Multi-level living-recreation space structure identification method based on overlapping community discovery algorithm

Abstract

The invention relates to a multi-level living-recreation space structure identification method based on an overlapping community discovery algorithm. The method comprises the following steps: S1, constructing a living-recreation network based on mobile phone signaling data; s2, converting the living-recreation network into a weighted edge network; s3, based on the weighted edge network, identifying an edge community structure in the weighted edge network; s4, mapping the side community structure back to the living-recreation network, and obtaining a node community structure corresponding to each original node; s5, constructing a hypergraph based on the node community structure, and identifying and obtaining an Nth-layer community structure by using a louvain community discovery algorithm based on the hypergraph; and S6, eliminating communities of which the community scale is smaller than a preset value in the Nth layer of community structure, and iteratively executing the step S5 for the reserved communities until an iteration ending condition is met, so as to obtain the multi-layer living-recreation space structure of the city. Compared with the prior art, the invention provides the method for automatically identifying the urban multi-level living-recreation space structure based on the resident activity rule.

Description

Technical Field

[0001] This invention relates to the field of urban spatial structure recognition technology, and in particular to a multi-level residential-recreation spatial structure recognition method based on an overlapping community detection algorithm. Background Technology

[0002] The Central Urban Work Conference pointed out the need to focus on building comfortable and convenient livable cities and optimizing urban spatial structure. Traditional urban spatial structure research has largely focused on the urban morphology level, emphasizing the static distribution of various spatial elements while neglecting the dynamic structural characteristics presented by the flow of elements within the city. This static perspective leads to insufficient understanding of the functional organization mechanisms within the city, resulting in a series of practical problems. Taking recreational activities as an example, without effective organization of residents' recreational behavior, public activity centers are prone to homogeneous competition, leading to inefficient land resource allocation and insufficient spatial vitality.

[0003] Supported by flow space theory and big data technology, identifying urban spatial structure through residents' dynamic activity data has become possible, which helps analyze the functional affiliation and community attributes of each spatial unit. However, existing identification methods still have significant limitations: First, spatial unit classification methods are exclusive, and units can only be classified into a single category, failing to reflect the actual functional flow and overlapping characteristics; second, urban space itself has multi-level and systematic structural characteristics, while conventional community detection algorithms struggle to effectively identify multi-level spatial organizational relationships at the urban scale. To address the problems in existing technologies regarding community overlap identification, Chinese patent application CN108520471A provides a method for identifying overlapping communities. Specifically, it purifies the network by deleting unstable edges, and then divides the purified network into several connected, tightly packed community cores and some isolated discrete points. The similarity between discrete points and communities is then calculated, and finally, the community to which a discrete point belongs is determined based on the similarity. In addition, information such as the user's geographical location and time is also introduced when identifying overlapping communities. Although the division of overlapping communities is ultimately achieved, information loss may occur during the identification process. Furthermore, the community structure in the identification process forces the network to be divided into a binary structure of community core and discrete points, which pre-limits the communities to grow only around a fixed core. This non-overlapping assumption contradicts the natural overlapping characteristics that are common in real networks, resulting in inaccurate division results. It can be seen that although the existing technology can achieve the identification of overlapping communities, there is still no relevant technology for the identification of the residential-recreational space structure of urban residents.

[0004] Therefore, providing a technical means to automatically identify the multi-level residential-recreation space structure in cities based on residents' activity patterns is a technical problem that needs to be solved. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a multi-level residential-recreation space structure identification method based on the overlapping community detection algorithm. This invention provides a method for identifying the urban residential-recreation space structure based on the overlapping community algorithm, which effectively solves the shortcomings of traditional identification technology in terms of dynamism, overlap and hierarchy, and provides a scientific basis for urban spatial structure optimization and refined management.

[0006] The objective of this invention can be achieved through the following technical solutions: According to a first aspect of the present invention, a method for identifying multi-level residential-recreation spatial structures based on an overlapping community detection algorithm is provided, the method comprising: S1. Identify the residential-recreational function connection based on mobile phone signaling data, construct a residential-recreational network based on the residential-recreational function connection, and in the residential-recreational network, the original node corresponds to a geographic spatial unit; S2. Based on the residential-recreational network, the Linkcomm algorithm is used to transform it into a weighted edge network. In the weighted edge network, the node set corresponds to the original edge set in the residential-recreational network. The edge set in the weighted edge network includes the original edge pairs in the residential-recreational network that have common nodes. S3. Based on the weighted edge network, the Louvain community detection algorithm is used to identify the edge community structure in the weighted edge network. S4. Map the edge community structure back to the residential-recreational network, obtain the node community structure corresponding to each original node, and identify the original nodes with overlapping relationships based on the node community structure; the node community structure includes multiple node communities, and each community includes multiple original nodes. S5. Construct a hypergraph based on the aforementioned node community structure. Based on the hypergraph, use the Louvain community detection algorithm to identify the Nth layer community structure. In the hypergraph, each node corresponds to a node community, and edges are constructed based on community functional connections. The community functional connections are calculated based on the residential-recreational functional connections between all original nodes within the corresponding node community. For original nodes with overlap, the number of times they participate in the calculation of the community functional connections is equal to the number of overlaps. S6. Eliminate communities in the Nth layer community structure whose community size is smaller than the preset value. For the remaining communities, iterate through S5 until the community structure no longer changes, or there is only one community in the community structure. End the iteration to obtain the multi-layer residential-recreation space structure of the city.

[0007] As a preferred technical solution, the method for identifying the relationship between residential and recreational functions is as follows: Based on the aforementioned mobile signaling data, the user's dwell time at the base station within a preset time period is obtained, and the functional attributes of the corresponding base station are identified based on the dwell time. The functional attributes include residence, place of employment, and recreational location. The method for identifying the recreational location is as follows: obtaining mobile signaling data on non-holidays and weekends, obtaining the user's dwell time at base stations in non-residence and non-employment locations based on the mobile signaling data, and if the dwell time meets the recreational location criteria, then the functional attribute of the corresponding base station is recreational location. Within the statistical period, the frequency of residential-recreational connections at the base station level is obtained based on the mobile signaling data, and the frequency of residential-recreational connections at the base station level is mapped to geospatial units through spatial connections to realize the residential-recreational functional connections from residential space units to recreational space units; the geospatial unit includes at least one base station.

[0008] As a preferred technical solution, for base station A, whose functional attribute is residential, and base station B, whose functional attribute is recreation, the method for obtaining the base station-level residential-recreational connection frequency is as follows: within the statistical period, obtain the total number of people residing at base station A and engaging in recreational activities at base station B, and the total number of people is the base station-level residential-recreational connection frequency between base station A and base station B.

[0009] As a preferred technical solution, the method for constructing the residential-recreation network is as follows: Based on the aforementioned residential-recreational function connection, the frequency of residential-recreational connections for all users in geographic spatial unit pairs with residential-recreational function connections is obtained, and the frequency of residential-recreational connections is used as the connection strength of the corresponding geographic spatial unit. Using geospatial units as the original nodes, directed edges are constructed for pairs of original nodes that have residential-recreational functional connections. The edge weight of the directed edge is the connection strength, and a directed network is constructed. The directed network is transformed into an undirected network, which is the residential-recreational network; and in the residential-recreational functional relationship, the edge weight is the sum of the weights of the corresponding directed edges and the bidirectional edges.

[0010] As a preferred technical solution, in the weighted edge network, the edge weight of each side is calculated based on the similarity of the corresponding original edge pairs, and its expression is: , in, This represents the weighted edge network node corresponding to the original edge of the original node i and the original node j; This represents the weighted edge network node corresponding to the original edge of the original node k and the original node j; Represents the set of adjacent nodes of the original node i in the residential-recreational network; This represents the set of adjacent nodes of the original node k in the residential-recreational network; Representing the original edges in a residential-recreational network Edge weights; Representing the original edges in a residential-recreational network Edge weights.

[0011] As a preferred technical solution, the method for identifying the edge community structure is as follows: S31. Assign each node in the weighted edge network to an independent edge community; S32. For each node, the calculator calculates the modularity gain brought by moving to the neighboring edge community, and selects candidate movement schemes with positive modularity gain. S33. In the candidate moving schemes, select the corresponding node to migrate to the edge community where the modularity gain is the maximum value; S34. After the migration is completed, nodes belonging to the same side community will be aggregated into super nodes to build a new weighted edge network. S35. Iterate through S31 to S34 until the change in modularity gain is less than the change threshold, and output the final weighted edge network as the edge community structure.

[0012] As a preferred technical solution, the method for calculating the modularity is as follows: , in, Indicates modularity; This represents the total weight of the edge network, and , This represents the similarity weight between nodes u and v in a weighted edge network; Represents the set of nodes in a weighted edge network; Indicates precision parameters; and Let represent the weighted degrees of nodes u and v in the weighted edge network, respectively, and have . , ; Represents the community membership function; This represents the community corresponding to node u in a weighted edge network; This represents the community corresponding to node v in a weighted edge network.

[0013] As a preferred technical solution, the method for obtaining the node community structure is as follows: For the acquisition Border-like communities Obtain the original set of edges corresponding to it in the residential-recreational network. ; Based on the aforementioned original edge set The original nodes of all the original edges included in the node community are the building nodes. .

[0014] As a preferred technical solution, in S5, the original node with overlap means that for the original edge b belonging to multiple different node communities, its corresponding original node also belongs to multiple different node communities; the node community containing the original node with overlap is called the overlapping community.

[0015] As a preferred technical solution, in the hypergraph, the edge weight is the sum of the connection strengths of all original nodes between the two node communities, that is: , in, This represents the edge weight of the edge connecting node i and node j in the hypergraph; Represents the set of nodes in node community i; Describes the set of nodes in node community j; This represents the strength of the connection between the original node u and the original node v.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1) This invention acquires dynamic data information based on dynamic finger signaling data, constructs original network information with geospatial units as nodes based on this dynamic data information, and introduces the LinkComm algorithm to promote the edges in the original network to nodes in the edge graph network. The edges in the edge graph network are constructed based on the similarity of the original edges. In the whole method, the method of clustering edges is performed. Compared with clustering nodes, the method provided by this invention allows nodes to belong to multiple communities through the affiliation relationship of different communities of the edges, ensuring that the nodes have natural overlapping characteristics, thereby making the final identification result more accurate.

[0017] 2) This invention provides an iterative hierarchical clustering method. Specifically, the communities identified in the previous round are regarded as supernodes, and the functional connection strength between communities is used as edge weights to construct a hypergraph network at the next higher level. The LinkComm algorithm is then applied again to identify the community structure and obtain the community structure at the next higher level. This process is iterated until the number of communities is 1 or remains unchanged. This method automatically and data-drivenly reveals the multi-level hierarchical structure of urban space from fine-grained communities to coarse-grained functional clusters through a recursive aggregation process of community → supernode → community, thereby achieving effective identification of urban hierarchy. Attached Figure Description

[0018] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the first-layer community structure in an embodiment of the present invention; Figure 3 This is a schematic diagram of the second-layer community structure in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] To address the problems existing in current technologies, this invention provides a multi-level residential-recreational spatial structure identification method based on mobile phone signaling data and an overlapping community detection algorithm. First, based on mobile phone signaling data, the actual spatiotemporal behavior of residents is extracted, and a weighted network is constructed with spatial units as nodes and residential-recreational functional connections as edges. Then, the LinkComm overlapping community detection algorithm is used to cluster the nodes of this network, identifying the community category to which each spatial unit belongs and determining spatial units with multiple community affiliations. Furthermore, the resulting community structure is hierarchically clustered to extract a multi-scale hierarchical system of urban space. Specifically, the process is as follows: Figure 1 As shown, the detailed steps are as follows: S1. Identify residential-recreational function connections based on mobile phone signaling data, and construct a residential-recreational network based on these connections. ,in, This represents the set of nodes formed by each spatial unit. It represents the strength of residential-recreational connections between spatial units, is the set of edges in the residential-recreational network, and in the residential-recreational network, the original node corresponds to a geographic spatial unit.

[0021] S11. Obtain the duration of a user's stay at a base station within a preset time period based on mobile phone signaling data, and identify the functional attributes of the corresponding base station based on the duration of stay.

[0022] In this embodiment, the functional attributes include residence, place of employment, and recreation, specifically: The identification criteria for residential base stations are as follows: , Indicates that the user is at the base station Total nighttime stay This indicates the threshold for identifying the place of residence.

[0023] The identification criteria for the employment location base station are as follows: , in, Indicates that the user is at the base station Total duration of daytime stay Thresholds for identifying the place of employment.

[0024] The method and conditions for determining a recreational location are as follows: Mobile phone signaling data from non-holiday and weekend periods are obtained; based on this signaling data, the duration of a user's stay at a base station in a location other than their residence or workplace is obtained; if the duration of stay meets the recreational location criteria, the corresponding base station's functional attribute is designated as a recreational location. In this embodiment, the recreational location criteria are set as follows: , Indicates that the user is at the base station The length of stay, and .

[0025] S12. Within the statistical period, the frequency of residential-recreational connections at the base station level is obtained based on mobile signaling data, and the frequency of residential-recreational connections at the base station level is mapped to geospatial units through spatial connectivity to obtain residential spatial units. To the recreation space unit The residential-recreational functions are linked, and it needs to be clarified that the geospatial unit includes at least one base station.

[0026] Taking base station A, whose functional attribute is residential, and base station B, whose functional attribute is recreational, as examples, the method for obtaining the base station-level residential-recreational connection frequency is as follows: within the statistical period, obtain the total number of people residing at base station A and engaging in recreational activities at base station B. The total number of people is the base station-level residential-recreational connection frequency between base station A and base station B.

[0027] S13. Obtain all users' geographic spatial units with residential-recreational function relationships based on residential-recreational function relationships. The frequency of residential-recreational connections is used as the connection strength of the corresponding geographic spatial unit. .

[0028] S14. Using geographic spatial units as the original nodes For original node pairs with residential-recreational functional connections, directed edges are constructed, with the edge weight representing the connection strength, thus building a directed network. ,in, .

[0029] S15. Transform the directed network into an undirected network, which is a residential-recreational network. Furthermore, in the residential-recreation function relationship, the edge weight is the sum of the weights of the corresponding directed edges in both directions, i.e.: , Indicates from geographic spatial unit To geospatial units The edge weight of the directed edge. Indicates from geographic spatial unit To geospatial units The edge weight of the directed edge.

[0030] S2. Based on the resident-recreation network, it is transformed into a weighted edge network using the Linkcomm algorithm. The node set in the weighted edge network... The original edge set in the corresponding residential-recreational network ,Right now In a weighted edge network, the edge set includes the original edge pairs that share a common node in the residential-recreational network, i.e. , and All represent the original edges. and All represent edges in the weighted edge network corresponding to the original edges.

[0031] In a weighted edge network, the edge weight of each edge is calculated based on the corresponding original edge pair using weighted Jaccard similarity, and its expression is: , in, This represents the weighted edge network node corresponding to the original edge of the original node i and the original node j; This represents the weighted edge network node corresponding to the original edge of the original node k and the original node j; Represents the set of adjacent nodes of the original node i in the residential-recreational network; This represents the set of adjacent nodes of the original node k in the residential-recreational network; Representing the original edges in a residential-recreational network Edge weights; Representing the original edges in a residential-recreational network Edge weights.

[0032] S3. Based on weighted edge networks, use the Louvain community detection algorithm to identify edge community structures in weighted edge networks. .

[0033] S31. Assign each node in the weighted edge network to an independent edge community.

[0034] S32. For each node, the modularity gain brought by moving the calculator to its neighboring edge community. And select candidate move schemes with positive modularity gain. The method for calculating modularity is as follows: , in, Indicates modularity; This represents the total weight of the edge network, and , This represents the similarity weight between nodes u and v in a weighted edge network; Represents the set of nodes in a weighted edge network; Indicates precision parameters; and Let represent the weighted degrees of nodes u and v in the weighted edge network, respectively, and have . , ; Represents the community membership function; This represents the community corresponding to node u in a weighted edge network; This represents the community corresponding to node v in a weighted edge network.

[0035] S33. Select the candidate moving scheme to migrate the corresponding node to the edge community where the modularity gain is the maximum value.

[0036] S34. After the migration is completed, nodes belonging to the same side community are aggregated into super nodes to construct a new weighted edge network. The edge weights between side communities are transformed into edge weights between super nodes. In this embodiment, the edge weight of a super node is the weighted sum of the edge weights between the nodes included in the super node.

[0037] S35. Iterate through S31~S34 until the change in modularity gain is less than the change threshold, then output the final weighted edge network as the edge community structure. .

[0038] S4. Map the edge community structure back to the residential-recreational network, obtain the node community structure corresponding to each original node, and identify the original nodes with overlap based on the node community structure. The node community structure includes multiple node communities, and each community includes multiple original nodes.

[0039] S5, Node-based Community Structure Building a hypergraph Based on the hypergraph, the Louvain community detection algorithm was used to identify the Nth layer community structure. ,and .

[0040] In this step, the original nodes with overlapping properties refer to the original nodes that also belong to multiple different node communities for the original edge b. The corresponding node communities that contain original nodes with overlapping properties are called overlapping communities.

[0041] In a hypergraph, each node corresponds to a node community, which can be represented as: Edges are constructed based on community functional connections, which are the sum of residential-recreational functional connections between all original nodes within the corresponding node's community. For original nodes with overlap, the number of times they participate in community functional connection calculations is equal to the number of overlaps. In the hypergraph, the edge weight is the sum of the connection strengths between all original nodes in the two node communities, i.e.: , in, This represents the edge weight of the edge connecting node i and node j in the hypergraph; Represents the set of nodes in node community i; Describes the set of nodes in node community j; This represents the strength of the connection between the original node u and the original node v.

[0042] S6. Remove communities with a community size less than 1 (i.e., the number of original nodes contained in the community) from the Nth layer community structure. For the remaining communities, iterate through S5 until the community structure no longer changes, or there is only one community in the community structure. End the iteration to obtain the multi-layered residential-recreational space structure of the city, which can be represented as: .

[0043] To verify the feasibility of the method provided by this invention, this embodiment takes a city as an example, dividing it into 1km grid units. Grids with a cumulative resident population of less than 5% of the total resident population are removed to obtain geospatial units. Then, the method proposed in this invention is applied to identify the multi-level residential-recreational spatial structure. The identification results are then mapped. First, the spatial units are divided into two categories: one with overlapping communities (i.e., the number of identified communities is greater than 1), and the other with non-overlapping communities. For non-overlapping community units, the spatial units are colored according to the community structure classification results. For overlapping community units, they are uniformly colored red. Then, the spatial units of each community structure are extracted, and the standard deviation ellipse of the spatial units is drawn according to their location. In this process, the structure after one clustering (i.e., one execution of S5~S6) is drawn as follows. Figure 2 The structure shown in the diagram can be considered the lowest level of the community structure. Figure 2 Based on this, further clustering (i.e., executing S5~S6 again) yields... Figure 3 The structure shown is a second-layer community structure; in the context of... Figure 3 After the structural re-clustering (executing S5~S6 again), the number of communities in the final structure is either 1 or remains unchanged. Therefore, the final multi-level residential-recreation space structure of the city can be obtained as follows. Figure 3 As shown in the diagram, the method provided by this invention is feasible.

[0044] Furthermore, the present invention provides an electronic device including a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). Various programs and data required for device operation can also be stored in the RAM. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0045] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0046] The processing unit executes the various methods and processes described above, such as methods S1 to S6. For example, in some embodiments, methods S1 to S6 may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of methods S1 to S6 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S6 by any other suitable means (e.g., by means of firmware).

[0047] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0048] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0049] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0050] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for identifying multi-level residential-recreation spatial structures based on an overlapping community detection algorithm, characterized in that the method... include: S1. Identify the residential-recreational function connection based on mobile phone signaling data, construct a residential-recreational network based on the residential-recreational function connection, and in the residential-recreational network, the original node corresponds to a geographic spatial unit; S2. Based on the residential-recreational network, the Linkcomm algorithm is used to transform it into a weighted edge network. In the weighted edge network, the node set corresponds to the original edge set in the residential-recreational network. The edge set in the weighted edge network includes the original edge pairs in the residential-recreational network that have common nodes. S3. Based on the weighted edge network, the Louvain community detection algorithm is used to identify the edge community structure in the weighted edge network. S4. Map the edge community structure back to the residential-recreational network, obtain the node community structure corresponding to each original node, and identify the original nodes with overlapping relationships based on the node community structure; the node community structure includes multiple node communities, and each community includes multiple original nodes. S5. Construct a hypergraph based on the aforementioned node community structure. Based on the hypergraph, use the Louvain community detection algorithm to identify the Nth layer community structure. In the hypergraph, each node corresponds to a node community, and edges are constructed based on community functional connections. The community functional connections are calculated based on the residential-recreational functional connections between all original nodes within the corresponding node community. For original nodes with overlap, the number of times they participate in the calculation of the community functional connections is equal to the number of overlaps. S6. Eliminate communities in the Nth layer community structure whose community size is smaller than the preset value. For the remaining communities, iterate through S5 until the community structure no longer changes, or there is only one community in the community structure. End the iteration to obtain the multi-layer residential-recreation space structure of the city.

2. The method for identifying multi-level residential-recreation spatial structure based on an overlapping community detection algorithm according to claim 1, characterized in that, The method for identifying the residential-recreation function relationship is as follows: Based on the mobile signaling data, the user's dwell time at the base station within a preset time period is obtained. Based on the dwell time, the functional attributes of the corresponding base station are identified. The functional attributes include residence, place of employment, and recreational place. The method for identifying the recreational place is as follows: obtain mobile signaling data on non-holidays and weekends, obtain the user's dwell time at base stations in non-residence and non-employment locations based on the mobile signaling data, and if the dwell time meets the recreational place criteria, the functional attribute of the corresponding base station is recreational place. Within the statistical period, the frequency of residential-recreational connections at the base station level is obtained based on the mobile signaling data, and the frequency of residential-recreational connections at the base station level is mapped to geospatial units through spatial connections to realize the residential-recreational functional connections from residential space units to recreational space units; the geospatial unit includes at least one base station.

3. The method for identifying multi-level residential-recreation spatial structure based on an overlapping community detection algorithm according to claim 2, characterized in that, For base station A, whose functional attribute is residential, and base station B, whose functional attribute is recreation, the method for obtaining the base station-level residential-recreational connection frequency is as follows: within the statistical period, obtain the total number of people residing at base station A and engaging in recreational activities at base station B. The total number of people is the base station-level residential-recreational connection frequency between base station A and base station B.

4. The method for identifying multi-level residential-recreation spatial structures based on an overlapping community detection algorithm according to claim 1, characterized in that, The method for constructing the residential-recreation network is as follows: Based on the aforementioned residential-recreational function connection, the frequency of residential-recreational connections for all users in geographic spatial unit pairs with residential-recreational function connections is obtained, and the frequency of residential-recreational connections is used as the connection strength of the corresponding geographic spatial unit. Using geospatial units as the original nodes, directed edges are constructed for pairs of original nodes that have residential-recreational functional connections. The edge weight of the directed edge is the connection strength, and a directed network is constructed. The directed network is transformed into an undirected network, which is the residential-recreational network; and in the residential-recreational functional relationship, the edge weight is the sum of the weights of the corresponding directed edges and the bidirectional edges.

5. A method for identifying multi-level residential-recreation spatial structures based on an overlapping community detection algorithm according to claim 1, characterized in that, In the weighted edge network described above, the edge weight of each side is calculated based on the similarity of the corresponding original edge pair, and its expression is: ， in, This represents the weighted edge network node corresponding to the original edge of the original node i and the original node j; This represents the weighted edge network node corresponding to the original edge of the original node k and the original node j; Represents the set of adjacent nodes of the original node i in the residential-recreational network; This represents the set of adjacent nodes of the original node k in the residential-recreational network; Representing the original edges in a residential-recreational network Edge weights; Representing the original edges in a residential-recreational network Edge weights.

6. The method for identifying multi-level residential-recreation spatial structure based on an overlapping community detection algorithm according to claim 1, characterized in that, The method for identifying the aforementioned edge community structure is as follows: S31. Assign each node in the weighted edge network to an independent edge community; S32. For each node, the calculator calculates the modularity gain brought by moving to the neighboring edge community, and selects candidate movement schemes with positive modularity gain. S33. In the candidate moving schemes, select the corresponding node to migrate to the edge community where the modularity gain is the maximum value; S34. After the migration is completed, nodes belonging to the same side community will be aggregated into super nodes to build a new weighted edge network. S35. Iterate through S31 to S34 until the change in modularity gain is less than the change threshold, and output the final weighted edge network as the edge community structure.

7. A method for identifying multi-level residential-recreation spatial structures based on an overlapping community detection algorithm according to claim 6, characterized in that, The method for calculating the modularity is as follows: ， in, Indicates modularity; This represents the total weight of the edge network, and , This represents the similarity weight between nodes u and v in a weighted edge network; Represents the set of nodes in a weighted edge network; Indicates precision parameters; and Let represent the weighted degrees of nodes u and v in the weighted edge network, respectively, and have . , ; Represents the community membership function; This represents the community corresponding to node u in a weighted edge network; This represents the community corresponding to node v in a weighted edge network.

8. A method for identifying multi-level residential-recreation spatial structures based on an overlapping community detection algorithm according to claim 1, characterized in that, The method for obtaining the node community structure is as follows: For the acquisition Border-like communities Obtain the original set of edges corresponding to it in the residential-recreational network. ; Based on the aforementioned original edge set The original nodes of all the original edges included in the node community are the building nodes. .

9. A method for identifying multi-level residential-recreation spatial structures based on an overlapping community detection algorithm according to claim 1, characterized in that, In S5, an original node with overlap means that for an original edge b belonging to multiple different node communities, its corresponding original node also belongs to multiple different node communities; a node community containing original nodes with overlap is an overlapping community.

10. A method for identifying multi-level residential-recreation spatial structure based on an overlapping community detection algorithm according to claim 1, characterized in that, In the hypergraph described above, the edge weight is the sum of the connection strengths of all original nodes between the two node communities, which is: ， in, This represents the edge weight of the edge connecting node i and node j in the hypergraph; Represents the set of nodes in node community i; Describes the set of nodes in node community j; This represents the strength of the connection between the original node u and the original node v.

Images