Commuting area identification method, device, and equipment and storage medium

Abstract

This application provides a method, apparatus, device, and storage medium for commuter area identification, relating to the field of big data processing technology. It integrates commuter traffic volume data, commuter time data, and housing cost data to calculate a comprehensive commuter connection weight, quantifying residents' actual behavioral trade-offs between commuting time burden and housing cost advantages in daily life. After preliminary clustering to obtain candidate commuter communities, a spatial consistency correction step is introduced to deeply refine the clustering results to obtain continuous commuter communities, ensuring the continuity of the delineated commuter area boundaries. Finally, the internal commuter retention rate, which measures the interpretability of commuter flows, and the spatial continuity, which measures geographical morphological characteristics, are jointly used as the target fitness. An optimization algorithm dynamically filters and proliferates various weight parameter combinations in each iteration, seeking the global optimal solution among mutually constraining objective functions, improving identification accuracy and structural rationality, and making the commuter area delineation results accurate and reliable.

Description

Technical Field

[0001] This application relates to the field of big data processing technology, and in particular to commuter area identification methods, apparatus, devices, and storage media. Background Technology

[0002] A commuting zone is a functionally significant aggregated area that combines basic spatial units based on the spatial characteristics of commuting activities. The scientific identification of commuting zones has important theoretical and practical value for transportation planning, public service allocation, and land use optimization.

[0003] In related technologies, commuter zone identification methods typically construct spatial connection networks based on commuter traffic between residences and workplaces, and delineate commuter zones through interaction intensity metrics or graph theory indices. However, most methods rely excessively on commuter traffic volume and spatial distance, lacking consideration of deeper factors such as population attributes, employment structure, and housing costs, resulting in low accuracy in commuter zone delineation. Summary of the Invention

[0004] The main objective of this application is to propose a commuter area identification method, apparatus, device, and storage medium to improve the accuracy of commuter area division results.

[0005] To achieve the above objectives, a first aspect of this application proposes a commuter area identification method, comprising: Obtain multiple commuting weight parameter sets to initialize the parameter population; The optimization algorithm iteratively optimizes the parameter population. In each iteration, the target fitness is calculated for each commuting weight parameter group in the current population. If the termination condition is not met, the optimization algorithm generates a new parameter population based on the target fitness and enters the next iteration. If the termination condition is met, the target weight parameter group and its corresponding continuous commuting community are output as the commuting area division result. The calculation of the target fitness for each of the commuting weight parameter groups in the current population includes the following steps: calculating the comprehensive commuting connection weight between each pair of spatial units based on the commuting flow scale data, commuting time data, housing cost data, and the commuting weight parameter groups of each spatial unit in the target area; constructing a weighted commuting network based on the comprehensive commuting connection weights; performing preliminary clustering on the weighted commuting network to obtain candidate commuting communities; performing spatial consistency correction on the candidate commuting communities to obtain continuous commuting communities; and calculating the internal commuting retention rate and spatial continuity as the target fitness based on the current continuous commuting communities.

[0006] In some embodiments, calculating the comprehensive commuting connection weight between any two spatial units based on commuting traffic volume data, commuting time data, housing cost data, and the commuting weight parameter set for each spatial unit within the target area includes: Based on the commuting flow data, commuting time data, and housing cost data of each pair of spatial units, the basic commuting connection strength, commuting time impedance characteristics, and housing price difference constraint characteristics are calculated, and the two spatial units are defined as a residential unit and an employment unit, respectively. The comprehensive commuting connection weight is obtained by weighting the basic commuting connection strength, the commuting time impedance characteristics, and the housing price difference constraint characteristics using the commuting weight parameter group.

[0007] In some embodiments, calculating the basic commuter connectivity strength, commuter time impedance characteristics, and housing price difference constraint characteristics based on the commuter flow volume data, commuter time data, and housing cost data for each pair of spatial units includes: The number of commuters from the residential unit to the employment unit is obtained from the commuter traffic volume data, the number of commuters is standardized, and the basic commuter connection intensity is calculated. A first penalty parameter is obtained based on the commuting time data and the first attenuation parameter, and the commuting time impedance characteristic is obtained by multiplying the first penalty parameter and the basic commuting connection strength. The housing price difference constraint features are derived from the commuting time data and the housing cost data.

[0008] In some embodiments, obtaining the housing price difference constraint features based on the commuting time data and the housing cost data includes: The second penalty parameter is obtained based on the commuting time data and the second attenuation parameter; The housing price difference constraint feature is obtained by obtaining the first housing cost of the residential unit and the second housing cost of the employment unit from the housing cost data, obtaining the cost difference between the second housing cost and the first housing cost, obtaining the maximum value between the cost difference and zero, calculating the quotient of the maximum value and the first housing cost, and calculating the product of the quotient and the second penalty parameter.

[0009] In some embodiments, the weighted commuting network uses the spatial unit as a node and the corresponding comprehensive commuting connection weight as the edge weight of the directed edge. The preliminary clustering of the weighted commuting network to obtain candidate commuting communities includes: The total weight and the weighted degree corresponding to each node are obtained based on the edge weights. The nodes are selected in pairs to form node pairs. A first intermediate parameter is obtained by multiplying the corresponding weighted degree by the total weight. The first intermediate parameter is subtracted from the comprehensive commuting connection weight of the node pair to obtain a second intermediate parameter. A community identifier parameter is set to indicate whether the node pairs are located in the same community. The node pair parameter is obtained by multiplying the community identifier parameter and the second intermediate parameter. All the node pair parameters are accumulated to obtain an accumulated parameter. The modularity objective function is obtained by quotienting the accumulated parameter and the total weight. With the objective function of maximizing modularity as the goal, the community identifier parameters are solved to obtain the target community identifier, and the weighted commuting network is clustered based on the target community identifier to obtain the candidate commuting communities.

[0010] In some embodiments, the step of spatially consistent correcting the candidate commuting communities to obtain continuous commuting communities includes: Traverse the commuter areas in the candidate commuter communities, determine at least one continuous spatial entity in each commuter area, and determine the continuous spatial entity with the largest area as the largest continuous spatial entity. Identify adjacent units that are inside the largest continuous spatial entity but are assigned to other commuter areas, and merge the adjacent units into the commuter area to which the largest continuous spatial entity belongs. Identify isolated units corresponding to the commuting area. If the area of ​​the isolated unit is less than a preset area threshold, merge the isolated unit into the adjacent commuting area.

[0011] In some embodiments, calculating internal commuting retention rate and spatial continuity as target fitness based on the current continuous commuting community includes: For the commuter zone in the continuous commuting community, calculate the number of first commuters whose starting point and ending point are both in the commuter zone, and the number of second commuters whose starting point or ending point is either in the commuter zone. Based on the number of first commuters and the number of second commuters, obtain the internal commuting retention rate, so as to maximize the internal commuting retention rate as the internal commuting retention rate target. For the commuter areas in the continuous commuter community, the adjacency value and the corresponding commuter area identifier of the spatial unit are obtained, and the spatial continuity parameter is obtained based on the adjacency value and the commuter area identifier, with maximizing the spatial continuity parameter as the spatial continuity objective; The internal commuting retention rate and the spatial continuity parameter are combined as the target fitness.

[0012] To achieve the above objectives, a second aspect of this application provides a commuter area identification device, comprising: Parameter acquisition module: used to acquire multiple commuting weight parameter sets to initialize the parameter population; The optimization iteration module is used to iteratively optimize the parameter population using an optimization algorithm. In each iteration, the target fitness is calculated for each commuting weight parameter group in the current population. If the termination condition is not met, the optimization algorithm generates a new parameter population based on the target fitness and enters the next iteration. If the termination condition is met, the target weight parameter group and its corresponding continuous commuting community are output as the commuting area division result. The calculation of the target fitness for each of the commuting weight parameter groups in the current population includes the following steps: calculating the comprehensive commuting connection weight between each pair of spatial units based on the commuting flow scale data, commuting time data, housing cost data, and the commuting weight parameter groups of each spatial unit in the target area; constructing a weighted commuting network based on the comprehensive commuting connection weights; performing preliminary clustering on the weighted commuting network to obtain candidate commuting communities; performing spatial consistency correction on the candidate commuting communities to obtain continuous commuting communities; and calculating the internal commuting retention rate and spatial continuity as the target fitness based on the current continuous commuting communities.

[0013] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0014] To achieve the above objectives, a fourth aspect of the present application provides a storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0015] The commuter area identification method, apparatus, device, and storage medium proposed in this application initialize a parameter population by acquiring multiple commuter weight parameter groups; iteratively optimizes the parameter population using an optimization algorithm. In each iteration, a target fitness is calculated for each commuter weight parameter group in the current population. If the termination condition is not met, the optimization algorithm generates a new parameter population based on the target fitness and enters the next iteration; if the termination condition is met, the target weight parameter group and its corresponding continuous commuter community are output as the commuter area division result. The calculation of the target fitness for each commuter weight parameter group in the current population includes the following steps: calculating the comprehensive commuter connection weight between each pair of spatial units based on commuter traffic volume data, commuter time data, housing cost data, and commuter weight parameter groups within the target area; constructing a weighted commuter network based on the comprehensive commuter connection weight; performing preliminary clustering on the weighted commuter network to obtain candidate commuter communities; performing spatial consistency correction on the candidate commuter communities to obtain continuous commuter communities; and calculating the internal commuter retention rate and spatial continuity as the target fitness based on the current continuous commuter community. This application's embodiments calculate comprehensive commuter connection weights by integrating commuter flow scale data, commuter time data, and housing cost data. This quantifies residents' actual behavioral trade-offs between commuter time burden and housing cost advantages in daily life, making the spatial connection modeling more consistent with real socioeconomic logic and significantly improving the accuracy of the identification results. Secondly, after initially clustering candidate commuter communities, a spatial consistency correction step is introduced to deeply correct the clustering results to obtain continuous commuter communities. This effectively eliminates fragmented patches that do not conform to geographical norms, ensuring the continuity of the delineated commuter zone boundaries. Finally, the internal commuter retention rate, which measures the interpretability of commuter flows, and the spatial continuity, which measures geographical morphological characteristics, are jointly used as the target fitness. Through optimization algorithms, multiple weight parameter combinations are dynamically filtered and proliferated in each iteration, thereby finding the global optimal solution among mutually constraining objective functions. This further improves the identification accuracy and structural rationality, making the final commuter zone delineation results more accurate and reliable. Attached Figure Description

[0016] Figure 1 This is a flowchart of the commuter area identification method provided in the embodiments of this application.

[0017] Figure 2 This is a flowchart of calculating the target fitness for each commuting weight parameter group in the current population, provided in an embodiment of this application.

[0018] Figure 3 This is a flowchart provided in an embodiment of the present application for calculating the comprehensive commuting connection weight between any two spatial units based on commuting traffic volume data, commuting time data, housing cost data, and commuting weight parameter groups of each spatial unit within the target area.

[0019] Figure 4 This is a flowchart provided in an embodiment of the present application for calculating basic commuting connection strength, commuting time impedance characteristics, and housing price difference constraint characteristics based on commuting flow volume data, commuting time data, and housing cost data of each pair of spatial units.

[0020] Figure 5 This is a flowchart illustrating the characteristics of housing price difference constraints derived from commuting time data and housing cost data, provided in an embodiment of this application.

[0021] Figure 6 This is a flowchart of obtaining candidate commuter communities by performing preliminary clustering of a weighted commuter network, as provided in an embodiment of this application.

[0022] Figure 7 This is a flowchart of obtaining a continuous commuting community by performing spatial consistency correction on candidate commuting communities, as provided in an embodiment of this application.

[0023] Figure 8 This is a flowchart provided in an embodiment of the present application for calculating internal commuting retention rate and spatial continuity as target fitness based on the current continuous commuting community.

[0024] Figure 9 This is an example of commuter area identification provided in the embodiments of this application.

[0025] Figure 10 This is a structural block diagram of a commuter area identification device provided in another embodiment of this application.

[0026] Figure 11 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0028] It should be noted that although functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0030] Commuter zones are functionally integrated areas formed by combining basic spatial units based on the spatial characteristics of commuting activities. This process primarily involves two core issues: first, the spatial clustering of residential and work areas; and second, the construction and characterization of the spatial connections between them. As one of the most important spatial behaviors in daily life, the scientific identification of commuter zones has significant theoretical and practical value for transportation planning, public service allocation, and land use optimization.

[0031] In related technologies, commuter zone identification methods primarily rely on commuter flow between residential and workplace locations to characterize the strength of spatial connections. In spatial connection modeling, the research unit typically refers to a spatial unit, which can be an administrative unit such as a city or district, or a more refined grid unit. The finer the division of the research unit, the better it is for revealing the internal structure of the city and further aggregating it into several areas with commuter function characteristics. Spatial connections are usually based on the number of commuters or their flow, and specific algorithms are used to aggregate spatial units. Therefore, commuter zone identification is essentially a process of functionally aggregating basic units based on the dual attributes of residential and workplace areas through spatial connections.

[0032] Regarding the spatial scale of commuter zone identification, most research in related technologies is based on administrative scales, such as using county-level units in different countries as the basic research unit. In this type of research, commuter zones are usually divided within established administrative boundaries to ensure the spatial consistency of the identification results and to maintain consistency with the administrative management system. In addition to administrative scales, some studies have begun to explore using more refined spatial units, such as using point-of-interest data to identify urban functional areas, in order to improve the accuracy of spatial characterization.

[0033] From an algorithmic perspective, research on commuter zone identification in related technologies mainly includes weighted network analysis, hierarchical clustering analysis, evolutionary algorithms, and various hybrid algorithms. These methods typically construct a spatial connection network based on commuter traffic between residences and workplaces, and delineate commuter zones using interaction strength metrics or graph theory indices. Some studies also incorporate distance constraints or threshold rules to further optimize the identification results.

[0034] However, commuter zone identification methods in related technologies still have the following significant limitations: First, the spatial analysis scale is too coarse, masking internal differences. Most methods are limited by administrative boundaries (such as districts, counties, and municipal units), resulting in high aggregation of commuter data during the data preprocessing stage. This approach ignores the heterogeneity within administrative units, causing the model to fail to identify multiple commuter cores and their complex interactions within a single administrative region. Second, the modeling indicators are too simplistic, neglecting socioeconomic driving mechanisms. The models rely excessively on commuter flow scale and spatial distance, lacking consideration of deeper factors such as population attributes, employment structure, and housing costs. In reality, residents often weigh commuting time against housing costs; ignoring such economic constraints leads to a lack of real-world explanatory power in the identification results. Third, the identification process lacks structural constraints, making spatial morphology easily fragmented. These algorithms often employ a one-time partitioning scheme using a single-layer network, lacking a mechanism for identifying the substructures within commuter zones. This results in outputs that often only represent generalized boundaries, and without spatial continuity constraints, they are prone to spatial fragmentation, isolated spots, and other phenomena that do not conform to geographical logic. Fourth, the lack of a global optimization mechanism makes it difficult to balance multi-dimensional indicators. When defining boundaries, these methods often fail to simultaneously consider both "consistency of commuting behavior" and "continuity of geographic space," lacking a systematic evaluation and screening framework that can find the optimal balance among multiple objective functions.

[0035] Based on this, embodiments of this application provide a commuter area identification method, apparatus, device, and storage medium. By fusing commuter flow scale data, commuter time data, and housing cost data to calculate comprehensive commuter connection weights, it quantifies the real behavioral trade-offs residents make between commuting time burden and housing cost advantages in daily life. This makes the modeling of spatial connections more consistent with real socioeconomic logic, significantly improving the accuracy of the identification results. Secondly, after obtaining candidate commuter communities through preliminary clustering, a spatial consistency correction step is introduced to deeply correct the clustering results to obtain continuous commuter communities. This effectively eliminates fragmented patches that do not conform to geographical norms, ensuring the continuity of the delineated commuter area boundaries. Finally, the internal commuter retention rate, which measures the interpretability of commuter flow, and the spatial continuity, which measures geographical morphological characteristics, are jointly used as the target fitness. Through optimization algorithms, multiple weight parameter combinations are dynamically filtered and proliferated in each iteration, thereby finding the global optimal solution among mutually constraining objective functions. This further improves the identification accuracy and structural rationality, making the final commuter area delineation results more accurate and reliable.

[0036] This application provides a commuter area identification method, apparatus, device, and storage medium, which are specifically described through the following embodiments. First, the commuter area identification method in this application embodiment is described.

[0037] The commuting area identification method provided in this application relates to the field of big data processing technology. This method can be applied to a terminal, a server, or a computer program running on either the terminal or the server. For example, the computer program can be a native program or software module in an operating system; it can be a native application (APP), i.e., a program that needs to be installed in the operating system to run, such as a client that supports commuting area identification, i.e., a program that only needs to be downloaded to a browser environment to run; or it can be a small program that can be embedded in any APP. In short, the above-mentioned computer program can be any form of application, module, or plugin. The terminal communicates with the server via a network. This commuting area identification method can be executed by the terminal or the server, or by the terminal and the server working together.

[0038] In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, or smartwatch, etc. The server can be a standalone server, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms; it can also be a service node in a blockchain system, where the service nodes form a peer-to-peer (P2P) network. The P2P protocol is an application layer protocol running on top of the Transmission Control Protocol (TCP). The terminal and server can connect via Bluetooth, Universal Serial Bus (USB), or a network, etc., and this embodiment does not impose any limitations.

[0039] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0040] The commuter area identification method in the embodiments of this application is described below.

[0041] Figure 1 This is an optional flowchart of the commuter area identification method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps 110 to 120. It is also understood that this embodiment... Figure 1 The order of steps 110 to 120 is not specifically limited. The order of steps can be adjusted or some steps can be reduced or added according to actual needs.

[0042] Step 110: Obtain multiple commuting weight parameter groups to initialize the parameter population.

[0043] In one embodiment, the commuting weight parameter set specifically includes three weighting factors for constructing the comprehensive commuting connection weights, namely, the commuting time impedance weight parameter. Housing price difference constraint weight parameters and basic commuter traffic weight parameters ,in, The initial value range of the commuting weight parameter group can be preset using the analytic hierarchy process (AHP) or expert scoring method.

[0044] Because commuter zone identification models in related technologies often rely excessively on single commuter flow scale or spatial distance, lacking consideration of deeper socio-economic driving mechanisms, and because in the real logic of urban operation, residents' daily cross-district commuting behavior is essentially the result of a trade-off between commuting time burden and the advantages of housing costs (such as housing price differences), ignoring such economic constraints leads to a lack of real-world explanatory power in the identification results. Therefore, in order to overcome the limitations of traditional single indicators, this embodiment introduces a commuting time impedance weight parameter. This study characterizes the punitive effect of long-distance or long-duration commutes on residents' travel, and aims to suppress the interference of unrealistically long commutes on regional delineation. It introduces a weighting parameter to constrain housing price differences. This study quantifies the reality that residents are forced to endure longer commutes in pursuit of lower housing costs, reflecting the separation of work and residence caused by economic pressure. A basic commuter flow weight parameter is introduced. This is used to characterize the strength of the actual physical connection between the basic residence and workplace, preserving the basic framework of the commuting network. It is evident that by combining these three dimensions, this embodiment can transform a single transportation network into a trade-off network that reflects the real socio-economic logic.

[0045] In one embodiment, a genetic algorithm, simulated annealing, particle swarm optimization (PSO), tabu search, or ant colony optimization (ACO) is selected as the optimization algorithm to search for the global optimum. When initializing the parameter population, to ensure that the optimization algorithm has sufficient global search capability and genetic diversity in the initial stage, multiple different combinations of weight parameters are generated through random sampling within a preset numerical range. For example, 30 different combinations are randomly generated. The numerical values ​​serve as the initial parameter population, providing an evolutionary basis for finding the optimal balance point among multiple objective functions. Furthermore, in megacities with volatile housing prices or severe job-housing separation, it is preferable to appropriately increase the weight parameter of housing price difference constraints in the commuting weight parameter group to more accurately quantify residents' long-distance commuting behavior forced by economic pressure.

[0046] Step 120: Use an optimization algorithm to iteratively optimize the parameter population. In each iteration, calculate the target fitness for each commuting weight parameter group in the current population. If the termination condition is not met, the optimization algorithm generates a new parameter population based on the target fitness and enters the next iteration. If the termination condition is met, output the target weight parameter group and its corresponding continuous commuting community as the commuting area division result.

[0047] In one embodiment, when dividing commuter zones, it is necessary to simultaneously consider two mutually constraining objectives: maximizing internal commuter retention rate and maximizing spatial morphological continuity. Therefore, an optimization algorithm is used for joint optimization, preferably a non-dominated sorting genetic algorithm with an elite retention strategy (such as NSGA-II). The following is a detailed description using a genetic algorithm as an example.

[0048] In each generation of iterative optimization, for each commuting weight parameter set within the current parameter population, it is substituted into the underlying model to calculate the target fitness of that parameter set in the current generation. Subsequently, based on the calculated target fitness, the algorithm performs non-dominated sorting and crowding calculations on all individuals in the current population (i.e., the commuting weight parameter sets). Then, through genetic operations such as selection, crossover, and mutation, a next generation of parameter populations with new feature combinations is generated. During this evolutionary process, an elite preservation strategy is used to directly retain and pass on the best-performing parent parameter sets to the offspring, effectively avoiding the loss of excellent solutions and ensuring stable convergence of the algorithm towards Pareto optimality. At the end of each iteration, the system determines whether the current evolutionary state meets the preset termination conditions. These termination conditions can be that the current evolutionary generation has reached the set maximum number of iterations (e.g., 15 generations), or that the average fitness of the parameter population has not changed significantly over multiple consecutive generations. If the termination condition is not met, the newly generated parameter population is used to return to the fitness evaluation step, starting the next iteration. When the termination condition is met, the algorithm stops iterating and outputs the Pareto optimal solution set that achieves the best balance between commuting behavior and spatial morphology. Finally, a set of target weight parameters is selected from this solution set according to actual decision preferences, and the continuous commuting communities corresponding to this target weight parameter set are taken as the final commuting area division result.

[0049] The calculation process for the target fitness will be described in detail below. (Refer to...) Figure 2 , Figure 2 This is a flowchart illustrating the calculation of target fitness for each commuting weight parameter set in the current population, provided in an embodiment of this application. The flowchart specifically includes the following steps: Step 210: Calculate the comprehensive commuting connection weight between each pair of spatial units based on the commuting flow volume data, commuting time data, housing cost data, and commuting weight parameter group of each spatial unit in the target area.

[0050] In one embodiment, the purpose of calculating the integrated commuting connection weights is to transform multi-source traffic flows and socioeconomic constraints into edge weights in the network topology. Specifically, using a weighted average of... Taking a research area composed of spatial units as an example, these spatial units include residential units and employment units. Indicates the residential unit index. Represents the employment unit index. Indicates from residential unit To the employment unit The number of commuters.

[0051] It is understood that although this embodiment supports multiple spatial scales, a standard 1km×1km geographic grid is preferred as the spatial unit. This scale can balance computational load and spatial resolution, avoiding internal heterogeneity loss due to excessively large spatial units, or computational redundancy due to excessively small spatial units. In addition, this embodiment is also applicable to irregular areas such as traffic analysis zones (TAZs), polygons, administrative regions, or Thiessen polygons generated based on points of interest (POIs) as the basic spatial units.

[0052] In one embodiment, reference is made to Figure 3 , Figure 3 This is a flowchart provided in this application embodiment for calculating the comprehensive commuting connection weight between any two spatial units based on commuting traffic volume data, commuting time data, housing cost data, and commuting weight parameter groups of each spatial unit within the target area. The flowchart specifically includes the following steps: Step 310: Calculate the basic commuting connection strength, commuting time impedance characteristics, and housing price difference constraint characteristics based on the commuting flow volume data, commuting time data, and housing cost data of each pair of spatial units.

[0053] In one embodiment, these two spatial units are defined as a residential unit and an employment unit, respectively. To transform raw traffic trajectories and multi-source socioeconomic data into quantitative indicators reflecting real human activity patterns, basic commuting connection strength, commuting time impedance characteristics, and housing price difference constraint characteristics are set. (Refer to...) Figure 4 , Figure 4 This is a flowchart provided in an embodiment of the present application for calculating basic commuting connection strength, commuting time impedance characteristics, and housing price difference constraint characteristics based on commuting flow volume data, commuting time data, and housing cost data of each pair of spatial units. The flowchart specifically includes the following steps: Step 410: Obtain the number of commuters from residential units to employment units from the commuter traffic volume data, standardize the commuter numbers, and calculate the basic commuter connection intensity.

[0054] In one embodiment, commuter traffic volume data is acquired. To adapt to the data infrastructure and application scenarios of different cities, the acquisition methods for commuter traffic volume data include, but are not limited to, any one or a combination of mobile phone signaling data, public transportation card swipe data, floating car and location trajectory data, and travel survey data. Mobile phone signaling data is collected from anonymized mobile phone signaling data, cleaned by an operator platform or a third-party location data platform, and extracted into "residence-workplace" OD (Origin-Destination) pairs to accurately obtain cross-grid commuting information. Public transportation card swipe data is obtained by acquiring the card swipe entry and exit times and station coordinate records of smart transportation cards such as bus cards and subway cards, and combining spatial clustering to estimate the number of commuters between residential and employment units. Floating car and location trajectory data utilizes anonymized trajectory data from vehicle terminals with GPS positioning capabilities or shared mobility platforms to extract origin and destination information for statistical analysis. Travel survey data is obtained from statistical data such as household travel surveys and population censuses conducted by statistical departments or transportation management departments to obtain the commuter distribution baseline. It is understandable that the method of obtaining commuter traffic volume data here is only for illustration and does not represent a limitation.

[0055] Then, from the acquired commuter traffic volume data, the actual number of commuters from residential unit i to employment unit j is extracted. To eliminate the absolute numerical differences caused by the population size of different spatial units and to accurately measure the relative proportions of different flow directions, this embodiment standardizes the acquired commuter numbers to calculate the basic commuter connection intensity. This indicator describes the relative importance of commuting trips from residential unit i to a specific employment unit j, specifically expressed as:

[0056] in, This represents the total number of commuters traveling from residential unit i to all employment units m within the study area.

[0057] Step 420: Obtain the first penalty parameter based on the commuting time data and the first attenuation parameter, and obtain the commuting time impedance characteristic based on the product of the first penalty parameter and the basic commuting connection strength.

[0058] In one embodiment, considering that commuting time significantly restricts the feasibility of commuting in real daily activities, this embodiment obtains commuting time data through various means to quantify this spatial impedance, such as calling map APIs, theoretical calculations based on road network structure, or average travel time based on historical trajectories. Calling map APIs involves accessing an external map service interface, inputting the coordinates of the center points of the residential and employment units, and directly obtaining the actual driving time between the two points, or obtaining the actual public transportation or cycling time as the commuting time. Theoretical calculations based on road network structure integrate open-source road network data, and calculate the theoretical commuting time using the shortest path algorithm based on road grade, theoretical speed limit, and topological distance between nodes. Average travel time based on historical trajectories utilizes historical GPS trajectory data from floating cars to statistically calculate the actual average travel time from the residential area to the employment area during specific commuting periods such as morning and evening rush hours. It is understood that the methods for obtaining commuting time data here are for illustrative purposes only and do not represent limitations.

[0059] Next, based on the obtained commuting time and the set first attenuation parameter Construct an exponential decay function This serves as the first penalty parameter. Subsequently, the first penalty parameter is multiplied by the calculated baseline commuting connection strength to calculate the commuting time impedance characteristic. , is represented as:

[0060] Among them, the first attenuation parameter As a time impedance parameter, it is used to suppress and penalize the interference of excessively long commutes on the delineation of commuter zone structures. In practical applications, the value of this parameter is preferably obtained by nonlinear regression estimation based on historical commuting big data. In the absence of local statistical data, it can be set to 0.5 to ensure the robustness of the exponential decay function in describing long-distance penalties.

[0061] In addition to the exponential decay function, the first penalty parameter can also be mathematically characterized using a power-law function, a Gaussian function, or a logistic function. Alternatively, it can be replaced with alternative indicators such as distance impedance or reachability impedance for penalty purposes.

[0062] Step 430: Obtain housing price difference constraint characteristics based on commuting time data and housing cost data.

[0063] In one embodiment, reference is made to Figure 5 , Figure 5This is a flowchart illustrating the process of obtaining housing price difference constraint characteristics based on commuting time data and housing cost data, provided in an embodiment of this application. The flowchart specifically includes the following steps: Step 510: Obtain the second penalty parameter based on the commuting time data and the second attenuation parameter.

[0064] In one embodiment, considering that in real-world urban spatial economics, although suburbs often offer significant cost advantages in terms of housing, excessively long commuting times can offset these economic benefits, this embodiment constructs a penalty mechanism specifically designed to measure time burden before introducing housing cost differences. Specifically, for commuting links from residential unit i to employment unit j, a penalty mechanism is established based on pre-acquired commuting time data. and the set second attenuation parameter Construct an exponential decay function As the second penalty parameter. Wherein, the second attenuation parameter... This is used to quantify the interaction effect between housing cost differences and commuting time burden; that is, the longer the commuting time, the greater the decline in the willingness to endure cross-regional commuting in exchange for lower housing prices. Additionally, the second penalty parameter can also be expressed mathematically using power-law decay or Gaussian decay.

[0065] Step 520: Obtain the first residential cost of the residential unit and the second residential cost of the employment unit from the residential cost data, obtain the cost difference between the second residential cost and the first residential cost, obtain the maximum value between the cost difference and zero, calculate the quotient of the maximum value and the first residential cost, and calculate the product of the quotient and the second penalty parameter to obtain the housing price difference constraint feature.

[0066] In one embodiment, the housing price difference constraint feature is used to quantify the reality that residents are forced to endure longer commutes in pursuit of lower housing costs, i.e., the separation of work and residence caused by housing price gradient pressure. To adapt to a multi-source data environment, the acquisition methods for residential cost data include, but are not limited to: real estate transaction platform data, proxy variable calculation, and publicly available government statistical data. Real estate transaction platform data is obtained through web crawlers or API interfaces, using average housing prices at the community level or average rent at the building level on real estate transaction platforms. Proxy variable calculation, when direct housing price or rent data is lacking, can collect socioeconomic indicators such as land transfer prices, building volume ratio, or per capita living area, and generate proxy variables for residential costs for each spatial unit through spatial interpolation or regression models. Publicly available government statistical data uses benchmark land prices or housing guidance prices for each street / community published by the housing and urban-rural development department or the statistics bureau.

[0067] During the calculation process, the first residential cost of residential unit i is extracted from the residential cost data. And the cost of second residence for employment unit j First, calculate the cost difference between the second residence cost and the first residence cost. Subsequently, the maximum value between this cost difference and zero is obtained using the maximum value function. The purpose of introducing the maximum value here is to establish the economic boundary condition: only when the cost of living in the place of employment is strictly greater than the cost of living in the place of residence (…) Only when the cost of employment is low do residents have the economic incentive to move to lower-cost areas in the periphery; if the place of employment is cheaper, there is no such specific economic constraint driving them.

[0068] Next, the maximum value is calculated in relation to the first housing cost. The quotient is used to reflect the relative proportion of cost disparity. Finally, this quotient is multiplied by the second penalty parameter to obtain the housing price difference constraint feature. , is represented as:

[0069] This transforms the housing price gradient, which reflects socioeconomic attributes, into a key constraint feature in network space connections, thereby greatly enhancing the practical explanatory power of the final commuter zone division scheme.

[0070] Step 320: Use the commuting weight parameter group to perform a weighted calculation on the basic commuting connection strength, commuting time impedance characteristics and housing price difference constraint characteristics to obtain the comprehensive commuting connection weight.

[0071] In one embodiment, the aforementioned basic commuting connection strength, commuting time impedance characteristics, and housing price difference constraint characteristics are fused to obtain a single quantitative indicator that can truly reflect the comprehensive behavioral trade-offs of residents. Specifically, in the current iteration of the optimization algorithm, the system extracts a specific set of commuting weight parameters being evaluated. This set of parameters contains three weight adjustment factors greater than zero, namely the first weight parameter. Second weighting parameter and the third weight parameter Subsequently, using these parameters, a linear weighted calculation is performed on the three feature dimensions calculated in the preceding steps, for example, a weighted summation, to determine the overall commuting connection weight. Represented as:

[0072] in: For commuting time impedance characteristics, These are the corresponding weighting parameters, used to adjust the proportion of sound generation in the overall connection of spatial-temporal impedance. The characteristic constrained by housing price differences The corresponding weight parameters are used to adjust the dominance of cross-regional commuting behavior caused by economic pressure in the overall connectivity. Based on basic commuting connection intensity, The corresponding weight parameters are used to control the basic proportion of the original traffic flow. Through the above weighted summation, the originally isolated multi-source heterogeneous data is reduced in dimension and fused into a single absolutely definite scalar value. The weight of this comprehensive commuting connection. It can accurately quantify the ultimate combined attractiveness between residential unit i and employment unit j after overcoming time resistance and being driven by economic compensation.

[0073] In one embodiment, in addition to the linear weighted combination method described above, the weighted calculation can also be a product-type combination method, expressed as: Alternatively, the basic commuting connection strength, commuting time impedance characteristics, and housing price difference constraint characteristics can be used as input variables and fed into a pre-trained machine learning model, such as random forest or neural network, to output a comprehensive nonlinear evaluation score as the comprehensive commuting connection weight.

[0074] Step 220: Construct a weighted commuting network based on the comprehensive commuting connection weights, perform preliminary clustering on the weighted commuting network to obtain candidate commuting communities, and perform spatial consistency correction on the candidate commuting communities to obtain continuous commuting communities.

[0075] In one embodiment, the research units based on geospatial distribution and their socioeconomic connections are abstracted and transformed into a network topology in graph theory. Specifically, based on all the comprehensive commuting connection weights calculated in the previous steps, a global weighted directed commuting network G=(V,E,W) is constructed, where each node in the network constitutes a node set V, and each node represents a spatial unit within the research area. The total number of nodes in the network equals the total number of spatial units within the research area, and nodes i and j correspond to the actual residential unit and employment unit, respectively. Since daily commuting behavior has a clear directionality, i.e., from residence to workplace, this network is constructed as a directed graph structure, and the actual commuting directions between spatial units constitute the edge set E. A directed edge from node i to node j represents the corresponding cross-regional commuting flow. In addition, each directed edge is assigned a specific value as an edge weight, which is the comprehensive commuting connection weight. The weight matrix W, composed of all edge weights, globally represents the intensity of comprehensive commuting connections between various micro-regions in the entire urban network.

[0076] Through the graph mapping described above, this embodiment transforms traditional geospatial clustering into a complex network community partitioning problem. In this constructed weighted directed commuting network, spatial units that have high-frequency commuting flows and conform to the economic principle of cross-regional residence to reduce costs will have significantly larger edge weights.

[0077] In one embodiment, reference is made to Figure 6 , Figure 6 This is a flowchart of preliminary clustering of a weighted commuter network to obtain candidate commuter communities, provided in an embodiment of this application. The flowchart specifically includes the following steps: Step 610: Obtain the total weight and the weighted degree corresponding to each node based on the edge weights.

[0078] In one embodiment, based on the constructed weighted commuting network G=(V,E,W), the basic topological parameters of the network are first extracted to provide a benchmark for community partitioning. Specifically, all nodes and directed edges in the network are traversed, and the total network weight m and the weighted degree of each node i are calculated based on the edge weights. Among them, weighted degree It represents the sum of the comprehensive commuting connection strengths between node i and all other nodes in the network, reflecting the total commuting throughput or activity of this spatial unit in the entire city network.

[0079] Next, the total network weight m is calculated, expressed as:

[0080] The total network weight m represents the total scale of integrated commuting connections between all spatial units within the entire study area.

[0081] Step 620: Select nodes in pairs to form node pairs. Obtain the first intermediate parameter by multiplying the corresponding weighted degrees and the total weight. Subtract the first intermediate parameter from the comprehensive commuting connection weight of the node pair to obtain the second intermediate parameter. Set a community identifier parameter to indicate whether the node pairs are located in the same community. Calculate the node pair parameter by multiplying the community identifier parameter and the second intermediate parameter. Accumulate all node pair parameters to obtain the accumulated parameter. Obtain the modularity objective function by quotient of the accumulated parameter and the total weight.

[0082] In one embodiment, a modularity objective function Q is constructed based on the total network weights to evaluate the quality of commuter zone partitioning. First, for any pair of nodes i and j selected in the network, the product of their weighted degrees ki and kj is calculated and divided by twice the total network weights to obtain a first intermediate parameter, expressed as: The first intermediate parameter characterizes the theoretically expected connection strength between node i and node j, assuming city-wide commuting trips occur completely randomly. Subsequently, the actual overall commuting connection weights between nodes are determined. Subtracting the first intermediate parameter yields the second intermediate parameter. The second intermediate parameter represents the degree to which actual commuting connections exceed random probability. If this value is greater than zero, it indicates that the actual commuting connections between the two locations are exceptionally close, exhibiting a strong tendency to cluster together. Next, a community identifier parameter is set to indicate whether the nodes are located in the same community. .in, and These are the community numbers assigned to nodes i and j, respectively. If they are assigned to the same candidate community, then... , =1; otherwise =0, this parameter acts as a filter switch. Finally, the product of the community identifier parameter and the second intermediate parameter is calculated to obtain the node pair parameter. The parameters of all node pairs in the entire network are then summed to obtain the accumulated parameter. This accumulated parameter is then divided by twice the total weight 2m to finally obtain the modularity objective function Q, expressed as:

[0083] Step 630: With the objective function of maximizing modularity as the goal, solve the community identifier parameters to obtain the target community identifier, and cluster the weighted commuting network according to the target community identifier to obtain candidate commuting communities.

[0084] In one embodiment, a community detection algorithm, such as the Louvain algorithm based on modularity optimization, is used. This algorithm has extremely high modularity optimization efficiency when processing large-scale grid networks and can quickly identify potential candidate commuter community prototypes. Heuristic clustering is performed on the weighted commuter network with the sole optimization objective of maximizing the constructed modularity objective function Q. Specifically, in the initial stage of the algorithm, each node in the network is treated as an independent community. Then, an iterative process is entered. During the iteration, each node in the network is traversed, attempting to move it into the community of its neighboring nodes, and the change in modularity resulting from this movement is calculated. .like If the value is greater than 0, the move is accepted, and the node's community identifier parameter is updated to the identifier of its neighboring nodes; otherwise, the original state is retained. This process is repeated until any move of any node in the network can no longer increase the overall modularity Q. At this point, the algorithm converges, and the final target community identifier ci of each node is extracted. Clusters of spatial units with the same target community identifier constitute a candidate commuting community. The candidate commuting communities partitioned through this optimization process can meet the structural requirement of having the densest internal commuting connections and the sparsest external connections.

[0085] Understandably, when performing preliminary clustering on a weighted commuter network to obtain candidate commuter communities, in addition to the Louvain algorithm, algorithms such as Infomap, Label Propagation, Fast Unfolding, or Spectral Clustering based on the Laplacian matrix can also be used for preliminary clustering.

[0086] In one embodiment, the initial clustering based on network topology only considers the comprehensive connection weights between units without applying strict two-dimensional geospatial constraints. This leads to the fragmentation of candidate commuter communities in physical space, such as internal gaps or external enclaves. To ensure the spatial coherence and geographical logic of commuter area boundaries, this embodiment introduces a spatial consistency correction mechanism. (Refer to...) Figure 7 , Figure 7 This is a flowchart of obtaining continuous commuting communities by performing spatial consistency correction on candidate commuting communities according to an embodiment of this application, specifically including the following steps: Step 710: Traverse the commuter areas in the candidate commuter communities, identify at least one continuous spatial entity in each commuter area, and identify the continuous spatial entity with the largest area as the maximum continuous spatial entity. Identify the adjacent units that are inside the maximum continuous spatial entity but have been assigned to other commuter areas, and merge the adjacent units into the commuter area to which the maximum continuous spatial entity belongs.

[0087] In one embodiment, a spatial adjacency matrix is ​​constructed by combining the geographic coordinates of each spatial unit, such as latitude and longitude or raster row and column numbers. First, all candidate commuter communities obtained from the initial segmentation are traversed. For each candidate commuter area, spatial analysis methods such as connected component labeling algorithms are used to identify all continuous spatial entities on the map formed by stitching together adjacent units. Then, the area or number of spatial units contained in each continuous spatial entity is calculated, and the entity with the largest area is determined as the largest continuous spatial entity of the commuter area. Next, neighboring units that are geographically completely surrounded by the largest continuous spatial entity (i.e., located within it) but were incorrectly assigned to other commuter areas during the network clustering stage due to individual data noise are identified. The community identifiers of these neighboring units are forcibly changed to the community identifier of the current largest continuous spatial entity, thereby re-merging them into the commuter area and ensuring the spatial integrity of the core area of ​​the commuter area.

[0088] Step 720: Identify isolated units corresponding to commuting areas. If the area of ​​an isolated unit is less than a preset area threshold, merge the isolated unit into an adjacent commuting area.

[0089] In one embodiment, after filling the internal holes, further depth correction is performed on the external enclaves that are detached from the main structure. Specifically, isolated units that are spatially distinct from their largest continuous spatial entity are identified within each commuting area. Isolated units refer to fragmented units that do not have any spatial adjacency with the main structure. To determine the affiliation of these isolated units, a preset area threshold judgment mechanism is introduced. Here, the preset area threshold can be set to 1% of the total area of ​​the commuting area where the isolated unit is located. The size of each isolated unit is judged. If the area of ​​an isolated unit is smaller than the preset area threshold, it is considered an interference patch with no obvious independent function and too small an area. In this case, it is separated from the original commuting area and forcibly merged into the surrounding commuting area that is directly adjacent to it in geographical space and has the closest connection with it.

[0090] This embodiment uses a set area threshold for forced merging, which can eliminate enclave phenomena to the greatest extent and ensure the continuity and integrity of the commuter area's execution boundary. It is understood that if an isolated unit's area is larger than the preset area threshold or is not adjacent to any main structure, it can be considered an independent area and retained or removed as invalid noise.

[0091] In one embodiment, mathematical morphology operators in the geographic information system can also be directly invoked for consistency judgment. For example, spatial dilation operations can be used to fill boundary gaps, spatial erosion operations can be used to remove small burrs, or closing operations can be used to smooth and close the boundaries of candidate commuter areas, ultimately outputting a continuous commuter community with coherent boundaries.

[0092] Step 230: Calculate the internal commuting retention rate and spatial continuity as target fitness based on the current continuous commuting community.

[0093] In one embodiment, reference is made to Figure 8 , Figure 8 This is a flowchart provided in an embodiment of the present application for calculating internal commuting retention rate and spatial continuity as target fitness based on the current continuous commuting community, specifically including the following steps: Step 810: For the commuter zones in the continuous commuting community, calculate the number of first commuters whose starting point and destination are both in the commuter zone, and the number of second commuters whose starting point or destination is either in the commuter zone. Based on the number of first commuters and the number of second commuters, obtain the internal commuting retention rate, so as to maximize the internal commuting retention rate as the internal commuting retention rate target.

[0094] In one embodiment, to measure the ability of the defined commuter zones to encompass and interpret real-world daily commuting trajectories, a first evaluation objective, namely the internal commuting retention rate, is constructed from the dimension of human behavioral logic. Specifically, for any continuous commuting community Rr obtained after spatial consistency correction, all commuting flow data occurring within that area are extracted. .

[0095] First, calculate the total number of commuters whose starting point i and ending point j are both located within the commuting zone Rr. This number is designated as the first commuter, representing the commuter volume that achieves a complete work-life balance within the commuter zone. Subsequently, the total number of commuters is calculated, with at least one end of either the starting point i or the ending point j located within the commuter zone Rr. This number is designated as the second commuter, which represents the total potential commuter count associated with that commuter zone.

[0096] Next, the ratio of the first commuter to the second commuter is calculated to obtain the internal commuting retention rate ISC(r) of the commuting area Rrr. The closer the internal commuting retention rate is to 1, the more accurate the boundary is, and the more successfully the daily commuting life of the vast majority of residents is included.

[0097] Maximizing the internal commuting retention rate across all partitioned areas is the primary optimization objective, expressed as:

[0098] Step 820: For commuter zones in a continuous commuter community, obtain the adjacency value and the corresponding commuter zone identifier for the spatial unit, and obtain the spatial continuity parameter based on the adjacency value and the commuter zone identifier, with maximizing the spatial continuity parameter as the spatial continuity objective.

[0099] In one embodiment, in addition to the behavioral dimension assessment, a second evaluation objective is constructed from the geographical morphology dimension, namely the spatial continuity parameter, to prevent the algorithm from forcibly piecing together far apart independent plots in order to pursue an extremely high commuting retention rate.

[0100] Specifically, a spatial adjacency matrix A={ is introduced to describe the underlying grid topology of the study area. If spatial cell i and spatial cell j are geographically directly adjacent, then the adjacency value is... If they are not adjacent, then Simultaneously, obtain the commuter zone identifier for each spatial unit under the current partitioning scheme. and If spatial unit i belongs to commuting area Rr, then =1, otherwise 0; similarly, if spatial unit j belongs to commuter zone Rr, then =1.

[0101] Next, calculate the adjacency value. With two commuter zone signs , product of products The product is only valid if the two spatial units are geographically adjacent. ), and were simultaneously assigned to the same commuter zone ( =1 and This item is counted as 1 only when the value is 1. The product results of all spatial unit pairs in the whole are summed up to obtain the spatial continuity parameter reflecting the degree of contiguousness of the overall plots. The higher the score of this parameter, the more compact and coherent the commuter zone is on the map, and the more it meets the objective requirements of urban physical planning.

[0102] Maximizing the spatial continuity parameter as the second optimization objective is expressed as:

[0103] Step 830: Combine internal commuting retention rate and spatial continuity parameters as the target fitness.

[0104] In one embodiment, within a real geographic space and transportation network, the aforementioned internal commuting retention rate and spatial continuity parameter are mutually constrained. To find the optimal balance between these two, the internal commuting retention rate and spatial continuity parameter are jointly used as the target fitness of the multi-objective optimization algorithm. Furthermore, the internal commuting retention rate, the primary optimization objective when calculating the target fitness, can be replaced by calculating indicators such as internal modularity, compactness, or mutual information based on information theory. These replaced indicators are then combined with the spatial continuity parameter to obtain the final target fitness.

[0105] Next, in each generation of the genetic algorithm's propagation and evaluation process, the current weighted parameter group population is non-dominated and ranked according to the fitness of these two objectives. Individuals that can achieve an advantage in both objectives simultaneously, or those that excel in one objective without severely harming the other, are given higher retention priority and passed on to the next generation. Through this multi-objective joint evaluation mechanism, a set of Pareto optimal solutions can ultimately be evolved and selected, resulting in the optimal commuting zone division that achieves an extreme balance in both commuting behavior and geographic space.

[0106] In one embodiment, reference is made to Figure 9 , Figure 9 This is an example of commuter area identification provided in the embodiments of this application. Figure 9 The example uses data from two cities (city cluster, region, bay area). The commuting data in the figure includes commuting traffic volume data and commuting time data.

[0107] One of the studies focuses on research area 1, which includes nine cities.

[0108] First, a basic geospatial grid was constructed for the study area, dividing the entire area of ​​the nine cities into standard 1km×1km geographic grids as the smallest spatial unit. Anonymized mobile signaling data was collected and cleaned by the operator platform to extract "residence-workplace" OD pairs, covering approximately 900,000 commuting travel records across spatial units. Average housing prices at the residential community level were collected from multiple platforms for subsequent extraction of the first and second residential costs for each spatial unit. The actual driving time between the center points of each spatial unit was obtained by calling the map API. Road network data and map POI data were integrated to help assess the attractiveness of the spatial units.

[0109] Next, the system settings are configured based on the parameters obtained from the actual regression model. The first decay parameter used to calculate the first penalty parameter is set to 0.415; the second decay parameter used to constrain housing costs is set to 0.152. A genetic algorithm is used to randomly sample within the numerical range of [0.01, 2.0] to obtain multiple commuting weight parameter sets to initialize the parameter population.

[0110] Then, a weighted commuting network is constructed based on the calculated comprehensive commuting connection weights, and the Louvain algorithm is applied to the weighted commuting network for community detection to obtain candidate commuting communities. The commuting areas within the candidate commuting communities are traversed, and the largest continuous grid cell within each community is extracted and identified as the largest continuous spatial entity. Isolated units located at the edges of commuting areas are identified; for example, a spatial adjacency tolerance of Tolerance=2 is set to identify fragmented units without direct adjacencies. These isolated units are forcibly merged into adjacent commuting areas, eliminating spatial fragmentation and thus obtaining continuous commuting communities.

[0111] To obtain the target weight parameter set, an optimization algorithm, such as the NSGA-II multi-objective genetic algorithm, is used to iteratively optimize the parameter population. The population size is set to 30, and the maximum number of iterations (termination condition) is set to 15 generations. The objective is to maximize the internal commuting retention rate, ensuring that the identified continuous commuting communities have strong functional cohesion. The objective is to maximize the spatial continuity parameter, ensuring that the region is geographically compact and free of redundant isolated units. In each iteration, the internal commuting retention rate and spatial continuity parameter are combined as the target fitness. If the termination condition is met (after 15 generations of evolution), the Pareto optimal solution set is obtained, yielding the commuting area division result.

[0112] According to the commuter zone division results, under the selected set of target weight parameters and their corresponding continuous commuter communities, the average internal commuting retention rate can reach over 90%, and the maximum continuous spatial entity within each commuter zone increases significantly. This proves that this embodiment can perfectly balance behavioral logic and spatial form when processing complex urban agglomeration data. The optimized target weight parameter set reflects the role of different dimensions of weight factors in the commuter zone identification process. For example, under a specific parameter combination, the constraint of housing cost becomes the most critical dimension for commuter zone identification, with 90.40% of commuting behavior occurring within the finally identified commuter zones.

[0113] Secondly, nine counties in other regions were selected as the research area 2.

[0114] First, the entire area of ​​the nine counties was divided into standard 1km×1km geographic grids as the smallest spatial unit. Anonymized mobile signaling data was collected, cleaned, and extracted into "residence-workplace" OD pairs, covering approximately 140,000 commuting travel information records across spatial units. Average rent at the building level was collected from different platforms. The actual driving time between the center points of each spatial unit was obtained by calling the map API. Road network data and location POI data were integrated to help assess the attractiveness of spatial units. Next, the system settings were configured based on the parameters obtained from the actual regression model. The first decay parameter used to calculate the first penalty parameter was set to 0.746; the second decay parameter used to constrain housing costs was set to 0.389. A genetic algorithm was used to randomly sample within the numerical range of [0.01, 2.0] to obtain multiple commuting weight parameter groups to initialize the parameter population. Based on the above data and parameters, the same steps as in study area 1 are performed: calculating the comprehensive commuting connection weights to construct the network, performing preliminary clustering to obtain candidate commuting communities, correcting spatial consistency to obtain continuous commuting communities, and iteratively optimizing based on the calculated target fitness, finally outputting the target weight parameter set and the corresponding continuous commuting communities.

[0115] As can be seen from the above, the commuter area identification method provided in this application has the following significant comprehensive benefits.

[0116] Firstly, in terms of recognition accuracy, spatial resolution is improved by an order of magnitude. This embodiment breaks through the limitations of traditional coarse-grained modeling based on administrative units such as districts and counties, and identifies commuter areas using extremely fine-grained spatial units. This enables precise capture of the micro-level commuter cores and their dynamic interaction characteristics within the city. By introducing a first penalty parameter to suppress excessively long-distance commuting interference, and by introducing housing price difference constraints combined with a second penalty parameter, the embodiment accurately depicts the real behavioral trade-offs between residents' housing costs and commuting time burdens. This makes the underlying network structure built based on comprehensive commuting connection weights more consistent with real socio-economic logic. By performing a spatial consistency correction step on the candidate commuter communities obtained from the initial clustering, incorrectly assigned adjacent units and isolated units with areas smaller than a preset area threshold are forcibly merged. This effectively solves the spatial fragmentation phenomenon commonly seen in network clustering, outputting continuous commuter communities with coherent boundaries and conforming to geographical logic. By using optimization algorithms to iteratively optimize the parameter population, and combining the internal commuting retention rate and spatial continuity parameters as the target fitness for dynamic screening, the final partitioning result output in this embodiment shows a significant increase in the internal commuting retention rate index compared to traditional methods, and completely eliminates meaningless enclaves, thereby greatly enhancing the spatial compactness of urban commuting areas.

[0117] Secondly, in terms of economic benefits, it can optimize resource allocation and reduce social costs. Based on the continuous commuter communities output in this embodiment, it can clearly reveal the excessive cross-regional commuting phenomenon caused by the pressure of tiered housing prices. Government and planning departments can then use this information to precisely plan affordable housing or adjust industrial spatial layout within specific commuter zones, thereby fundamentally and effectively reducing the overall urban commuting economic costs, transportation energy consumption, and carbon emissions. Using the identification results of the real comprehensive commuter connection weights reflected in this embodiment for the overall planning of cross-regional infrastructure such as rail transit and expressways can effectively avoid the resource misallocation caused by blind construction based on rigid administrative boundaries in the past, and significantly improve the investment return and utilization rate of public infrastructure.

[0118] Thirdly, in terms of social benefits, it can improve the precision of governance and residents' sense of well-being. This embodiment, through multi-objective optimization, outputs high-fidelity and refined commuter zone boundaries, realistically reflecting residents' cross-regional travel patterns and residential compromises. This provides a highly valuable spatial benchmark for the government to formulate cross-administrative region talent attraction policies, allocate employment subsidies, and ensure the equitable distribution of public services. By revealing the real work-residence connection network to assist in optimizing the urban commuting structure and shortening residents' average commuting time, this embodiment not only helps to significantly reduce urban transportation carbon emissions but also directly contributes to the city's sustainable development strategy, effectively improving residents' daily travel experience and overall life satisfaction.

[0119] The technical solution provided in this application initializes a parameter population by acquiring multiple commuting weight parameter groups; iteratively optimizes the parameter population using an optimization algorithm. In each iteration, the target fitness is calculated for each commuting weight parameter group in the current population. If the termination condition is not met, the optimization algorithm generates a new parameter population based on the target fitness and enters the next iteration. If the termination condition is met, the target weight parameter group and its corresponding continuous commuting community are output as the commuting area division result. The calculation of the target fitness for each commuting weight parameter group in the current population includes the following steps: calculating the comprehensive commuting connection weight between each pair of spatial units based on the commuting flow scale data, commuting time data, housing cost data, and commuting weight parameter groups of each spatial unit within the target area; constructing a weighted commuting network based on the comprehensive commuting connection weight; performing preliminary clustering on the weighted commuting network to obtain candidate commuting communities; performing spatial consistency correction on the candidate commuting communities to obtain continuous commuting communities; and calculating the internal commuting retention rate and spatial continuity as the target fitness based on the current continuous commuting community. This application's embodiments calculate comprehensive commuter connection weights by integrating commuter flow scale data, commuter time data, and housing cost data. This quantifies residents' actual behavioral trade-offs between commuter time burden and housing cost advantages in daily life, making the spatial connection modeling more consistent with real socioeconomic logic and significantly improving the accuracy of the identification results. Secondly, after initially clustering candidate commuter communities, a spatial consistency correction step is introduced to deeply correct the clustering results to obtain continuous commuter communities. This effectively eliminates fragmented patches that do not conform to geographical norms, ensuring the continuity of the delineated commuter zone boundaries. Finally, the internal commuter retention rate, which measures the interpretability of commuter flows, and the spatial continuity, which measures geographical morphological characteristics, are jointly used as the target fitness. Through optimization algorithms, multiple weight parameter combinations are dynamically filtered and proliferated in each iteration, thereby finding the global optimal solution among mutually constraining objective functions. This further improves the identification accuracy and structural rationality, making the final commuter zone delineation results more accurate and reliable.

[0120] This application also provides a commuter area identification device, which can implement the above-described commuter area identification method, referring to... Figure 10 The device includes: Parameter acquisition module 1010: Used to acquire multiple commuting weight parameter groups to initialize the parameter population.

[0121] The optimization iteration module 1020 is used to iteratively optimize the parameter population using an optimization algorithm. In each iteration, the target fitness is calculated for each commuting weight parameter group in the current population. If the termination condition is not met, the optimization algorithm generates a new parameter population based on the target fitness and enters the next iteration. If the termination condition is met, the target weight parameter group and its corresponding continuous commuting community are output as the commuting area division result.

[0122] The calculation of target fitness for each commuting weight parameter set in the current population includes the following steps: Calculate the comprehensive commuting connection weight between each pair of spatial units based on the commuting flow scale data, commuting time data, housing cost data, and commuting weight parameter set of each spatial unit in the target area; construct a weighted commuting network based on the comprehensive commuting connection weight; perform preliminary clustering on the weighted commuting network to obtain candidate commuting communities; perform spatial consistency correction on the candidate commuting communities to obtain continuous commuting communities; and calculate the internal commuting retention rate and spatial continuity as target fitness based on the current continuous commuting communities.

[0123] The specific implementation of the commuter area identification device in this embodiment is basically the same as the specific implementation of the commuter area identification method described above, and will not be repeated here.

[0124] This application also provides an electronic device, including: At least one memory; At least one processor; At least one program; The program is stored in a memory, and the processor executes the at least one program to implement the commuter area identification method described above. The electronic device can be any smart terminal, including mobile phones, tablets, personal digital assistants (PDAs), and in-vehicle computers.

[0125] Please see Figure 11 , Figure 11 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 1101 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1102 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1102 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1102 and is called and executed by the processor 1101 to execute the commuting area identification method of the embodiments of this application. Input / output interface 1103 is used to implement information input and output; The communication interface 1104 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1105 transmits information between various components of the device (e.g., processor 1101, memory 1102, input / output interface 1103, and communication interface 1104); The processor 1101, memory 1102, input / output interface 1103 and communication interface 1104 are connected to each other within the device via bus 1105.

[0126] This application embodiment also provides a storage medium that stores a computer program, which, when executed by a processor, implements the above-described commuting area identification method.

[0127] Memory, as a non-transitory storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0128] The commuter area identification method, apparatus, device, and storage medium proposed in this application initialize a parameter population by acquiring multiple commuter weight parameter groups; iteratively optimizes the parameter population using an optimization algorithm. In each iteration, a target fitness is calculated for each commuter weight parameter group in the current population. If the termination condition is not met, the optimization algorithm generates a new parameter population based on the target fitness and enters the next iteration; if the termination condition is met, the target weight parameter group and its corresponding continuous commuter community are output as the commuter area division result. The calculation of the target fitness for each commuter weight parameter group in the current population includes the following steps: calculating the comprehensive commuter connection weight between each pair of spatial units based on commuter traffic volume data, commuter time data, housing cost data, and commuter weight parameter groups within the target area; constructing a weighted commuter network based on the comprehensive commuter connection weight; performing preliminary clustering on the weighted commuter network to obtain candidate commuter communities; performing spatial consistency correction on the candidate commuter communities to obtain continuous commuter communities; and calculating the internal commuter retention rate and spatial continuity as the target fitness based on the current continuous commuter community. This application's embodiments calculate comprehensive commuter connection weights by integrating commuter flow scale data, commuter time data, and housing cost data. This quantifies residents' actual behavioral trade-offs between commuter time burden and housing cost advantages in daily life, making the spatial connection modeling more consistent with real socioeconomic logic and significantly improving the accuracy of the identification results. Secondly, after initially clustering candidate commuter communities, a spatial consistency correction step is introduced to deeply correct the clustering results to obtain continuous commuter communities. This effectively eliminates fragmented patches that do not conform to geographical norms, ensuring the continuity of the delineated commuter zone boundaries. Finally, the internal commuter retention rate, which measures the interpretability of commuter flows, and the spatial continuity, which measures geographical morphological characteristics, are jointly used as the target fitness. Through optimization algorithms, multiple weight parameter combinations are dynamically filtered and proliferated in each iteration, thereby finding the global optimal solution among mutually constraining objective functions. This further improves the identification accuracy and structural rationality, making the final commuter zone delineation results more accurate and reliable.

[0129] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0130] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0131] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0132] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0133] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0134] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0135] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0136] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0137] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0138] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0139] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for identifying commuter areas, characterized in that, include: Obtain multiple commuting weight parameter sets to initialize the parameter population; The optimization algorithm iteratively optimizes the parameter population. In each iteration, the target fitness is calculated for each commuting weight parameter group in the current population. If the termination condition is not met, the optimization algorithm generates a new parameter population based on the target fitness and enters the next iteration. If the termination condition is met, the target weight parameter group and its corresponding continuous commuting community are output as the commuting area division result. The calculation of the target fitness for each of the commuting weight parameter groups in the current population includes the following steps: calculating the comprehensive commuting connection weight between each pair of spatial units based on the commuting flow scale data, commuting time data, housing cost data, and the commuting weight parameter groups of each spatial unit in the target area; constructing a weighted commuting network based on the comprehensive commuting connection weights; performing preliminary clustering on the weighted commuting network to obtain candidate commuting communities; performing spatial consistency correction on the candidate commuting communities to obtain continuous commuting communities; and calculating the internal commuting retention rate and spatial continuity as the target fitness based on the current continuous commuting communities.

2. The commuter area identification method according to claim 1, characterized in that, The calculation of the comprehensive commuting connection weight between any two spatial units within the target area, based on commuting traffic volume data, commuting time data, housing cost data, and the commuting weight parameter set, includes: Based on the commuting flow data, commuting time data, and housing cost data of each pair of spatial units, the basic commuting connection strength, commuting time impedance characteristics, and housing price difference constraint characteristics are calculated, and the two spatial units are defined as a residential unit and an employment unit, respectively. The comprehensive commuting connection weight is obtained by weighting the basic commuting connection strength, the commuting time impedance characteristics, and the housing price difference constraint characteristics using the commuting weight parameter group.

3. The commuter area identification method according to claim 2, characterized in that, The calculation of basic commuter connectivity strength, commuter time impedance characteristics, and housing price difference constraint characteristics based on the commuter flow volume data, commuter time data, and housing cost data of each pair of spatial units includes: The number of commuters from the residential unit to the employment unit is obtained from the commuter traffic volume data, the number of commuters is standardized, and the basic commuter connection intensity is calculated. A first penalty parameter is obtained based on the commuting time data and the first attenuation parameter, and the commuting time impedance characteristic is obtained by multiplying the first penalty parameter and the basic commuting connection strength. The housing price difference constraint features are derived from the commuting time data and the housing cost data.

4. The commuter area identification method according to claim 3, characterized in that, The process of obtaining the housing price difference constraint features based on the commuting time data and the housing cost data includes: The second penalty parameter is obtained based on the commuting time data and the second attenuation parameter; The housing price difference constraint feature is obtained by obtaining the first housing cost of the residential unit and the second housing cost of the employment unit from the housing cost data, obtaining the cost difference between the second housing cost and the first housing cost, obtaining the maximum value between the cost difference and zero, calculating the quotient of the maximum value and the first housing cost, and calculating the product of the quotient and the second penalty parameter.

5. The commuter area identification method according to claim 1, characterized in that, In the weighted commuting network, the spatial unit is used as a node, and the corresponding comprehensive commuting connection weight is used as the edge weight of the directed edge. The preliminary clustering of the weighted commuting network to obtain candidate commuting communities includes: The total weight and the weighted degree corresponding to each node are obtained based on the edge weights. The nodes are selected in pairs to form node pairs. A first intermediate parameter is obtained by multiplying the corresponding weighted degree by the total weight. The first intermediate parameter is subtracted from the comprehensive commuting connection weight of the node pair to obtain a second intermediate parameter. A community identifier parameter is set to indicate whether the node pairs are located in the same community. The node pair parameter is obtained by multiplying the community identifier parameter and the second intermediate parameter. All the node pair parameters are accumulated to obtain an accumulated parameter. The modularity objective function is obtained by quotienting the accumulated parameter and the total weight. With the objective function of maximizing modularity as the goal, the community identifier parameters are solved to obtain the target community identifier, and the weighted commuting network is clustered based on the target community identifier to obtain the candidate commuting communities.

6. The commuter area identification method according to claim 5, characterized in that, The step of spatially consistent correcting the candidate commuting communities to obtain continuous commuting communities includes: Traverse the commuter areas in the candidate commuter communities, determine at least one continuous spatial entity in each commuter area, and determine the continuous spatial entity with the largest area as the largest continuous spatial entity. Identify adjacent units that are inside the largest continuous spatial entity but are assigned to other commuter areas, and merge the adjacent units into the commuter area to which the largest continuous spatial entity belongs. Identify isolated units corresponding to the commuting area. If the area of ​​the isolated unit is less than a preset area threshold, merge the isolated unit into the adjacent commuting area.

7. The commuter area identification method according to claim 1, characterized in that, The calculation of internal commuting retention rate and spatial continuity as target fitness based on the current continuous commuting community includes: For the commuter zone in the continuous commuting community, calculate the number of first commuters whose starting point and ending point are both in the commuter zone, and the number of second commuters whose starting point or ending point is either in the commuter zone. Based on the number of first commuters and the number of second commuters, obtain the internal commuting retention rate, so as to maximize the internal commuting retention rate as the internal commuting retention rate target. For the commuter areas in the continuous commuter community, the adjacency value and the corresponding commuter area identifier of the spatial unit are obtained, and the spatial continuity parameter is obtained based on the adjacency value and the commuter area identifier, with maximizing the spatial continuity parameter as the spatial continuity objective; The internal commuting retention rate and the spatial continuity parameter are combined as the target fitness.

8. A commuter area identification device, characterized in that, include: Parameter acquisition module: used to acquire multiple commuting weight parameter sets to initialize the parameter population; The optimization iteration module is used to iteratively optimize the parameter population using an optimization algorithm. In each iteration, the target fitness is calculated for each commuting weight parameter group in the current population. If the termination condition is not met, the optimization algorithm generates a new parameter population based on the target fitness and enters the next iteration. If the termination condition is met, the target weight parameter group and its corresponding continuous commuting community are output as the commuting area division result. The calculation of the target fitness for each of the commuting weight parameter groups in the current population includes the following steps: calculating the comprehensive commuting connection weight between each pair of spatial units based on the commuting flow scale data, commuting time data, housing cost data, and the commuting weight parameter groups of each spatial unit in the target area; constructing a weighted commuting network based on the comprehensive commuting connection weights; performing preliminary clustering on the weighted commuting network to obtain candidate commuting communities; performing spatial consistency correction on the candidate commuting communities to obtain continuous commuting communities; and calculating the internal commuting retention rate and spatial continuity as the target fitness based on the current continuous commuting communities.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the commuter area identification method according to any one of claims 1 to 7.

10. A storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the commuter area identification method according to any one of claims 1 to 7.

Images