Method and system for analyzing industrial correlation degree between cities based on industrial migration data

Abstract

The invention discloses an inter-city industrial correlation analysis method and system based on industrial migration data, and relates to the technical field of industrial correlation analysis. The method comprises the following steps: collecting industry classification data of a first target city and a second target city, and constructing an industry chain graph according to the industry classification data; performing data field design according to the industrial chain atlas by adopting a preset structured field model, and constructing an industrial dynamic data set based on data fields; and according to the industrial dynamic data set, calculating industrial association strength among different target cities, and generating an industrial association spatial pattern visualization graph among the different target cities based on the industrial association strength. Through the method provided by the invention, fine-grained analysis of upstream, middle and downstream links of an industrial chain can be realized by fusing multi-source industrial migration big data, so that a real-time, multi-dimensional and operable decision basis is provided for collaborative development, investment attraction and policy making of regional industries.

Description

Technical Field

[0001] This invention relates to the technical field of industrial linkage analysis, and in particular to a method and system for analyzing industrial linkages between cities based on industrial migration data. Background Technology

[0002] Against the backdrop of globalization, regional integration, and industrial transformation and upgrading, clearly depicting the industrial connections between cities (such as transfer, cooperation, and complementarity) is crucial for formulating regional coordinated development policies, investment attraction strategies, industrial layout optimization, and industrial chain and supply chain security assessments.

[0003] Traditional analyses based on static statistical data struggle to capture complex upstream and downstream collaborative relationships. More intuitive and granular spatial visualization tools are needed to understand the patterns of industrial linkages between cities.

[0004] Existing industry correlation analysis mainly focuses on the connections between upstream and downstream industries within a certain region, or on factors such as topography and urban development of different cities (mostly geographically connected), or establishes correlation indicators to obtain city correlation analysis. Existing correlation analysis only targets a single characteristic (between cities or between upstream and downstream industries), and cannot handle massive data processing with precision down to specific regions or specific industry segments. Especially when there are huge differences in industrial structure and development level between different cities, it cannot quickly and effectively conduct correlation analysis or accurate data display at the multi-city and fine-scale industry scale. It cannot guide further industrial planning or implement various measures to upgrade the city's level. There is an urgent need for those skilled in the art to solve the corresponding technical problems. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the present invention aims to propose a method and system for analyzing the industrial correlation between cities based on industrial migration data. This method can achieve fine-grained analysis of the upstream, midstream, and downstream links of the industrial chain by integrating multi-source industrial migration big data, thereby providing real-time, multi-dimensional, and operable decision-making basis for regional industrial collaborative development, investment promotion, and policy formulation.

[0006] To achieve the objectives of this invention, the following technical solution is adopted: One of the objectives of this invention is: A method for analyzing the industrial linkages between cities based on industrial migration data, the method comprising the following steps: Collect industry classification data for the first and second target cities, and construct an industry chain map based on the industry classification data; The data fields are designed based on the industry chain map using a pre-defined structured field model, and an industry dynamic dataset is constructed based on the data fields. The industrial linkage strength between different target cities is calculated based on the industrial dynamic dataset, and a spatial pattern visualization map of industrial linkage between different target cities is generated based on the industrial linkage strength.

[0007] In the above technical solution, by collecting industry classification data from different cities or urban agglomerations and constructing an industry chain map based on the industry classification data, it is possible to distinguish specific links in the industry chain throughout the entire process of data analysis field design, data aggregation and processing, and visualization. For the industry chain map, a pre-set structured field model is used, and the data field design is carried out based on the industry chain map using the structured field model, thereby further refining the classification of the collected industry classification data. This can accurately and intuitively reveal the spatiotemporal evolution of industrial transfer, cooperation, and survival relationships between cities. The industrial linkage strength between different target cities is calculated based on the industrial dynamic dataset, and a visualization map of the spatial pattern of industrial linkages between different target cities is generated based on the industrial linkage strength. This enables fine-grained analysis of the upstream, midstream, and downstream links of the industry chain, thereby providing real-time, multi-dimensional, and operable decision-making basis for regional industrial collaborative development, investment promotion, and policy formulation.

[0008] Furthermore, the process of constructing the industry chain map based on the industry classification data includes: The industry code standardization matching of the industry classification data of the first target city and the industry classification data of the second target city is performed using the national economic industry classification rules to obtain the first industry industry map and the second industry industry map. The first industry industry map and the second industry industry map are divided and reconstructed according to the three links of upstream, midstream and downstream to obtain the industry chain map.

[0009] Furthermore, based on the aforementioned industry chain map, the industries in the upstream, midstream, and downstream segments are further subdivided into multiple levels until each level of industry is matched with the industry sub-category code in the National Economic Industry Classification. This process includes: When a specified industry has a clear classification within its industry and can be matched with the standard national economic industry name, the classification result is directly matched; When a specified industry has a clear classification within its respective sector but cannot be directly matched, text similarity calculation is used to find the most similar industry in the knowledge base and replace the keywords. When a designated industry is not clearly classified within its respective sector, natural language processing technology is used to identify industry segments or entities from the industry description, match them with the national economic industry classification standards, and assign industry codes.

[0010] In the above technical solution, industry codes are standardized and matched according to the national economic industry classification rules to obtain the first industry industry map and the second industry industry map. The first industry industry map and the second industry industry map are then divided and reconstructed according to the upstream, midstream and downstream links to obtain the industry chain map. This can integrate multi-source industrial migration big data to achieve fine-grained analysis of the upstream, midstream and downstream links of the industry chain.

[0011] Furthermore, the data fields designed based on the industrial chain map using a preset structured field model include, but are not limited to, various combinations of: industrial transfer statistics fields, enterprise investment statistics fields, enterprise branch statistics fields, enterprise survival statistics fields, newly established enterprise statistics fields, and defunct enterprise statistics fields.

[0012] Furthermore, in the process of constructing the industry dynamic dataset based on the aforementioned data fields, the industry dynamic dataset is constructed with industry classification code-administrative division code-time as the core dimensions to support the related query and analysis of different data field themes.

[0013] Furthermore, the process of calculating the industrial linkage strength between different target cities based on the aforementioned industrial dynamics dataset includes: Construct an industry association network among city nodes based on the aforementioned industry dynamics dataset; Based on the aforementioned industry linkage network, the intensity of industry linkages between cities is calculated using a pre-defined linkage analysis model.

[0014] Furthermore, the correlation analysis model uses a weighted algorithm to calculate the industrial correlation strength value between cities, and the expression is as follows: R[ab]= α×(T[ab] / T[a]) + β×(I[ab] / I[a]) + γ×(B[ab] / B[a]) Where R[ab] represents the industrial linkage strength from city a to city b; T[ab] represents the number of industrial transfers from city a to city b; T[a] represents the total number of industrial transfers generated by city a; I[ab] represents the total investment amount from city a to city b; I[a] represents the total investment amount from city a to all cities; B[ab] represents the number of branch offices established by corporate headquarters in city a to branch offices in city b; B[a] represents the total number of corporate headquarters established in city a; α, β, and γ all represent industry characteristic weight coefficients.

[0015] Furthermore, the industry characteristic weighting coefficients α, β, γ are configured differently based on the technology intensity, capital intensity, or supply chain characteristics of the target industry.

[0016] Furthermore, the ArcGIS spatial geographic information platform is used to generate a visualization map of the spatial pattern of industrial associations between different target cities based on the aforementioned industrial association strength.

[0017] The above technical solution comprehensively considers multiple dimensions such as migration, cooperation, and survival, and pre-sets a correlation analysis model to more closely reflect the complex processes of industrial operations in real life. Based on the dynamic industrial dataset, it provides more comprehensive and objective analysis results. Then, it fully utilizes the spatial analysis and visualization advantages of ArcGIS to present the abstract and complex inter-city industrial connections in a highly intuitive and clear multi-dimensional map form, which greatly improves the efficiency of understanding the results and the value of decision support.

[0018] The second objective of this invention is: A system for analyzing inter-city industrial linkages based on industrial migration data, the system comprising: The data acquisition module is used to collect industry classification data for the first and second target cities. The graph construction module is used to construct an industry chain graph based on the industry classification data. The field design module is used to design data fields based on the industry chain map using a preset structured field model, and to construct an industry dynamic dataset based on the data fields. The correlation analysis module is used to calculate the industrial correlation strength between different target cities based on the industrial dynamic dataset, and to generate a visualization map of the spatial pattern of industrial correlation between different target cities based on the industrial correlation strength.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes a method for analyzing the industrial linkages between cities based on industrial migration data. By collecting industry classification data from different cities or urban clusters and constructing an industrial chain map based on this data, it enables the differentiation of specific links in the industrial chain throughout the entire process of data analysis field design, data aggregation and processing, and visualization. For the industrial chain map, a pre-defined structured field model is used, and this model is employed to design data fields based on the industrial chain map, further refining the collected industry classification data. This accurately and intuitively reveals the spatiotemporal evolution patterns of industrial transfer, cooperation, and survival relationships between cities. The method calculates the industrial linkage strength between different target cities based on the industrial dynamic dataset and generates a visualization map of the spatial pattern of industrial linkages between different target cities based on this strength. This allows for fine-grained analysis of the upstream, midstream, and downstream links of the industrial chain, providing real-time, multi-dimensional, and actionable decision-making support for regional industrial collaborative development, investment attraction, and policy formulation. Attached Figure Description

[0020] Figure 1 A flowchart illustrating the steps of a method for analyzing the industrial linkages between cities based on industrial migration data, provided in this application embodiment; Figure 2 This is a schematic diagram of the upstream segment of the industrial chain in the industrial chain map provided in the embodiments of this application; Figure 3 This is a schematic diagram of the midstream segment of the industry chain in the industry chain map provided in the embodiments of this application; Figure 4 This is a schematic diagram of the downstream segment of the industrial chain in the industrial chain map provided in the embodiments of this application; Figure 5 A schematic diagram of the industry code standard matching table provided for embodiments of this application; Figure 6 This is a schematic diagram of a portion of the data table in the industry dynamics dataset provided in an embodiment of this application; Figure 7 A schematic diagram of the industrial transfer layer provided in the embodiments of this application; Figure 8 A schematic diagram of the leading industry investment layer provided in the embodiments of this application; Figure 9 A schematic diagram of the branch layer for the industry provided in this application embodiment; Figure 10 This is a schematic diagram of the structure of a system for analyzing the industrial correlation between cities based on industrial migration data, provided in an embodiment of this application. Detailed Implementation

[0021] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0022] 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 invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0023] Example 1: This embodiment provides a method for analyzing the industrial linkages between cities based on industrial migration data. (See also...) Figure 1 The method includes the following steps: Step S1: Collect industry classification data for the first target city and the second target city, and construct an industry chain map based on the industry classification data; Step S2: Design data fields based on the industry chain map using a preset structured field model, and construct an industry dynamic dataset based on the data fields; Step S3: Calculate the industrial linkage strength between different target cities based on the industrial dynamic dataset, and generate a visualization map of the spatial pattern of industrial linkage between different target cities based on the industrial linkage strength.

[0024] In a preferred embodiment, step S1, the process of constructing the industry chain map based on the industry classification data, includes: The industry code standardization matching of the industry classification data of the first target city and the industry classification data of the second target city is performed using the national economic industry classification rules to obtain the first industry industry map and the second industry industry map. The first industry industry map and the second industry industry map are divided and reconstructed according to the three links of upstream, midstream and downstream to obtain the industry chain map.

[0025] Specifically, we will separately analyze the industry landscape of city clusters A and B, including sectors such as nuclear technology applications, power transmission and transformation equipment, next-generation information technology, mobile internet, non-ferrous metals and new alloys, advanced steel materials, brine chemicals, new materials, new energy, green agricultural products and food processing, modern logistics, cultural tourism, biomedicine, textiles and apparel, and watchmaking. The industry landscape is broken down into upstream, midstream, and downstream segments, resulting in the following industry chain diagram: Figures 2-4 As shown.

[0026] The specific industry chain division criteria are as follows: the upstream is basic materials and core facilities, with all items belonging to the raw material supply or infrastructure links as the basis for judgment; the midstream is core equipment and conversion production, with special equipment manufacturing and technology-intensive conversion production as the basis for judgment; and the downstream is multi-field application scenarios, with direct access to end-user scenarios as the basis for judgment.

[0027] In a preferred embodiment, based on the aforementioned industry chain map, the industries in the upstream, midstream, and downstream segments are further subdivided into multiple levels until each level of industry is matched with an industry sub-category code in the National Economic Industry Classification. The process includes: When a specified industry has a clear classification within its industry and can be matched with the standard national economic industry name, the classification result is directly matched; When a specified industry has a clear classification within its respective sector but cannot be directly matched, text similarity calculation is used to find the most similar industry in the knowledge base and replace the keywords. When a designated industry is not clearly classified within its respective sector, natural language processing technology is used to identify industry segments or entities from the industry description, match them with the national economic industry classification standards, and assign industry codes.

[0028] Specifically, the method for three-level classification and name code matching of the automation industry is as follows: 1. When the industry has a clear classification and can be matched with the standard national economic industry name, the classification result is obtained according to the above classification standards. 2. When the industry has a clear classification but cannot be matched with the standard national economic industry name, text similarity calculation (based on the cosine similarity of the industry name) is used to find the most similar industry in the knowledge base, and keyword replacement is performed based on the name of that industry. 3. When the industry has no clear classification, NLP technology is used to identify possible industry links (such as "raw materials", "equipment manufacturing", "application services", etc.) or entities in the industrial chain (such as upstream materials, midstream manufacturing, downstream applications, etc.) from industry descriptions, industry terms, or existing text, and match them with the national economic industry classification standards (pre-loaded industry classification table) to assign industry codes. The generated industrial chain division is output in tabular form (including industry links, first-level subdivisions, second-level subdivisions, national economic industry names, and codes), then verified using a knowledge graph, and finally manually reviewed and corrected the verification results. The above method yields a relatively accurate three-level classification of the industry and its corresponding industry codes.

[0029] Furthermore, the first-level sub-sectors obtained from the breakdown of the industrial chain are further subdivided into third-level sub-sectors, and standard national economic industry names and corresponding industry codes are matched for each third-level industry. For example, the nuclear fuel industry in the upstream segment of the nuclear technology application industry is further subdivided into nuclear fuel processing and other non-metallic mineral product manufacturing, with corresponding industry codes 2530 and 3099, respectively. A detailed industry code standard matching table is available, such as... Figure 5 As shown.

[0030] Understandably, by standardizing and matching industry codes through the national economic industry classification rules, we can obtain the first industry industry map and the second industry industry map. Then, by splitting and reconstructing the first industry industry map and the second industry industry map according to the upstream, midstream and downstream links, we can obtain the industry chain map. This can integrate multi-source industrial migration big data and achieve fine-grained analysis of the upstream, midstream and downstream links of the industry chain.

[0031] As a preferred embodiment, in step S2, the data fields designed according to the industrial chain map using a preset structured field model include, but are not limited to, various combinations of industrial transfer statistics fields, enterprise investment statistics fields, enterprise branch statistics fields, enterprise survival statistics fields, established enterprise statistics fields, and defunct enterprise statistics fields.

[0032] Specifically, the statistical fields for industrial transfer include: the location of the industry transfer, the city codes of the transfer location, the year of transfer, the name of the major category, sub-category, and minor category of the industry, the industry code, and the number of transferred enterprises.

[0033] The enterprise investment statistics fields include: the location of the investing and invested enterprises, the location code, the name of the major category, sub-category and minor category of the industry, the industry code, the number of investing enterprises, the year of investment, the number of invested enterprises, and the number of investments.

[0034] The statistical fields for enterprise branch offices include: enterprise location, location code, name of sub-category within industry category, industry code, number of headquarters enterprises, year of establishment of branch offices, location of branch offices, name of sub-category within industry category of branch offices, number of branch offices, and number of times branch offices have been established.

[0035] The enterprise survival statistics fields include: enterprise location, location code, operating status, industry category name, industry code, and number of enterprises.

[0036] The statistical fields for newly established enterprises include: enterprise location, location code, establishment date, industry category name, industry code, and number of enterprises.

[0037] The statistical fields for defunct enterprises include: enterprise location, location code, defunct date, industry category name, industry code, and number of enterprises.

[0038] In the process of designing data fields based on the industry chain map using a preset structured field model, all data are embedded with a three-dimensional core dimension of "industry code-administrative division code-time", which supports cross-theme data correlation analysis. For example, investment data and data of defunct enterprises can be directly linked through "industry code + year".

[0039] In a preferred embodiment, during the process of constructing the industry dynamic dataset based on the data fields, the industry dynamic dataset is constructed with industry classification code-administrative division code-time as the core dimensions to support the related query and analysis of different data field themes.

[0040] In a preferred embodiment, based on the aforementioned organized fields, relevant data such as the location of industry transfer in / out, number of investing companies, investment year, and branch office location are intelligently collected for each industry. Real-time capture of company establishment / deregistration / branch office information is automatically matched with the aforementioned fields; cross-regional investment flow data is obtained through an API gateway, dynamically populating investment field groups; and operating status and industry transfer description text are extracted and converted into structured fields.

[0041] Furthermore, the unstructured data such as the sub-category name code and city of the corresponding industry category are processed. An industry terminology-code conversion knowledge graph is constructed based on the S1 method described above. The result is: input "New Energy Battery Material Manufacturing", output {Industry Category Code: C30, Sub-category Code: C309, Minor Category Code: C3099} Furthermore, outlier handling and data aggregation across various industries and scenarios are addressed. An extensible quality rule library is defined for data in each field category. For example, rules are established for the industrial transfer statistics field: the total number of enterprises must be a positive integer, and the transfer direction must include two different city name fields, etc.

[0042] Finally, the industry dynamic dataset constructed from all fields was automatically imported into the database (i.e., imported into data blocks in the ArcGIS spatial geographic information platform). The resulting data table is as follows: Figure 6 As shown.

[0043] In a preferred embodiment, step S3, which calculates the industrial correlation strength between different target cities based on the industrial dynamic dataset, includes: An industrial linkage network between city nodes is constructed based on the aforementioned industrial dynamics dataset. Specifically, data tables containing information on industrial transfer, investment, branches, existence, establishment, and demise, compiled from the industrial dynamics dataset, are imported into the ArcGIS spatial geographic information platform. Subsequently, an industrial linkage network between city nodes is constructed based on the data on industrial transfer, investment, branches, existence, establishment, and demise. Then, based on the aforementioned industry linkage network, the intensity of industry linkage between cities is calculated using a preset linkage analysis model.

[0044] In a preferred embodiment, the correlation analysis model uses a weighted algorithm to calculate the industrial correlation strength value between cities, and the expression is as follows: R[ab]= α×(T[ab] / T[a]) + β×(I[ab] / I[a]) + γ×(B[ab] / B[a]) Where R[ab] represents the industrial linkage strength from city a to city b; T[ab] represents the number of industrial transfers from city a to city b; T[a] represents the total number of industrial transfers generated by city a; I[ab] represents the total investment amount from city a to city b; I[a] represents the total investment amount from city a to all cities; B[ab] represents the number of branch offices established by corporate headquarters in city a to branch offices in city b; B[a] represents the total number of corporate headquarters established in city a; α, β, and γ all represent industry characteristic weight coefficients.

[0045] In a preferred embodiment, the industry characteristic weight coefficients α, β, γ are configured differently according to the technology intensity, capital intensity, or supply chain characteristics of the target industry; for example, the coefficients for the new energy industry are 0.4, 0.3, and 0.3, respectively; and the coefficients for the traditional manufacturing industry are 0.2, 0.5, and 0.3, respectively.

[0046] In this embodiment, after inputting the names of city cluster A and city cluster B, the administrative boundary clipping of the provinces corresponding to city cluster A and city cluster B can be automatically output to obtain the provincial base map range that highlights the connectivity between cities.

[0047] In this embodiment, the city clusters that are connected under different circumstances are further cropped according to administrative boundaries to automatically obtain adaptive boundary display of city cluster A and city cluster B that highlight industrial clusters.

[0048] As a preferred embodiment, the ArcGIS spatial information platform is used to generate a visualization map of the spatial pattern of industrial associations between different target cities based on the industrial association strength. Furthermore, during the layer design process, the correlation analysis results from the ArcGIS spatial geographic information platform are used as a basis for: Industrial Transfer Layer: Automatically generates display elements such as cities of origin and destination for enterprise transfers, total transfer volume, and transfer path. For destination cities, different circle sizes represent different total data volumes, colors distinguish between the origin and destination, arrows indicate the transfer path, and different arrow colors distinguish the total transfer volume. Standardized visual display rules are used, such as: arrow width W=5+15×log10(T[ab]); color gradient: red represents outbound transfer, using RGB(255, 200-50×T[ab] / T[a], 200); blue represents inbound transfer, using RGB(200, 200, 255-50×T[ab] / T[a]), resulting in a layer like this. Figure 7 As shown.

[0049] Leading Industry Investment Layer: This layer automatically generates a display of leading industry investment information, showcasing the location, number, and investment connections of invested and invested companies. City locations are represented by circles of varying sizes to indicate different data volumes. Colors differentiate between investing and invested companies, and lines represent connections between companies, with line thickness indicating the degree of connection. Standardized visual display rules are applied, such as: the transparency of the connection line is K = 0.3 + 0.7 × (I[ab] / max(I)). 2 The final layer is as follows Figure 8 As shown.

[0050] The industry-specific branch layer automatically generates display elements showing the number, location, and connectivity of headquarters and branches. The number of companies is indicated by circles of varying sizes representing different cities, colors differentiate between headquarters and branches, and lines represent the connectivity between them. The final layer looks like this. Figure 9 As shown.

[0051] Understandably, a correlation analysis model is pre-set to comprehensively consider multiple dimensions such as migration, cooperation, and survival, so as to more closely reflect the complex process of industrial operation in real life. Based on the dynamic industrial dataset, a more comprehensive and objective analysis result is given. Then, by making full use of ArcGIS's spatial analysis and visualization advantages, the abstract and complex industrial connections between cities are presented in a highly intuitive and clear multi-dimensional map form, which greatly improves the efficiency of understanding the results and the value of decision support.

[0052] In this embodiment, by collecting industry classification data from different cities or urban clusters and constructing an industry chain map based on the industry classification data, it is possible to distinguish specific links in the industry chain throughout the entire process of data analysis field design, data aggregation and processing, and visualization. For the industry chain map, a pre-set structured field model is used, and the data field design is carried out based on the industry chain map using the structured field model, thereby further refining the classification of the collected industry classification data. This can accurately and intuitively reveal the spatiotemporal evolution of industrial transfer, cooperation, and survival relationships between cities. The industrial correlation strength between different target cities is calculated based on the industrial dynamic dataset, and a visualization map of the spatial pattern of industrial correlation between different target cities is generated based on the industrial correlation strength. This enables fine-grained analysis of the upstream, midstream, and downstream links of the industry chain, thereby providing real-time, multi-dimensional, and operable decision-making basis for regional industrial collaborative development, investment promotion, and policy formulation.

[0053] Example 2: This embodiment provides a system for analyzing the industrial linkages between cities based on industrial migration data. (See also...) Figure 10 The system includes: The data acquisition module is used to collect industry classification data for the first and second target cities. The graph construction module is used to construct an industry chain graph based on the industry classification data. The field design module is used to design data fields based on the industry chain map using a preset structured field model, and to construct an industry dynamic dataset based on the data fields. The correlation analysis module is used to calculate the industrial correlation strength between different target cities based on the industrial dynamic dataset, and to generate a visualization map of the spatial pattern of industrial correlation between different target cities based on the industrial correlation strength.

[0054] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for analyzing the industrial linkages between cities based on industrial migration data, characterized in that, The method includes the following steps: Collect industry classification data for the first and second target cities, and construct an industry chain map based on the industry classification data; The data fields are designed based on the industry chain map using a pre-defined structured field model, and an industry dynamic dataset is constructed based on the data fields. The industrial linkage strength between different target cities is calculated based on the industrial dynamic dataset, and a spatial pattern visualization map of industrial linkage between different target cities is generated based on the industrial linkage strength.

2. The method for analyzing the industrial correlation between cities based on industrial migration data according to claim 1, characterized in that, The process of constructing a supply chain map based on the industry classification data includes: The industry code standardization matching of the industry classification data of the first target city and the industry classification data of the second target city is performed using the national economic industry classification rules to obtain the first industry industry map and the second industry industry map. The first industry industry map and the second industry industry map are divided and reconstructed according to the three links of upstream, midstream and downstream to obtain the industry chain map.

3. The method for analyzing the industrial correlation between cities based on industrial migration data according to claim 2, characterized in that, Based on the aforementioned industry chain map, the industries in the upstream, midstream, and downstream segments are further subdivided into multiple levels until each industry in each level is matched with an industry subclass code in the National Economic Industry Classification. The process includes: When a specified industry has a clear classification within its industry and can be matched with the standard national economic industry name, the classification result is directly matched; When a specified industry has a clear classification within its respective sector but cannot be directly matched, text similarity calculation is used to find the most similar industry in the knowledge base and replace the keywords. When a designated industry is not clearly classified within its respective sector, natural language processing technology is used to identify industry segments or entities from the industry description, match them with the national economic industry classification standards, and assign industry codes.

4. The method for analyzing the industrial correlation between cities based on industrial migration data according to claim 1, characterized in that, The data fields designed based on the industry chain map using a preset structured field model include, but are not limited to, various combinations of: industry transfer statistics fields, enterprise investment statistics fields, enterprise branch statistics fields, enterprise survival statistics fields, newly established enterprise statistics fields, and defunct enterprise statistics fields.

5. The method for analyzing the industrial correlation between cities based on industrial migration data according to claim 4, characterized in that, In the process of constructing the industry dynamic dataset based on the aforementioned data fields, the industry dynamic dataset is constructed with industry classification code-administrative division code-time as the core dimensions to support the related query and analysis of different data field themes.

6. The method for analyzing the industrial correlation between cities based on industrial migration data according to claim 1, characterized in that, The process of calculating the industrial linkage strength between different target cities based on the aforementioned industrial dynamics dataset includes: Construct an industry association network among city nodes based on the aforementioned industry dynamics dataset; Based on the aforementioned industry linkage network, the intensity of industry linkages between cities is calculated using a pre-defined linkage analysis model.

7. The method for analyzing the industrial correlation between cities based on industrial migration data according to claim 6, characterized in that, The correlation analysis model uses a weighted algorithm to calculate the strength of industrial correlations between cities, and the expression is as follows: R[ab]= α×(T[ab] / T[a]) + β×(I[ab] / I[a]) + γ×(B[ab] / B[a]) Where R[ab] represents the industrial linkage strength from city a to city b; T[ab] represents the number of industrial transfers from city a to city b; T[a] represents the total number of industrial transfers generated by city a; I[ab] represents the total investment amount from city a to city b; I[a] represents the total investment amount from city a to all cities; B[ab] represents the number of branch offices established by corporate headquarters in city a to branch offices in city b; B[a] represents the total number of corporate headquarters established in city a; α, β, and γ all represent industry characteristic weight coefficients.

8. The method for analyzing the industrial correlation between cities based on industrial migration data according to claim 6, characterized in that, The industry characteristic weighting coefficients α, β, γ are configured differently based on the technology intensity, capital intensity, or supply chain characteristics of the target industry.

9. The method for analyzing the industrial correlation between cities based on industrial migration data according to claim 1, characterized in that, The ArcGIS spatial geographic information platform is used to generate a visualization map of the spatial pattern of industrial associations between different target cities based on the strength of industrial associations.

10. A system for analyzing the industrial linkages between cities based on industrial migration data, characterized in that, The system includes: The data acquisition module is used to collect industry classification data for the first and second target cities. The graph construction module is used to construct an industry chain graph based on the industry classification data. The field design module is used to design data fields based on the industry chain map using a preset structured field model, and to construct an industry dynamic dataset based on the data fields. The correlation analysis module is used to calculate the industrial correlation strength between different target cities based on the industrial dynamic dataset, and to generate a visualization map of the spatial pattern of industrial correlation between different target cities based on the industrial correlation strength.

Images