves the efficiency of the system. Various POI interest points are extracted, a basis is provided for functional complementarity index construction and POI similarity calculation, residence-work, residence-business and work-business functional complementarity indexes are constructed, the functional relationship among subway station domain units is comprehensively reflected, an explanation basis is provided for the interactive passenger flow volume among subway stations, and meanwhile, the method has the advantages of being simple in structure, convenient to operate and high in practicability. A user-based collaborative filtering algorithm is applied to construct a POI semantic similarity matrix, the similarity between subway station domain units is quantified, a functional complementarity index and the POI semantic similarity matrix are introduced on the basis of a gravity model, a corrected gravity model is constructed, and accurate prediction and analysis of the interactive passenger flow volume between subway stations are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of passenger flow analysis, and in particular to a method and system for analyzing interactive passenger flow between subway stations. Background Art

[0002] With the acceleration of urbanization, subways, as an important part of urban public transportation, have important significance for urban transportation planning, operation management and policy formulation. The interactive passenger flow between subway stations not only reflects the travel needs and travel patterns of urban residents, but also reveals deep information such as urban spatial structure, economic development level and population distribution.

[0003] Traditional methods for analyzing interactive passenger flow between subway stations mainly rely on historical passenger flow data and use statistical analysis, time series analysis and other methods to predict future passenger flow. However, these methods often ignore the impact of the socioeconomic attributes and spatial structure of the areas surrounding the subway stations on passenger flow, resulting in inaccurate and incomplete analysis results.

[0004] In recent years, with the rapid development of big data and artificial intelligence technologies, more and more researchers have begun to try to apply new technologies to the analysis of interactive passenger flow between subway stations. For example, by collecting building census data and POI (Point of Interest) data around subway stations, the socioeconomic attributes and spatial structure of the areas around subway stations can be more carefully portrayed. At the same time, methods based on machine learning and data mining also provide new ideas for the prediction and analysis of passenger flow.

[0005] However, there are still some problems in the existing big data-based analysis methods of interactive passenger flow between subway stations. On the one hand, how to effectively integrate and utilize multi-source data in the areas surrounding subway stations to construct a comprehensive and accurate description of the attributes of subway station domain units is an urgent problem to be solved. On the other hand, how to establish an effective correlation model between these attribute descriptions and passenger flow to achieve accurate prediction and analysis of passenger flow is also a challenging task. Summary of the invention

[0006] In view of this, the present invention proposes a method and system for analyzing interactive passenger flow between subway stations, which can improve the accuracy and comprehensiveness of interactive passenger flow analysis between subway stations.

[0007] The technical solution of the present invention is achieved in this way: A method for analyzing interactive passenger flow between subway stations, specifically comprising: Divide subway station area units: With the subway station as the center, select the area within a preset radius as the subway station area unit, and further divide the subway station area unit using the Thiessen polygon method to determine the spatial range of each subway station area unit; Collect and process data: collect building census data and POI data within the subway station area unit, process the building census data and divide it into three types of land use functions: residential, work, and commercial, process the POI data and extract various POI points of interest; Construct functional complementarity indicators: construct functional complementarity indicators for residence-work, residence-commercial, and work-commercial based on the land use function types within the subway station area unit; Calculate POI similarity: divide the POI semantics in the subway station area into multiple categories, apply the user-based collaborative filtering recommendation algorithm, and construct the POI semantic similarity matrix; Construct a modified gravity model: Based on the gravity model, the functional complementarity index and POI semantic similarity matrix are introduced to construct a modified gravity model of interactive passenger flow between subway stations; Prediction and analysis: The constructed modified gravity model is applied to the prediction and analysis of the interactive passenger flow between actual subway stations to obtain the analysis results of the interactive passenger flow between subway stations.

[0008] As a further optional solution of the method for analyzing the interactive passenger flow between subway stations, the subway station is taken as the center, and an area within a preset radius is selected as a subway station area unit. The subway station area unit is further divided by the Thiessen polygon method to determine the spatial range of each subway station area unit, which specifically includes: According to the influence range of the subway station, a circular area with a preset radius and the station as the center is determined as the scope of preliminary division; In the initially divided circular area, select a preset number of control points; According to the control points, the Thiessen polygon algorithm is used to construct polygons, so that each polygon contains only one control point, and the boundary of the polygon is composed of the perpendicular bisectors of adjacent control points; The constructed polygons are optimized to ensure that they completely cover the initially divided circular area and there is no overlap or gap between adjacent polygons; The generated Thiessen polygons are superimposed on the initially divided circular areas to determine the final subway station area units.

[0009] As a further optional solution of the method for analyzing the interactive passenger flow between subway stations, the functional complementarity index of residence-work, residence-commercial, and work-commercial is constructed according to the land use function type within the subway station area unit, specifically including: Calculate the residential-work function complementarity RWC_ij, wherein the residential-work function complementarity RWC_ij reflects the degree of complementarity between the residential and work functions of the subway station area unit i and the subway station area unit j; Calculate the residential-commercial functional complementarity REC_ij, wherein the residential-commercial functional complementarity REC_ij reflects the degree of complementarity between residential and commercial functions between subway station domain unit i and subway station domain unit j; The work-commercial function complementarity WEC_ij is calculated, and the work-commercial function complementarity WEC_ij reflects the degree of complementarity of work and commercial functions between subway station domain unit i and subway station domain unit j.

[0010] As a further optional solution of the method for analyzing the interactive passenger flow between subway stations, the semantics of POIs in the subway station area are divided into multiple categories, and a recommendation algorithm based on user collaborative filtering is applied to construct a POI semantic similarity matrix, which specifically includes: Collect POI data of subway station area units; Semantically divide POI data according to thirteen predefined categories; For each subway station area unit, a 13-dimensional vector is constructed according to the number and category of POIs it contains. Each element in the vector represents the number of POIs of the corresponding category. The cosine similarity algorithm is used to calculate the POI similarity between two subway station area units; According to the calculated similarity values, a POI similarity matrix between subway station area units is constructed, and each element in the matrix represents the POI similarity between the corresponding two subway station area units.

[0011] As a further optional solution of the interactive passenger flow analysis method between subway stations, the modified gravity model of the interactive passenger flow between subway stations is constructed by introducing the functional complementarity index and the POI semantic similarity matrix on the basis of the gravity model, which specifically includes: According to the basic gravity model, combined with the functional complementarity index and POI semantic similarity, the formula of the modified gravity model is set; Using actual passenger flow data, the parameters in the model are calculated through regression analysis; According to the set formula and calculated parameters, the basic model is optimized and adjusted to obtain the modified gravity model.

[0012] As a further optional solution of the method for analyzing the interactive passenger flow between subway stations, the formula of the modified gravity model is specifically: ; in, From the starting point i To the end j of passenger flow, and are the floor area ratios of the starting and ending stations, is the station spacing between the starting and ending points, is the functional complementarity index between the starting and ending points, is the POI similarity matrix between the starting and ending points, and the α, β, γ, and k are adjustable parameters of the model.

[0013] An interactive passenger flow analysis system between subway stations, comprising: A partitioning module is used to select an area within a preset radius with the subway station as the center as the subway station area unit, and further divide the subway station area unit by the Thiessen polygon method to determine the spatial range of each subway station area unit; The extraction module is used to collect the building census data and POI data within the subway station area unit, process the building census data, divide it into three types of land use functions: residential, work, and commercial, and process the POI data to extract various POI points of interest; The first construction module is used to construct functional complementarity indicators of residence-work, residence-commercial, and work-commercial according to the land use function types within the subway station area unit; The second construction module is used to divide the POI semantics in the subway station area into multiple categories, apply the user-based collaborative filtering recommendation algorithm, and construct the POI semantic similarity matrix; The third construction module is used to introduce the functional complementarity index and the POI semantic similarity matrix on the basis of the gravity model to construct a modified gravity model of interactive passenger flow between subway stations; The prediction and analysis module is used to apply the constructed modified gravity model to the prediction and analysis of the interactive passenger flow between actual subway stations to obtain the analysis results of the interactive passenger flow between subway stations.

[0014] A computing device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for analyzing the interactive passenger flow between subway stations are implemented.

[0015] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for analyzing interactive passenger flow between subway stations.

[0016] The beneficial effects of the present invention are as follows: by dividing the subway station area unit through the Thiessen polygon method, the spatial range of each subway station area unit can be determined more accurately, so as to more accurately capture the land use function and population distribution characteristics around the subway station, collect and process building census data and POI data, divide the building census data into three types of land use functions: residence, work, and business, extract various types of POI points of interest, provide rich basic data for constructing functional complementarity indicators and calculating POI similarity, construct functional complementarity indicators of residence-work, residence-business, and work-business, which can comprehensively reflect the functional complementarity relationship between subway station area units, and provide a strong basis for explaining the interactive passenger flow between subway stations, divide POI semantics into multiple categories, and apply a user-based collaborative filtering recommendation algorithm to construct a POI semantic similarity matrix , which can quantify the degree of similarity between subway station domain units, further enrich the explanatory power of the model, introduce functional complementarity indicators and POI semantic similarity matrix on the basis of the gravity model, and construct a modified gravity model of interactive passenger flow between subway stations. This model can more accurately reflect the influencing factors and changing laws of interactive passenger flow between subway stations, and improve the accuracy and reliability of prediction. By applying the constructed modified gravity model to the prediction and analysis of interactive passenger flow between actual subway stations, more accurate and comprehensive analysis results of interactive passenger flow between subway stations can be obtained, which provides powerful decision-making support for urban planning, traffic management and other fields. Through refined spatial unit division, comprehensive data collection and processing, scientific indicator construction, precise model construction and prediction, accurate prediction and analysis of interactive passenger flow between subway stations are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A schematic diagram of a flow chart of a method for analyzing interactive passenger flow between subway stations according to the present invention; Figure 2 The present invention is a schematic diagram of the composition of an interactive passenger flow analysis system between subway stations. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] refer to Figure 1 to Figure 2 , a method for analyzing interactive passenger flow between subway stations, specifically comprising: Divide subway station area units: With the subway station as the center, select the area within a preset radius as the subway station area unit, and further divide the subway station area unit using the Thiessen polygon method to determine the spatial range of each subway station area unit; Collect and process data: collect building census data and POI data within the subway station area unit, process the building census data and divide it into three types of land use functions: residential, work, and commercial, process the POI data and extract various POI points of interest; Construct functional complementarity indicators: construct functional complementarity indicators for residence-work, residence-commercial, and work-commercial based on the land use function types within the subway station area unit; Calculate POI similarity: divide the POI semantics in the subway station area into multiple categories, apply the user-based collaborative filtering recommendation algorithm, and construct the POI semantic similarity matrix; Construct a modified gravity model: Based on the gravity model, the functional complementarity index and POI semantic similarity matrix are introduced to construct a modified gravity model of interactive passenger flow between subway stations; Prediction and analysis: The constructed modified gravity model is applied to the prediction and analysis of the interactive passenger flow between actual subway stations to obtain the analysis results of the interactive passenger flow between subway stations.

[0023] In this embodiment, the spatial scope of each subway station area unit is determined more accurately by dividing the subway station area unit through the Thiessen polygon method, so as to more accurately capture the land use function and population distribution characteristics around the subway station, collect and process the building census data and POI data, divide the building census data into three types of land use functions: residential, work, and commercial, extract various POI points of interest, and provide rich basic data for constructing functional complementarity indicators and calculating POI similarity. The functional complementarity indicators of residential-work, residential-commercial, and work-commercial are constructed, which can comprehensively reflect the functional complementary relationship between subway station area units and provide a strong basis for explaining the interactive passenger flow between subway stations. The POI semantics are divided into multiple categories, and the POI semantic similarity matrix is ​​constructed by applying the user-based collaborative filtering recommendation algorithm. It can quantify the similarity between subway station domain units, further enrich the explanatory power of the model, introduce functional complementarity index and POI semantic similarity matrix on the basis of gravity model, and construct a modified gravity model of interactive passenger flow between subway stations. This model can more accurately reflect the influencing factors and changing laws of interactive passenger flow between subway stations, and improve the accuracy and reliability of prediction. By applying the constructed modified gravity model to the prediction and analysis of interactive passenger flow between actual subway stations, more accurate and comprehensive analysis results of interactive passenger flow between subway stations can be obtained, which provides powerful decision-making support for urban planning, traffic management and other fields. Through refined spatial unit division, comprehensive data collection and processing, scientific indicator construction, precise model construction and prediction, accurate prediction and analysis of interactive passenger flow between subway stations are achieved.

[0024] Preferably, taking the subway station as the center, selecting an area within a preset radius as a subway station area unit, further dividing the subway station area unit by the Thiessen polygon method, and determining the spatial range of each subway station area unit specifically includes: According to the influence range of the subway station, a circular area with a preset radius and the station as the center is determined as the scope of preliminary division; In the initially divided circular area, select a preset number of control points; According to the control points, the Thiessen polygon algorithm is used to construct polygons, so that each polygon contains only one control point, and the boundary of the polygon is composed of the perpendicular bisectors of adjacent control points; The constructed polygons are optimized to ensure that they completely cover the initially divided circular area and there is no overlap or gap between adjacent polygons; The generated Thiessen polygons are superimposed on the initially divided circular areas to determine the final subway station area units.

[0025] In this embodiment, by determining a circular area with a preset radius and a subway station as the center as the preliminary division range, the influence range of the subway station can be more intuitively reflected, which provides a basis for subsequent detailed division. The Thiessen polygon algorithm is used to construct polygons, ensuring that each polygon contains only one control point and the boundary is composed of the perpendicular bisectors of adjacent control points. This division method makes each polygon more uniform in area and shape, thereby improving the accuracy of spatial division; selecting a preset number of control points to construct Thiessen polygons not only ensures the accuracy of division, but also avoids the increase in calculation complexity caused by too many control points. This method improves calculation efficiency while ensuring the quality of division. By optimizing polygons, it is ensured that they completely cover the preliminary divided circular area, and there is no overlap or gap between adjacent polygons, avoiding waste of resources and inaccurate division; It can be adjusted according to the actual influence range and preset radius of the subway station to adapt to the actual situation of different subway stations. The Thiessen polygon method itself has a certain flexibility, and can adapt to different division requirements by adjusting the number and position of control points, so as to obtain a subway station domain unit division result that is more in line with the actual situation.

[0026] Preferably, the functional complementarity indexes of residence-work, residence-commercial, and work-commercial are constructed according to the land use function types within the subway station area unit, specifically including: Calculate the residential-work function complementarity RWC_ij, wherein the residential-work function complementarity RWC_ij reflects the degree of complementarity between the residential and work functions of the subway station area unit i and the subway station area unit j; Calculate the residential-commercial functional complementarity REC_ij, wherein the residential-commercial functional complementarity REC_ij reflects the degree of complementarity between residential and commercial functions between subway station domain unit i and subway station domain unit j; The work-commercial function complementarity WEC_ij is calculated, and the work-commercial function complementarity WEC_ij reflects the degree of complementarity of work and commercial functions between subway station domain unit i and subway station domain unit j.

[0027] In this embodiment, the construction of the functional complementarity index helps to identify those subway station area units with strong functional complementarity, thereby guiding the construction and optimization of urban transportation facilities. For example, for areas with strong residential-work functional complementarity, public transportation connections can be strengthened to improve commuting efficiency; the work-commercial functional complementarity index can reflect the close connection between commercial activities and workplaces. By optimizing the transportation and facilities in these areas, more commercial activities and job opportunities can be attracted, thereby promoting an increase in passenger flow; the residential-commercial functional complementarity index helps to identify those subway station area units that are convenient for residents' daily shopping, leisure and entertainment. By identifying the commercial facilities and services in these areas, the impact on passenger flow can be understood.

[0028] It should be noted that the calculation of the residence-work function complementarity RWC_ij, which reflects the degree of complementarity between the residence and work functions between the subway station area unit i and the subway station area unit j, specifically includes the following steps: Calculate the land use function coefficient: for each metro station area unit i, calculate its land use function coefficient for residential function and work function, which needs to be calculated based on the building area of ​​each type of land in the metro station area unit and the total building area of ​​the same type of land in all metro station area units; Constructing functional complementarity index: Using the calculated land use function coefficient, construct the residential-work function complementarity index RWC_ij. This involves the product of the residential function and work function coefficients of the two sites i and j, reflecting the degree of complementarity between the residential and work functions of the two sites.

[0029] The above calculation of residential-commercial functional complementarity REC_ij, which reflects the degree of complementarity between residential and commercial functions between subway station area unit i and subway station area unit j, specifically includes the following steps: Calculate the land use function coefficient: For each metro station area unit i and j, calculate the land use function coefficient of its residential function and commercial function respectively. This usually involves counting the building area of ​​residential and commercial land in the metro station area unit and comparing it with the total building area of ​​the corresponding functions of all stations to obtain the function coefficient; Constructing functional complementarity index: Using the land use function coefficients of residential function and commercial function calculated above, construct the residential-commercial functional complementarity index REC_ij; Analyze the spatial distribution of REC_ij: By spatially visualizing the calculated REC_ij values, the spatial distribution characteristics of residential-commercial functional complementarity can be intuitively analyzed, which helps to identify which subway station area units have high residential and commercial functional complementarity.

[0030] Preferably, the semantic division of POIs in the subway station area into multiple categories, applying a user-based collaborative filtering recommendation algorithm, and constructing a POI semantic similarity matrix specifically includes: Collect POI data of subway station area units; Semantically divide POI data according to thirteen predefined categories; For each subway station area unit, a 13-dimensional vector is constructed according to the number and category of POIs it contains. Each element in the vector represents the number of POIs of the corresponding category. The cosine similarity algorithm is used to calculate the POI similarity between two subway station area units; According to the calculated similarity values, a POI similarity matrix between subway station area units is constructed, and each element in the matrix represents the POI similarity between the corresponding two subway station area units.

[0031] In this embodiment, by collecting POI data of subway station area units and performing semantic division, the functional distribution characteristics within the subway station area can be accurately captured. The method of constructing a thirteen-dimensional vector converts complex POI data into an easy-to-process numerical form to facilitate subsequent calculations and analysis. The cosine similarity algorithm is used to calculate the POI similarity between two subway station area units, which can accurately reflect their functional similarity. By constructing a POI similarity matrix, the functional relationship between each subway station area unit can be intuitively displayed, providing strong support for further analysis of functional complementarity. By analyzing the POI similarity matrix, subway station area units with unique functional characteristics can be identified.

[0032] It should be noted that the POI data of subway station area units are obtained from the Amap open platform or other data sources to ensure that the data covers 13 major categories such as residence, employment, and leisure.

[0033] In addition, the cosine similarity algorithm is used to calculate the POI similarity between two subway station area units. The specific formula is: ; in, and Represent the POI vectors of two subway station area units respectively, and They represent the magnitude of the vectors respectively.

[0034] Preferably, on the basis of the gravity model, the functional complementarity index and the POI semantic similarity matrix are introduced to construct a modified gravity model of the interactive passenger flow between subway stations, which specifically includes: According to the basic gravity model, combined with the functional complementarity index and POI semantic similarity, the formula of the modified gravity model is set; Using actual passenger flow data, the parameters in the model are calculated through regression analysis; According to the set formula and calculated parameters, the basic model is optimized and adjusted to obtain the modified gravity model.

[0035] In this embodiment, by introducing functional complementarity indicators (such as residential-work, residential-commercial, work-commercial and other functional complementarities) and POI semantic similarity matrix on the basis of the gravity model, the influencing factors of interactive passenger flow between subway stations can be considered more comprehensively. These factors reflect the functional connection and spatial interaction characteristics between subway station domain units, thereby improving the accuracy of the model in predicting actual passenger flow; the introduction of functional complementarity indicators and POI semantic similarity matrix enables the model to not only predict passenger flow, but also explain the causes of interactive passenger flow between subway stations, which helps urban planners and traffic managers to better understand the urban spatial structure and residents' travel behavior, and provide a scientific basis for urban traffic planning and policy formulation.

[0036] Preferably, the formula of the modified gravity model is specifically: ; in, is the passenger flow from starting point i to end point j, and are the floor area ratios of the starting and ending stations, is the station spacing between the starting and ending points, is the functional complementarity index between the starting and ending points, is the POI similarity matrix between the starting and ending points, and the α, β, γ, and k are adjustable parameters of the model.

[0037] In this embodiment, in the modified gravity model, α represents a coefficient parameter related to the floor area ratio of the starting station area, which is used to measure the influence of the floor area ratio of the starting station area on the interactive passenger flow; β represents a coefficient parameter related to the floor area ratio of the terminal station area, which is used to measure the influence of the floor area ratio of the terminal station area on the interactive passenger flow; γ represents a coefficient parameter related to the station distance between the starting and ending points, which is used to measure the influence of the distance attenuation effect on the interactive passenger flow; k is a coefficient parameter related to functional complementarity and similarity, which is used to measure the influence of these factors on the interactive passenger flow; δ represents other factors that are not explicitly listed but considered in the model that affect the interactive passenger flow.

[0038] An interactive passenger flow analysis system between subway stations, comprising: A partitioning module is used to select an area within a preset radius with the subway station as the center as the subway station area unit, and further divide the subway station area unit by the Thiessen polygon method to determine the spatial range of each subway station area unit; The extraction module is used to collect the building census data and POI data within the subway station area unit, process the building census data, divide it into three types of land use functions: residential, work, and commercial, and process the POI data to extract various POI points of interest; The first construction module is used to construct functional complementarity indicators of residence-work, residence-commercial, and work-commercial according to the land use function types within the subway station area unit; The second construction module is used to divide the POI semantics in the subway station area into multiple categories, apply the user-based collaborative filtering recommendation algorithm, and construct the POI semantic similarity matrix; The third construction module is used to introduce the functional complementarity index and the POI semantic similarity matrix on the basis of the gravity model to construct a modified gravity model of interactive passenger flow between subway stations; The prediction and analysis module is used to apply the constructed modified gravity model to the prediction and analysis of the interactive passenger flow between actual subway stations to obtain the analysis results of the interactive passenger flow between subway stations.

[0039] A computing device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for analyzing the interactive passenger flow between subway stations are implemented.

[0040] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for analyzing interactive passenger flow between subway stations.

[0041] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for analyzing interactive passenger flow between subway stations, characterized in that: Specifically include: Divide subway station area units: With the subway station as the center, select the area within a preset radius as the subway station area unit, and further divide the subway station area unit using the Thiessen polygon method to determine the spatial range of each subway station area unit; Collect and process data: collect building census data and POI data within the subway station area unit, process the building census data and divide it into three types of land use functions: residential, work, and commercial, process the POI data and extract various POI points of interest; Construct functional complementarity indicators: construct functional complementarity indicators for residence-work, residence-commercial, and work-commercial based on the land use function types within the subway station area unit; Calculate POI similarity: divide the POI semantics in the subway station area into multiple categories, apply the user-based collaborative filtering recommendation algorithm, and construct the POI semantic similarity matrix; Construct a modified gravity model: Based on the gravity model, the functional complementarity index and POI semantic similarity matrix are introduced to construct a modified gravity model of interactive passenger flow between subway stations; Prediction and analysis: The constructed modified gravity model is applied to the prediction and analysis of the interactive passenger flow between actual subway stations to obtain the analysis results of the interactive passenger flow between subway stations.

2. The method for analyzing the interactive passenger flow between subway stations according to claim 1, characterized in that: The method takes the subway station as the center, selects the area within a preset radius as the subway station area unit, further divides the subway station area unit by the Thiessen polygon method, and determines the spatial range of each subway station area unit, specifically including: According to the influence range of the subway station, a circular area with a preset radius and the station as the center is determined as the scope of preliminary division; In the initially divided circular area, select a preset number of control points; According to the control points, the Thiessen polygon algorithm is used to construct polygons, so that each polygon contains only one control point, and the boundary of the polygon is composed of the perpendicular bisectors of adjacent control points; The constructed polygons are optimized to ensure that they completely cover the initially divided circular area and there is no overlap or gap between adjacent polygons; The generated Thiessen polygons are superimposed on the initially divided circular areas to determine the final subway station area units.

3. The method for analyzing the interactive passenger flow between subway stations according to claim 2, characterized in that: The functional complementarity indicators of residence-work, residence-commercial, and work-commercial are constructed according to the land use function types within the subway station area unit, specifically including: Calculate the residential-work function complementarity RWC_ij, wherein the residential-work function complementarity RWC_ij reflects the degree of complementarity between the residential and work functions of the subway station area unit i and the subway station area unit j; Calculate the residential-commercial functional complementarity REC_ij, wherein the residential-commercial functional complementarity REC_ij reflects the degree of complementarity between residential and commercial functions between subway station domain unit i and subway station domain unit j; The work-commercial function complementarity WEC_ij is calculated, and the work-commercial function complementarity WEC_ij reflects the degree of complementarity of work and commercial functions between subway station domain unit i and subway station domain unit j.

4. The method for analyzing the interactive passenger flow between subway stations according to claim 3, characterized in that: The POI semantics in the subway station area are divided into multiple categories, and a recommendation algorithm based on user collaborative filtering is applied to construct a POI semantic similarity matrix, which specifically includes: Collect POI data of subway station area units; Semantically divide POI data according to thirteen predefined categories; For each subway station area unit, a 13-dimensional vector is constructed according to the number and category of POIs it contains. Each element in the vector represents the number of POIs of the corresponding category. The cosine similarity algorithm is used to calculate the POI similarity between two subway station area units; According to the calculated similarity values, a POI similarity matrix between subway station area units is constructed, and each element in the matrix represents the POI similarity between the corresponding two subway station area units.

5. The method for analyzing the interactive passenger flow between subway stations according to claim 4, characterized in that: On the basis of the gravity model, the functional complementarity index and the POI semantic similarity matrix are introduced to construct a modified gravity model of interactive passenger flow between subway stations, which specifically includes: According to the basic gravity model, combined with the functional complementarity index and POI semantic similarity, the formula of the modified gravity model is set; Using actual passenger flow data, the parameters in the model are calculated through regression analysis; According to the set formula and calculated parameters, the basic model is optimized and adjusted to obtain the modified gravity model.

6. The method for analyzing the interactive passenger flow between subway stations according to claim 5, characterized in that: The formula of the modified gravity model is specifically: ； in, is the passenger flow from starting point i to end point j, and are the floor area ratios of the starting and ending stations, is the station spacing between the starting and ending points, is the functional complementarity index between the starting and ending points, is the POI similarity matrix between the starting and ending points, α, β, γ, k is an adjustable parameter of the model.

7. A subway station interactive passenger flow analysis system, characterized in that: include: A partitioning module is used to select an area within a preset radius with the subway station as the center as the subway station area unit, further divide the subway station area unit by the Thiessen polygon method, and determine the spatial range of each subway station area unit; The extraction module is used to collect the building census data and POI data in the subway station area unit, process the building census data, divide it into three types of land use functions: residential, work, and commercial, and process the POI data to extract various POI points of interest; The first construction module is used to construct functional complementarity indicators of residence-work, residence-commercial, and work-commercial according to the land use function types within the subway station area unit; The second construction module is used to divide the POI semantics in the subway station area into multiple categories, apply the user-based collaborative filtering recommendation algorithm, and construct the POI semantic similarity matrix; The third construction module is used to introduce the functional complementarity index and the POI semantic similarity matrix on the basis of the gravity model to construct a modified gravity model of interactive passenger flow between subway stations; The prediction and analysis module is used to apply the constructed modified gravity model to the prediction and analysis of the interactive passenger flow between actual subway stations to obtain the analysis results of the interactive passenger flow between subway stations.

8. A computing device, characterized in that The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for analyzing interactive passenger flow between subway stations described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for analyzing interactive passenger flow between subway stations described in any one of claims 1 to 6.

Images