A method for recommending location of urban terminal buildings based on hypergraph network

Abstract

The present invention provides a method for recommending the site selection of a city terminal building based on a hypergraph network. The method comprises: performing mixed uniform design processing on the attributes and levels of the defined city terminal building site selection problem to obtain a scenario for the site selection of the city terminal building; generating a questionnaire based on the scenario, and obtaining the weight of the attribute according to the respondents' scores on the questionnaire; defining the nodes and hyperedges of the hypergraph according to the attributes, assigning weights to the hyperedges of the hypergraph according to the respondents' scores on the questionnaire, calculating the eigenvectors of the Laplace matrix of the hypergraph, and using the eigenvectors as the willingness recommendation degree of the candidate points for the site selection of the city terminal building; collecting statistics on the itinerary data of airport passengers to obtain the demand recommendation degree of the candidate points, and combining the willingness recommendation degree and the willingness recommendation degree of the candidate points to obtain the comprehensive recommendation degree of the candidate points. The method of the present invention combines the hypergraph theory with the SP survey method, and comprehensively obtains the recommendation index of the candidate points from the perspective of the user's own experience, and gives a recommended site selection.

Description

Technical Field

[0001] The present invention relates to the field of computer application technology, and in particular to a method for recommending the location of an urban terminal building based on a hypergraph network. Background Art

[0002] Hypergraph theory refers to the principle that, in general graph theory, each edge is limited to only two nodes associated with it. However, in real networks, complex, multi-layered relationships exist between nodes, making it difficult to intuitively express these relationships using a general graph. Hypergraph theory, a theoretical approach that combines graph theory and set theory, is more suitable for solving complex real-world network problems than general graphs, and has gradually formed the foundation of hypergraph theory.

[0003] With the development of big data and information networks, hypergraph theory has gradually been used to deal with large and complex network problems, such as scientific research collaboration network evolution models and image co-segmentation algorithms. In particular, in recent years, hypergraph theory has also been used as a representation of network structure to analyze complex network structures.

[0004] An SP (Stated Preference) survey is an actual survey conducted to obtain "people's subjective preferences for multiple options under hypothetical conditions." The main advantage of an SP survey is that designers can purposefully set selection methods and factor levels, using different designed options to obtain multiple data points from a single respondent, thereby improving survey efficiency. The survey process uses predetermined attributes and varying levels of various attributes to form various scenarios in a certain manner. These scenarios then form alternative options, and respondents score and assess their overall preferences for each attribute. Therefore, the advantage of an SP survey is that it can weigh the varying importance of multiple factors, thereby measuring the importance passengers place on multiple factors in a city terminal.

[0005] In the general graph theory of the existing technology, each edge is limited to having only two nodes associated with it. However, in real networks, there are extensive and complex multi-layer relationships between nodes, making it difficult to intuitively express the relationships between nodes in real networks through general graphs.

[0006] Generally speaking, questionnaires have high requirements on respondents for research on more complex survey questions. The questionnaires are too complicated, resulting in low accuracy and difficulty in obtaining effective questionnaires.

[0007] Currently, there is no method in the prior art for recommending the location of urban terminal buildings by combining a hypermap with a questionnaire. Summary of the Invention

[0008] The embodiment of the present invention provides a method for recommending the location of a city terminal building based on a hypergraph network, so as to effectively recommend the location of a city terminal building.

[0009] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions.

[0010] A method for recommending the location of an urban terminal building based on a hypergraph network, comprising:

[0011] Define the attributes of the city terminal building site selection problem and the levels corresponding to the attributes, and perform mixed uniform design processing on the defined attributes and levels to obtain the city terminal building site selection scenario;

[0012] Generating a questionnaire based on the scenario of selecting a site for a city terminal, providing the questionnaire to respondents for scoring, and obtaining attribute weights based on the respondents' scores on the questionnaire;

[0013] Constructing a hypergraph, defining nodes and hyperedges of the hypergraph according to the attributes, assigning weights to the hyperedges of the hypergraph according to the respondents' scores on the questionnaire, defining a Laplace matrix of the hypergraph, calculating an eigenvector of the Laplace matrix, and using the eigenvector as a willingness to recommend candidate points for the site selection of the city terminal;

[0014] The itinerary data of airport passengers are collected to obtain the demand recommendation degree of the candidate point, and the willingness recommendation degree and the willingness recommendation degree of the candidate point are combined to obtain the comprehensive recommendation degree of the candidate point.

[0015] Preferably, the defining of the attributes of the city terminal building site selection problem and the levels corresponding to the attributes, and performing a mixed uniform design process on the defined attributes and levels to obtain the city terminal building site selection scenario, include:

[0016] Based on the database, the attributes of passengers' travel intentions are sorted out, and after integration and reduction, the attributes of the city terminal site selection problem are defined. The attributes include: the distribution of hotels near the terminal, the commercial coverage near the terminal, the level of catering near the terminal, the level of leisure and entertainment facilities near the terminal, the level of life service facilities near the terminal, and the convenience of transportation near the terminal. The levels of the corresponding attributes are defined, and the defined attributes and levels are mixed and uniformly designed using the DPS data processing system to obtain multiple different city terminal site selection scenarios.

[0017] Preferably, the step of generating a questionnaire based on the scenario of selecting a site for a city terminal, providing the questionnaire to respondents for scoring, and obtaining attribute weights based on the respondents' scores on the questionnaire includes:

[0018] A questionnaire is generated based on the optimal scenario. The questionnaire includes a series of scenarios with different levels of attributes related to the terminal building location. The questionnaire is provided to respondents for scoring, and the respondents' scores are obtained. A binomial logit model is used in the questionnaire to analyze the choice intention. The specific form of the utility function is as follows:

[0019]

[0020] Among them, X njk Represents the attribute variables in the jth scenario; θ j0 The relative superiority of the simulation scenario is the parameter to be estimated; θ jk Represents the relative weight of each attribute, which is the parameter to be estimated;

[0021] The probability P of the respondents choosing the jth simulation scenario is calculated as the proportion of the number of people who choose this scenario to the total number of people:

[0022]

[0023] Among them, P nj and P nk Indicates the proportion of respondents who choose the nth attribute in the jth scenario and the kth scenario to the total number of respondents, θ j0 and θ k0 The relative superiority of the simulated scenario j and scenario k is the parameter to be estimated, X nji and X nki Represent the attribute variables in scene j and scene k respectively.

[0024] Combine (1) and (2) to calculate the weight of each attribute, and use θ ji and θ ki Indicates the weight of attribute i in scenarios j and k.

[0025] Preferably, the constructing of a hypergraph, defining nodes and hyperedges of the hypergraph according to the attributes, assigning weights to hyperedges of the hypergraph according to respondents' scores on the questionnaire, defining a Laplace matrix of the hypergraph, calculating an eigenvector of the Laplace matrix, and using the eigenvector as a willingness to recommend candidate locations for a city terminal building, includes:

[0026] Let H = (V, E) represent a hypergraph, where V = {v1, v2, ..., v n} is a finite set of vertices of the hypergraph, is called a hyperedge of the hypergraph, The hypergraph is a finite set of hyperedges, the relevant attributes are determined as hyperedges of the hypergraph, the hyperedges are assigned the weights of the attributes calculated previously, and the initial candidate points for the city terminal site selection are defined as nodes of the hypergraph;

[0027] The structure of the hypergraph is represented by the incidence matrix H∈{0,1} |V|×|E| Indicates that each entry H(v,e) in the matrix is ​​used to indicate whether the vertex v is in the hyperedge e. The incidence matrix is ​​a continuous matrix with elements ranging from 0 to 1:

[0028]

[0029] H(v,e) is used to express the possibility that vertex v belongs to hyperedge e or the importance of vertex v to hyperedge e. The excess degree of hyperedge e δ(e) and the excess degree of vertex v d(v) are defined as:

[0030]

[0031]

[0032] The Laplacian matrix of the hypergraph is defined as:

[0033]

[0034] The normalized Laplace matrix is:

[0035]

[0036] Among them, D e is the diagonal matrix of hyperedge degree; D v is the diagonal matrix of vertex hyperdegrees; H is the incidence matrix representing the hypergraph structure; W is the diagonal matrix of hyperedge weights;

[0037] Extract the non-negative minimum eigenvalue of the Laplace matrix and its corresponding eigenvector: Υ=[Υ1,Υ2,……,Υ k ], and the eigenvector Υ is used as the passenger willingness of the candidate site selection for the city terminal.

[0038] Preferably, the statistical analysis of airport passenger travel data to obtain the demand recommendation degree of the candidate point and the combination of the willingness recommendation degree and the willingness recommendation degree of the candidate point to obtain the comprehensive recommendation degree of the candidate point include:

[0039] Based on the passenger clustering results of bus stops within a certain range of the candidate points for the city terminal, the distribution of the number of each attribute in this range is counted to obtain an association matrix. In the association matrix, each row represents an attribute corresponding to each candidate point, and each column represents all candidate points corresponding to an attribute. The number of rows is equal to the number of attributes, and the number of columns corresponds to the number of candidate points. Each element represents the importance of the candidate point relative to all candidate points under this attribute. The weight of the transportation attribute is calculated based on whether it can directly reach the airport, and the association matrix is ​​further modified.

[0040] Based on the geographic data and card swiping data of subway stations within the range, and combined with the clustering results, the number of card swipes from different locations to the airport is counted. The demand of the station is regarded as the demand of the nearest candidate point based on the proximity principle, and the demand distribution within the range is obtained. In the same way as calculating the recommendation intention, the demand is defined as an attribute, and the association matrix is ​​further modified. The Laplace matrix calculation is performed on the association matrix to obtain the non-negative minimum eigenvalue and eigenvector, and the eigenvector is used as the demand recommendation degree vector.

[0041] Preferably, the combining the willingness recommendation degree and the willingness recommendation degree of the candidate point to obtain the comprehensive recommendation degree of the candidate point includes:

[0042] The willingness recommendation degree and the demand recommendation degree are given the same weight, and the willingness recommendation degree and the demand recommendation degree are added together to obtain the comprehensive recommendation degree. The comprehensive recommendation degree includes both the subjective attributes of passengers for the location of the city terminal and the demand attributes derived from objective facts.

[0043] As can be seen from the technical solutions provided by the embodiments of the present invention described above, this invention provides a recommendation method that is suitable for multi-attribute, multi-dimensional site selection problems. By combining hypergraph theory with SP survey methods, and taking into account the user's own experience, this method considers as many valid attributes as possible in the case of multi-attribute site selection problems, comprehensively derives recommendation indicators for candidate points, and provides recommended sites. This method can serve as a methodology for complex site selection problems and also opens up new ideas for future multi-attribute, multi-dimensional site selection problems.

[0044] Additional aspects and advantages of the present invention will be set forth in part in the following description, will become apparent from the following description, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 any creative work.

[0046] Figure 1 A processing flow chart of a method for recommending the location of an urban terminal building based on a hypergraph network provided by an embodiment of the present invention;

[0047] Figure 2 A schematic diagram of a hypergraph provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limiting the present invention.

[0049] It will be understood by those skilled in the art that, unless expressly stated otherwise, the singular forms "a", "an", "said" and "the" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the description of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, "connected" or "coupled" as used herein may include wireless connections or couplings. The term "and / or" used herein includes any unit and all combinations of one or more associated listed items.

[0050] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art in the art to which the present invention pertains. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and, unless defined as such herein, will not be interpreted in an idealized or overly formal sense.

[0051] To facilitate understanding of the embodiments of the present invention, several specific embodiments will be further explained below with reference to the accompanying drawings, and each embodiment does not constitute a limitation on the embodiments of the present invention.

[0052] The processing flow of a method for recommending the location of an urban terminal building based on a hypergraph network provided by an embodiment of the present invention is as follows: Figure 1 As shown, the following processing steps are included:

[0053] Step S1: Based on the POI database, passenger travel intention attributes are collated and reduced to define relevant attributes for the city terminal site selection problem. These attributes include: hotel distribution near the terminal, commercial coverage near the terminal, catering quality near the terminal, leisure and entertainment facilities near the terminal, living service facilities near the terminal, and transportation convenience near the terminal. Levels for these attributes are defined, and the DPS data processing system performs mixed uniform design processing on these defined attributes and levels to generate multiple different city terminal site selection scenarios.

[0054] Step S2: Based on the scenarios selected after multiple rounds of mixed uniform processing in Step S1, a questionnaire is generated for the SP survey. The questionnaire includes scenarios with varying levels of attributes related to terminal location selection, such as traffic conditions around the terminal and nearby hotels. This questionnaire is provided to the respondents for rating. After the questionnaire is completed, the respondents' ratings are obtained to determine the weights of the relevant attributes.

[0055] Step S3: Construct a hypergraph, define its nodes and edges based on its attributes, and assign weights to its edges based on the questionnaire scores obtained in step S2. Define the hypergraph's Laplace matrix, calculate its eigenvectors, and use them as the willingness to recommend candidate locations for the city terminal.

[0056] Step S4: Count the itinerary data of airport passengers to obtain the demand recommendation degree of the candidate point, and combine the willingness recommendation degree and the willingness recommendation degree of the candidate point to obtain the comprehensive recommendation degree of the candidate point for the city terminal site selection.

[0057] The specific process of determining the attributes and levels in step S1 is as follows:

[0058] (1) Based on relevant literature and the various characteristics of the site selection problem, determine the attributes and level of the site selection problem.

[0059] (2) After determining the attributes and levels of the site selection problem, a mixed uniform design is used to construct the optimal scenario for the site selection of the urban terminal. During the design process, multiple mixed uniform processes are performed to ensure that the constructed scenario can well reflect the emphasis of different attributes and facilitate subsequent data processing.

[0060] In step S2, the specific process of questionnaire processing is as follows:

[0061] (1) The binomial logit (BL) model is used to analyze the choice intention. The specific form of the utility function is

[0062]

[0063] Among them, X njk Represents the attribute variables in the jth scenario; θ j0 The relative superiority of the simulation scenario is the parameter to be estimated; θ jk Represents the relative weight of each attribute and is the parameter to be estimated.

[0064] (2) Calculate the probability P of the respondents choosing the jth simulation scenario as the proportion of the number of people who choose this scenario to the total number of people:

[0065]

[0066] (3) Combine (7) and (8) to calculate the weight of each attribute, which is the parameter to be estimated.

[0067] The above step S3 specifically includes: Hypergraph theory stipulates that objects with the same attribute characteristics belong to the same set, and all nodes in the set are associated with other nodes. Figure 2 A schematic diagram of a hypergraph provided by an embodiment of the present invention is shown in FIG. Figure 2 As shown, according to the hypergraph theory, H = (V, E) represents a hypergraph, where V = {v1, v2, ..., v n} is a finite set of vertices of the hypergraph, is called the hyperedge of the hypergraph, and E={e1,e2,……,e n} is a finite set of hyperedges of the hypergraph.

[0068] The structure of a hypergraph is usually represented by the incidence matrix H∈{0,1} |V|×|E| To express it, each item H(v,e) in the matrix is ​​used to indicate whether the vertex v is in the hyperedge e:

[0069]

[0070] In complex situations such as the site selection of urban terminal buildings studied in this paper, the incidence matrix is ​​not a simple (0, 1) matrix, but a continuous matrix with elements ranging from 0 to 1. Therefore, H(v, e) is used to represent the possibility that vertex v belongs to hyperedge e or the importance of vertex v to hyperedge e. In addition, in a weighted hypergraph, each hyperedge e of the hypergraph is assigned a weight w(e) to represent the importance of its connection relationship in the hypergraph. Accordingly, the excess degree δ(e) of hyperedge e and the excess degree d(v) of vertex v are defined as:

[0071]

[0072]

[0073] Laplacian matrix: Spectral analysis of graphs is based on the eigenvalues ​​and eigenvectors of the graph’s Laplacian matrix. In general, the Laplacian matrix is ​​defined as:

[0074] Δ=DA (6)

[0075] Where D is the diagonal matrix of the vertex degree of a general graph; A is the adjacency matrix of a general graph. In a hypergraph, the Laplacian matrix is ​​more complex than in a general graph and is defined as:

[0076]

[0077] The normalized Laplace matrix is:

[0078]

[0079] Among them, D e is the diagonal matrix of hyperedge degree; D v is the diagonal matrix of vertex hyperdegrees; H is the incidence matrix representing the hypergraph structure; W is the diagonal matrix of hyperedge weights.

[0080] The basic definition process of the hypergraph in the present invention includes:

[0081] (1) Determine the hyperedges of the hypergraph, and determine the relevant attributes previously defined in step S1 as the hyperedges of the hypergraph.

[0082] (2) The candidate points for the initial city terminal site selection are defined as nodes of the hypergraph, and the hyperedges are assigned the weights of the attributes calculated previously.

[0083] The specific process of calculating the hypergraph related indicators and deriving the recommendation index is as follows:

[0084] (1) Using the incidence matrix, formula (1) expresses whether each node in the hypergraph is in the corresponding hyperedge.

[0085] The incidence matrix is ​​not a simple (0, 1) matrix, but a continuous matrix with elements ranging from 0 to 1. H(v, e) represents the probability that vertex v belongs to hyperedge e or the importance of vertex v to hyperedge e. In addition, in a weighted hypergraph, each hyperedge e of the hypergraph is assigned a weight w(e) to represent the importance of its connection relationship in the hypergraph. Accordingly, the excess degree δ(e) of hyperedge e is defined as formula (5) and the excess degree d(v) of vertex v is defined as formula (6).

[0086] (2) Definition of Laplacian matrix in hypergraph: In hypergraph, the Laplacian matrix is ​​more complex than that in general graph. It is defined as formula (7). The normalized Laplacian matrix is ​​formula (8), where D e is the diagonal matrix of hyperedge degree; D v is the diagonal matrix of vertex hyperdegrees; H is the incidence matrix representing the hypergraph structure; W is the diagonal matrix of hyperedge weights.

[0087] (3) Calculate the Laplace matrix and extract the non-negative minimum eigenvalue of the Laplace matrix and its corresponding eigenvector: Υ=[Υ1,Υ2,……,Υ k ], and the eigenvector Υ is used as the passenger willingness of the candidate site selection for the city terminal.

[0088] In step S4, the specific steps for processing the demand recommendation degree are as follows:

[0089] Based on the passenger clustering results for bus stops within a certain range of the candidate locations for the city terminal, the distribution of each attribute within this range was calculated to generate an association matrix. In this association matrix, each row represents an attribute corresponding to each candidate location, and each column represents all candidate locations corresponding to a given attribute. The number of rows equals the number of attributes, and the number of columns corresponds to the number of candidate locations. Each element represents the relative importance of a candidate location relative to all other candidate locations for that attribute. For each transportation attribute, a weight is calculated based on whether it provides fast, direct access to the airport, and the association matrix is ​​further modified.

[0090] Based on the geographic data and card swipe data of subway stations within the range, combined with clustering results, the number of card swipes from different locations to the airport was counted. Based on the principle of proximity, the demand for each station was treated as the demand for the nearest candidate point. This yielded the demand distribution within the range. Following the same method used to calculate the willingness to recommend, the demand was defined as an attribute, and the association matrix was further modified. A Laplace matrix calculation was performed on the above association matrix to obtain the non-negative minimum eigenvalue and eigenvector. The eigenvector was used as the demand recommendation vector.

[0091] The weights of willingness recommendation and demand recommendation are set to be the same, so the two are added together to obtain the comprehensive recommendation. Therefore, the comprehensive recommendation includes both the subjective attributes of passengers for the location of the city terminal and the demand attributes derived from objective facts.

[0092] To sum up, the embodiments of the present invention are based on the perspective of solving complex site selection problems. The method can construct a multi-attribute and multi-dimensional site selection problem using a hypergraph, and calculate the comprehensive recommendation degree of the candidate point in combination with the hypergraph, and finally determine the recommended site selection. However, most of the existing site selection problems only consider a single attribute, cannot solve multi-dimensional problems, and cannot give a more effective recommended site selection. The questionnaire survey process of the present invention adopts SP survey and constructs a virtual scene, so that the respondents can better understand the questionnaire, thereby obtaining more accurate questionnaire results. In general questionnaires, the questions are relatively redundant, which can easily confuse the respondents and make it difficult to obtain effective answers.

[0093] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0094] From the above description of the embodiments, it can be seen that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus the necessary general-purpose hardware platform. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present invention or certain parts of the embodiments.

[0095] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiments. The device and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. A person of ordinary skill in the art can understand and implement it without making any creative efforts.

[0096] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for recommending the location of an urban terminal building based on a hypergraph network, characterized in that: include: Define the attributes of the city terminal building site selection problem and the levels corresponding to the attributes, and perform mixed uniform design processing on the defined attributes and levels to obtain the city terminal building site selection scenario; Generating a questionnaire based on the scenario of selecting a site for a city terminal, providing the questionnaire to respondents for scoring, and obtaining attribute weights based on the respondents' scores on the questionnaire; Constructing a hypergraph, defining nodes and hyperedges of the hypergraph according to the attributes, assigning weights to the hyperedges of the hypergraph according to the respondents' scores on the questionnaire, defining a Laplace matrix of the hypergraph, calculating an eigenvector of the Laplace matrix, and using the eigenvector as a willingness to recommend candidate points for the site selection of the city terminal; Collecting data on the itinerary of airport passengers to obtain the demand recommendation degree of the candidate point, and combining the willingness recommendation degree and the willingness recommendation degree of the candidate point to obtain the comprehensive recommendation degree of the candidate point; The aforementioned definition of the attributes of the city terminal building site selection problem and the levels corresponding to the attributes, and performing a mixed uniform design process on the defined attributes and levels to obtain the city terminal building site selection scenario, include: The database is used to organize the attributes of passengers' travel intentions. After integration and reduction, the attributes of the city terminal site selection problem are defined. These attributes include: hotel distribution near the terminal, commercial coverage near the terminal, catering level near the terminal, leisure and entertainment facilities near the terminal, life service facilities near the terminal, and transportation convenience near the terminal. The levels of the corresponding attributes are defined, and the defined attributes and levels are mixed and uniformly designed using the DPS data processing system to obtain multiple different city terminal site selection scenarios. The step of generating a questionnaire based on the scenario of selecting a site for a city terminal, providing the questionnaire to respondents for scoring, and obtaining attribute weights based on the respondents' scores of the questionnaire includes: A questionnaire is generated based on the optimal scenario. The questionnaire includes a series of scenarios with different levels of attributes related to the terminal building location. The questionnaire is provided to respondents for scoring, and the respondents' scores are obtained. A binomial logit model is used in the questionnaire to analyze the choice intention. The specific form of the utility function is as follows: Among them, X njk Represents the attribute variables in the jth scenario; θ j0 The relative superiority of the simulation scenario is the parameter to be estimated; θ jk Represents the relative weight of each attribute, which is the parameter to be estimated; The probability P of the respondents choosing the jth simulation scenario is calculated as the proportion of the number of people who choose this scenario to the total number of people: Among them, P nj and P nk Indicates the proportion of respondents who choose the nth attribute in the jth scenario and the kth scenario to the total number of respondents, θ j0 and θ k0 The relative superiority of the simulated scenario j and scenario k is the parameter to be estimated, X nji and X nki Represent the attribute variables in scene j and scene k respectively; Combine (1) and (2) to calculate the weight of each attribute, and use θ ji and θ ki Indicates the weight of attribute i in scenarios j and k.

2. The method according to claim 1, characterized in that The steps of constructing a hypergraph, defining nodes and hyperedges of the hypergraph according to the attributes, assigning weights to hyperedges of the hypergraph according to respondents' scores on the questionnaire, defining a Laplace matrix of the hypergraph, calculating an eigenvector of the Laplace matrix, and using the eigenvector as a willingness to recommend a candidate site for a city terminal building include: Let H = (V, E) represent a hypergraph, where V = {v1, v2, ..., v n } is a finite set of vertices of the hypergraph, It is called the hyperedge of the hypergraph, E={e1,e2,……,e n } is a finite set of hyperedges of the hypergraph, the relevant attributes are determined as the hyperedges of the hypergraph, the hyperedges are assigned the weights of the attributes calculated previously, and the initial candidate points for the city terminal site selection are defined as the nodes of the hypergraph; The structure of the hypergraph is represented by the incidence matrix H∈{0,1} V×E Indicates that each entry H(v,e) in the matrix is ​​used to indicate whether the vertex v is in the hyperedge e. The incidence matrix is ​​a continuous matrix with elements ranging from 0 to 1: H(v,e) is used to express the possibility that vertex v belongs to hyperedge e or the importance of vertex v to hyperedge e. The excess degree of hyperedge e δ(e) and the excess degree of vertex v d(v) are defined as: d(v)=∑ e∈E w(e)*H(v,e) (5) Each hyperedge e of the hypergraph is assigned a weight w(e); The Laplacian matrix of the hypergraph is defined as: The normalized Laplace matrix is: Among them, D e is the diagonal matrix of hyperedge degree; D v is the diagonal matrix of vertex hyperdegrees; H is the incidence matrix representing the hypergraph structure; W is the diagonal matrix of hyperedge weights; Extract the non-negative minimum eigenvalue of the Laplace matrix and its corresponding eigenvector: γ=[Υ1,Υ2,……,Υ k ], and the eigenvector Υ is used as the passenger willingness of the candidate site selection for the city terminal.

3. The method according to claim 2, characterized in that The statistical analysis of airport passenger itinerary data to obtain the demand recommendation degree of the candidate point and the combination of the willingness recommendation degree and the willingness recommendation degree of the candidate point to obtain the comprehensive recommendation degree of the candidate point include: Based on the passenger clustering results of bus stops within a certain range of the candidate points for the city terminal, the distribution of the number of each attribute in this range is counted to obtain an association matrix. In the association matrix, each row represents an attribute corresponding to each candidate point, and each column represents all candidate points corresponding to an attribute. The number of rows is equal to the number of attributes, and the number of columns corresponds to the number of candidate points. Each element represents the importance of the candidate point relative to all candidate points under this attribute. The weight of the transportation attribute is calculated based on whether it can directly reach the airport, and the association matrix is ​​further modified. Based on the geographic data and card swiping data of subway stations within the range, and combined with the clustering results, the number of card swipes from different locations to the airport is counted. The demand of the station is regarded as the demand of the nearest candidate point based on the proximity principle, and the demand distribution within the range is obtained. In the same way as calculating the recommendation intention, the demand is defined as an attribute, and the association matrix is ​​further modified. The Laplace matrix calculation is performed on the association matrix to obtain the non-negative minimum eigenvalue and eigenvector, and the eigenvector is used as the demand recommendation degree vector.

4. The method according to claim 3, characterized in that The step of combining the willingness recommendation degree and the willingness recommendation degree of the candidate point to obtain the comprehensive recommendation degree of the candidate point includes: The willingness recommendation degree and the demand recommendation degree are given the same weight, and the willingness recommendation degree and the demand recommendation degree are added together to obtain the comprehensive recommendation degree. The comprehensive recommendation degree includes both the subjective attributes of passengers for the location of the city terminal and the demand attributes derived from objective facts.

Images