Green watershed policy collaboration analysis system and method

Abstract

The invention relates to the technical field of ecological environment policy assessment, in particular to a green watershed policy collaboration analysis system and method, and the system comprises a text collection module, a text datamation module, a data processing module, a visual analysis module and a data storage module. Policy texts are obtained and screened to generate an original data set, and a four-dimensional target classification framework and a four-class tool division system are adopted to generate a structured basic database containing efficacy hierarchies; and calculating the weighted co-occurrence frequency of policy nodes based on the efficacy weight, analyzing the node degree and the network collaboration degree, and visually displaying the collaboration relationship through a force-oriented graph and a thermal matrix. The method comprises five steps of text collection, datamation, correlation analysis, visualization and hierarchical storage. Through a multi-dimensional analysis framework, automatic text processing, a dynamic weighting mechanism and visual interaction, the problems that an existing method is insufficient in quantification, low in efficiency, poor in adaptability and the like are solved, and scientific decision support is provided for policy optimization of the green watershed.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of ecological environment policy evaluation, and particularly relates to a green watershed policy synergy analysis system and method. BACKGROUND

[0002] With the deepening of the concept of ecological environment protection and sustainable development, green watershed construction has become an important part of ecological civilization construction. In the green watershed policy system, policy synergy directly affects resource allocation efficiency and management effectiveness. Scientific evaluation of the synergy between policy objectives and tools is a key prerequisite for optimizing the policy system and improving the overall effectiveness of the green watershed management support system. However, the existing technology has the following technical problems in the field of green watershed policy synergy analysis: (1) Lack of multi-dimensional analysis framework covering policy objectives, tools and associated relationships, making it difficult to build a synergy network between policy objectives and tools, resulting in an inability to fully quantify the synergy between policies. For example, traditional methods can only analyze policy objectives or tools in isolation, and cannot evaluate policy node correlation strength through standardized means such as co-occurrence frequency and network topology, leaving synergy analysis at the subjective qualitative level; (2) The classification and quantification of policy texts rely heavily on manual operations, making it difficult to quickly generate structured databases, resulting in low analysis efficiency. At the same time, the existing technology lacks a unified data storage hub, and the data exchange mechanism between modules is missing, making it impossible to seamlessly integrate policy text collection, data processing and analysis, severely restricting system integration; (3) Unable to visually present policy synergy relationships through network topology diagrams, heat maps and other visual forms, and lacking dynamic weighting mechanisms for policy effectiveness levels (such as central and local policies), making it difficult to conduct synergy analysis from multiple dimensions such as the global and sub-network, and unable to meet the needs of dynamic evaluation of green watershed policies; (4) The system is closed in architecture, unable to adjust analysis dimensions or index weights according to actual scenarios, especially lacking self-defined processing functions for policy data, making it difficult to meet the differentiated analysis needs of different management scenarios, limiting the application scope of policy synergy evaluation.

[0003] In view of the above, the present application is proposed. SUMMARY

[0004] In order to solve the problems in the prior art of lacking a multi-dimensional analysis framework, making it difficult to quantify synergy effect analysis and remaining at the qualitative level; heavy reliance on manual operations, lack of unified data storage and module interaction, low processing efficiency and poor integration; lack of visualization and dynamic weighting mechanisms, unable to visually display synergy relationships, and difficult to support multi-dimensional evaluation; closed system architecture, analysis dimensions and weights cannot be adjusted, and lack of self-defined functions and poor adaptability.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: Firstly, a green watershed policy synergy analysis system includes: Text acquisition module: Used to acquire policy texts related to green watersheds, generate raw datasets through multi-source acquisition and text filtering, and output them to the text datafication module; Text datafication module: used to perform semantic parsing on the original dataset, generate a basic database, and store it in the data storage module; Data processing module: used to perform correlation analysis on the basic database, generate collaborative data storage to the data storage module and output it to the visualization analysis module; Visualization and analysis module: used to visualize the collaborative data; Data storage module: Used to store the original dataset, basic database and collaborative data using a hierarchical storage architecture.

[0006] Custom function modules: used to write configuration parameters to the data storage module and trigger recalculation to the data processing module.

[0007] Furthermore, the text acquisition module includes: Data acquisition unit: used to acquire multi-source policy texts and transmit them to the text filtering unit; Text filtering unit: used to deduplicate and filter the types of the multi-source policy texts, and output the filtered valid text data to the text data digitization module.

[0008] Furthermore, the text digitization module includes: Automatic Policy Target Classification Unit: This unit receives valid text output from the text filtering unit, performs target dimension mapping based on a preset classification framework, and transmits the mapping results to the tool partitioning unit. Automatic Policy Tool Segmentation Unit: Used to extract and classify tool types from text after target classification, and transmit the classification results to the data structure unit; Data structuring unit: Used to integrate the results of target classification and tool segmentation, generate a basic database containing effectiveness level weights, and store it in the data storage module.

[0009] Furthermore, the data storage module includes: Storage structure design unit: used to build a hierarchical storage system, receive the basic database of data structure units and establish data association relationships; Data interaction mechanism unit: used to provide a write channel for the text acquisition module, a read / write channel for the data processing module, and a read channel for the visualization analysis module and custom function modules through standardized interfaces.

[0010] Further, the data processing module comprises: The co-occurrence calculation unit is configured to obtain the basic database from the data storage module, calculate the weighted co-occurrence frequency of the policy node pair based on the effectiveness weight, and transmit the result to the network topology analysis unit. The network topology analysis unit is configured to receive the result of the co-occurrence calculation unit, analyze the correlation structure of the policy system, and transmit the calculated node degree and the network average degree to the synergy quantification unit. The synergy quantification unit is configured to integrate the co-occurrence and topology analysis results, generate the synergy index of the node pair and the network level, and store the results in the data storage module.

[0011] The co-occurrence calculation unit is configured to obtain the basic database from the data storage module, calculate the weighted co-occurrence frequency of the policy node pair based on the effectiveness weight, and transmit the result to the network topology analysis unit.

[0012] Wherein, is the co-occurrence frequency of the policy node and , is the total number of policy files, is the weight of the kth policy file, is the policy target and tool node.

[0013] The calculation formula of the node degree is:

[0014] Wherein, is the degree of the node , is the total number of policy target and tool nodes, is the co-occurrence frequency of the node and , the higher the value, the stronger the core degree of the node in the policy network.

[0015] Further, the calculation formula of the network average degree is:

[0016] Wherein, is the network average degree (the average value of the degrees of all nodes, reflecting the overall correlation density of the policy system); is the total number of nodes, is the degree of each node.

[0017] Further, the synergy index includes the network synergy degree, and the calculation formula is:

[0018] Wherein, ​​​a coordination degree of the entire policy network, a total number of nodes, a coordination degree of a node pair, a number of all non-self-loop node pairs; Further, the coordination index includes a coordination degree of different sub-network structures, and a specific calculation formula is as follows:

[0019] wherein, a coordination degree of a sub-network, a number of nodes of the sub-network, a coordination degree of a node pair in the sub-network.

[0020] Further, the visualization analysis module includes: a graph generation unit, configured to obtain the coordination index from the data storage module, and generate a visualization graph based on a force-directed graph and a heat matrix algorithm; an interactive control unit, configured to receive a user operation instruction, drive the graph generation unit to update display content, and obtain the latest analysis result in real time through the data storage module.

[0021] In a second aspect, a green watershed policy coordination analysis method includes the following steps: S1, obtaining green watershed related policy texts, and generating an original data set through multi-source acquisition and text screening; S2, performing semantic analysis on the original data set to generate a basic database; S3, performing correlation analysis on the basic database to generate coordination data; S4, visualizing the coordination data; S5, storing the original data set, the basic database and the coordination data using a hierarchical storage architecture.

[0022] Compared with the prior art, the green watershed policy synergy analysis system and method provided by the application realizes the automatic processing and quantitative analysis of policy texts through multi-module synergy: the text acquisition module acquires and screens policy texts to generate an original data set; the text data module generates a structured basic database containing a hierarchy of effectiveness levels by using a four-dimensional target classification framework and a four-category tool division system; the data processing module calculates the weighted co-occurrence times of policy nodes based on effectiveness weights, analyzes node degrees and network synergy degrees, and realizes the quantitative analysis of policy synergy; the visual analysis module visually displays the synergy relationship through a force-directed graph and a heat matrix; the data storage module uses a hierarchical architecture to support full-process data interaction; and the custom function module supports weight adjustment and sub-network customization. The method includes five steps of text acquisition, data, correlation analysis, visualization, and hierarchical storage. The application solves the problems of insufficient quantification, low efficiency, and poor adaptability of existing methods by using a multi-dimensional analysis framework, automatic text processing, a dynamic weighting mechanism, and visual interaction, and provides scientific decision support for green watershed policy optimization. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 The application provides a green watershed policy synergy analysis system. DETAILED DESCRIPTION

[0024] The technical solutions of the application will be clearly described below with reference to the drawings. Obviously, the described embodiments are not all embodiments of the application, and all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the application.

[0025] It should be noted that, unless otherwise specified, the relative arrangement and numerical expression of components and steps described in the embodiments should not be understood as limiting the scope of the application.

[0026] The following description of exemplary embodiments is merely illustrative in nature and is in no way limiting on the application or its application or use. Techniques, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail herein, but should be considered part of the specification if applicable.

[0027] Embodiment one Reference Figure 1 , Figure 1 The application provides a green watershed policy synergy analysis system, and the architecture diagram of the system can specifically include: A1, a text acquisition module: used to acquire green watershed related policy texts to provide original data support for subsequent data processing and analysis. Specifically, it can include: Data collection unit: support manual import and automatic collection dual mode: manual import allows users to import policy text through local file upload (such as PDF, Word format) method, suitable for non-public channel policy file acquisition; automatic collection based on keyword search, access to public policy files through policy database (such as China policy and regulation database) and other platforms, covering central government regulations, local regulatory documents, industry standards and other types.

[0028] Text screening unit: identify duplicate files through text fingerprint algorithm, eliminate documents with more than 90% duplication rate; and determine whether the document belongs to policy text through text classification model, filter out non-policy content such as news reports and academic papers; store the effective policy files screened according to the unified format (including file ID, title, release time, effectiveness level and other metadata), and output the effective file quantity N to the data processing module.

[0029] A2, text data module: used for structured processing of the policy text to generate a basic database; by establishing a standardized policy target and tool classification system, a core index framework supporting collaborative analysis is constructed to provide standardized input for subsequent quantitative calculation and network modeling. Specifically, it can include: A21, policy target automatic classification unit: used for adopting PR pollution reduction ( ), CR carbon reduction ( ), GE green expansion ( ), and G growth ( ) four-dimensional classification framework to realize dimension mapping of text content through keyword matching and semantic analysis technology: presetting keywords such as wastewater treatment and pollutant emission corresponding to PR pollution reduction, low-carbon energy and carbon footprint corresponding to CR carbon reduction, forest coverage rate and wetland protection corresponding to GE green expansion, and green industry and ecological economy corresponding to G growth; the corresponding keywords can be increased according to actual needs.

[0030] With the help of natural language processing technology, the preprocessed policy text is automatically associated with the four-dimensional target framework, without manual intervention to complete the basic classification; at the same time, a single policy text can correspond to multiple target dimensions, and the structured output can meet the needs of identifying the correlation between multiple targets in policy collaborative analysis. Specifically, it can include: A211. Pre-training model construction: Use pre-trained language models such as BERT to fine-tune using manually annotated four-dimensional target classification datasets (covering real policy cases, such as annotating "vigorously developing photovoltaic industry to promote energy structure transformation" as CR carbon reduction, "promoting wetland ecological restoration project" as GE green expansion, etc.) to build a policy target classification base model. This model breaks through the surface limitations of traditional keyword matching and can identify the deep logic of "means-object-target" in the text (such as the causal chain of "circular transformation (means) → reduction of pollutant emissions (object) → PR pollution reduction (target)"). It can effectively distinguish the target attributes of expressions such as "energy transformation" in different contexts (such as determining whether it is a "CR carbon reduction" or a "G growth" industry support scenario in the context). By automatically identifying the deep semantics of the text through the pre-training model, traditional manual classification is replaced, reducing the classification time of a single policy text.

[0031] A212. Pre-set keyword library verification: Relying on an extensible target keyword library (consistent with existing functions, supporting dynamic addition or adjustment according to actual needs), the model's preliminary classification results are checked again to correct semantic ambiguity. For example, the keyword library is as follows: PR pollution reduction is associated with "wastewater treatment, total pollutant emission control, circular transformation"; CR carbon reduction is associated with "low-carbon energy replacement, carbon footprint accounting, photovoltaic industry"; GE green expansion is associated with "forest coverage rate improvement, wetland ecological restoration, biodiversity"; G growth is associated with "green industry cluster, ecological product value realization, circular economy". Through verification, misclassification caused by semantic ambiguity can be avoided.

[0032] A213. Multi-label classification and probability decision: For complex scenarios where a single policy text contains multiple target dimensions (such as a policy involving both "industrial pollution control (PR pollution reduction)" and "low-carbon technology subsidies (CR carbon reduction)"), the model outputs probability values for each target dimension (such as PR pollution reduction probability 0.92, CR carbon reduction probability 0.88), and filters valid target dimensions through a pre-set confidence threshold.

[0033] Classification results below the threshold are marked for manual verification to ensure the reliability of the output results. At the same time, the threshold parameter can be adjusted through the system's custom function module to adapt to the accuracy requirements of different analysis scenarios.

[0034] A22. Policy tool automatic division unit: used to define four types of tools: A government guidance, B market incentives, C public participation, and D digital empowerment. Through syntactic analysis, the tool description in the text is extracted and classified into: government guidance (A), market incentives (B), public participation (C), and digital empowerment (D) through a rule engine or fine-tuning model.

[0035] A221, Build a tool feature library: For the four types of tools, "government guidance, market incentives, public participation, and digital empowerment", establish an extensible feature library (including core words and extended words), for example: Government guidance: Core words "regulations, planning, approval, supervision, standards"; extended words "mandatory, order, assessment, supervision" and others; Market incentives: Core words "funds, subsidies, transactions, finance, taxes"; extended words "carbon quotas, green credit, price mechanism" and others; Public participation: Core words "propaganda, reporting, supervision, popular science, community"; extended words "public opinion collection, volunteer service, public notice" and others; Digital empowerment: Core words "monitoring, big data, sensors, platforms, digitalization"; extended words "smart supervision, blockchain, algorithm analysis" and others.

[0036] A222, Develop matching rules engine: Based on the frequency of key words and syntax structure in the text, set three layers of matching rules: First layer: Accurate matching of core words (such as the text containing "regulations" or "planning", directly triggering the "government guidance" type label); Second layer: Extended word combination matching (such as "carbon quota + transaction" combination, judged as "market incentives"); Third layer: Syntax logic verification (judged by subject-predicate-object structure, such as "government + develop + standards" → "government guidance"; "public + report + pollution" → "public participation").

[0037] A223, Conflict processing mechanism: When multiple types of tool feature words appear in the text at the same time (such as "government subsidies + public supervision"), sort by core word weight (core word weight is higher than extended word), take the tool type with the highest weight as the main classification, and the rest as secondary classification (such as "subsidies" belong to market incentives core words, priority higher than "supervision" of public participation extended words). The final output of the classified content is: Government guidance (A): including regulations mandatory (such as developing watershed pollution emission standards), planning approval (such as implementing green watershed construction planning) and other administrative means; Market incentives (B): covering green finance (such as setting up an ecological protection special fund), carbon trading (such as establishing a watershed carbon emissions trading mechanism) and other economic levers; Public participation (C): including environmental protection propaganda (such as carrying out watershed protection popular science activities), community supervision (such as establishing a public pollution reporting platform) and other forms of social force participation; Digital empowerment (D): involving smart monitoring (such as deploying a watershed water quality sensor network), big data platform (such as building a green watershed management cloud platform) and other technical tools.

[0038] A23, data structuring unit: for generating a structured basic database, specifically containing the following core fields: file ID, target classification, tool type, and effectiveness level; wherein the file ID records a unique identifier for each policy document, and associates the original text storage path; the target classification records a label set mapped to the four-dimensional target of PR pollution reduction, CR carbon reduction, etc. (supporting single file multi-target); the tool type records a label set corresponding to A-D type policy tools (supporting multi-tool combination); the effectiveness level considers the weight according to the release subject level and policy document type, providing a weighted basis for subsequent co-occurrence analysis.

[0039] The generated database adopts a relational storage structure, supports standardized interface calls with the data processing module, and ensures that the index data required for quantitative analysis can be efficiently read and calculated.

[0040] A3, data storage module: for storing policy texts and basic databases, and serving as a data interaction hub; responsible for the unified storage, management, and sharing of data throughout the process, and realizing seamless docking of text collection, data processing, visualization analysis, and other modules through standardized data structures and interaction interfaces, ensuring real-time flow and collaborative calling of data at each link. Specifically, it can include: A31, storage structure design unit: for adopting a MySQL relational database to build a three-layer data storage system, and establishing core tables and associated relationships: A311, policy text table: saves policy text raw data, provides classification basis for the text data module, and provides file weight reference for co-occurrence analysis. Storage fields: file ID, file title, release time, effectiveness level, text content, and metadata such as collection source URL.

[0041] A312, target tool classification table: maps policy texts and target tool classification systems, forms a structured basic database, and supports the associated analysis of the data processing module. Storage fields: file ID, policy target label (PR pollution reduction, CR carbon reduction, etc.), policy tool label (A government guidance, B market incentive, etc.), and classification confidence output.

[0042] A313, collaborative data table: real-time storage of quantitative analysis results output by the data processing module, providing graphical rendering data for the visualization module, and providing secondary analysis basis for the custom function module. Node pair ID ( , ), co-occurrence frequency ( ), co-occurrence frequency ( ), node degree ( ), collaboration degree ( 、 、 ), and calculation timestamp.

[0043] A32. Data Interaction Mechanism Unit: Provides a write interface for the text acquisition module, a read / write interface for the data processing module, and a read interface for the visualization module and custom function modules.

[0044] A transaction mechanism is adopted to ensure the consistency of data operations across modules. For example, after the data processing module completes a collaborative calculation, it automatically updates the collaborative data table by triggering a stored procedure and simultaneously notifies the visualization module to refresh the data. Real-time incremental updates are supported to avoid the performance loss caused by full data interaction, which is suitable for scenarios of continuous collection of policy texts.

[0045] A4. Data Processing Module: Used to perform correlation analysis on the basic database and generate collaborative data; specifically, it may include: A41. Co-occurrence Analysis Unit: Used to quantify the correlation strength between policy objectives / tools and generate a co-occurrence matrix. The formula for calculating the number of co-occurrences is:

[0046] in, As a policy node and The number of times they co-occur (e.g., the frequency of association between PR pollution reduction and government guidance); The total number of policy documents (the number of valid documents output by the text acquisition module). Weights for the k-th policy document (e.g., assigned according to level of authority: regulations) Regulations wait); For policy objectives / tool ​​nodes (such as PR pollution reduction, GE green expansion and government guidance, etc.); For each policy document The Boolean function Buller is used to determine whether a policy node is also included. (such as PR pollution reduction) and (e.g., government guidance): Return 1 if included, otherwise return 0. This is combined with the document's hierarchical weighting. (For example, national-level regulations have w=1, provincial-level regulations have w=0.8), and the weighted co-occurrence frequency of node pairs is accumulated. This mechanism gives higher weight to central policies in association analysis than local policies, enhancing the authority of the results.

[0047] By dividing by the total number of valid documents To eliminate the influence of dataset size on association strength, the formula for calculating co-occurrence frequency (association strength normalization) is as follows:

[0048] in, for and co-occurrence frequency (eliminate the difference of total number of files, correlation strength), co-occurrence frequency, total number of policy files.

[0049] A42, the degree of the network as a whole: for the perspective of network structure, analyze the correlation breadth and system density of policy nodes; the degree of network nodes (single node correlation breadth) calculation formula is:

[0050] wherein, the degree of node (the sum of co-occurrence frequency with all other nodes, measure the importance of node); total number of nodes of policy goals and tools (such as PR pollution reduction, CR carbon reduction, etc. Co nodes); co-occurrence frequency of node and . The higher the value, the stronger the core degree of the node in the policy network.

[0051] Network average degree (global correlation density), calculation formula is:

[0052] wherein, network average degree (the average of the degree of all nodes, reflect the overall correlation density of policy system); total number of nodes, the degree of each node.

[0053] A43, the coordination degree calculation unit: quantifying the coordination level of policy nodes and system, support global / local analysis; inter-node coordination (correlation strength normalization), calculation formula is:

[0054] wherein, the coordination of node and (co-occurrence frequency normalized to [0, 1] interval, highlight strong correlation), co-occurrence frequency, the maximum value of all node pairs co-occurrence frequency (global maximum correlation strength).

[0055] Network coordination degree (system level coordination level), calculation formula is:

[0056] wherein, the coordination degree of the whole policy network (the average of all node pairs coordination, measure the coordination level of system), This represents the total number of nodes. For node pair collaboration, This represents the number of all non-self-linked node pairs (excluding nodes that are themselves related).

[0057] The formula for calculating the degree of synergy among different sub-network structures (local synergy analysis) is as follows:

[0058] in, For the degree of synergy of sub-networks (policy objectives only or tool sub-networks only), The number of nodes in the subnetwork. This refers to the coordination between node pairs within a subnetwork.

[0059] Input: Basic data for this module , , , The policy objectives / tools classification results come from the text digitization module, and the policy text library comes from the data storage module.

[0060] Output: Calculation results , , , Collaborative data is stored in the data storage module and can be accessed by the visualization analysis module (for drawing collaborative network diagrams and heat matrices) and the custom function module (for weight adjustment and sub-network analysis).

[0061] A5. Visualization Analysis Module: Used to visualize collaborative data; through graphical and interactive technologies, it transforms the quantitative data on policy synergy generated by the data processing module (such as co-occurrence frequency, node degree, synergy degree, etc.) into intuitive and visual information carriers, helping users understand the strength of the correlation between policy objectives and tools and the level of system synergy, providing visualized decision support for the optimization of green watershed policies. Specifically, it may include: A51, Visualization Unit: Used for development based on the D3.js open-source visualization framework, employing the force-directed graph algorithm (…). Construct a policy node association network; including node design and edge design: Node diameter and node size in node design Proportional (e.g.) =1.0 corresponds to a diameter of 20px. =2.0 corresponds to a diameter of 40px), intuitively reflecting the core position of the node in the policy system; the color depth of the edges in the design relates to the node's synergy. Positive correlation (e.g.) ≥0.7 is displayed in red, 0.4≤ <0.7 is displayed in orange. <0.4 shows gray), the thickness of the edge is proportional to the co-occurrence frequency , facilitating the rapid identification of strongly associated node pairs.

[0062] The heat matrix adopts a two-dimensional matrix layout, with rows and columns corresponding to policy objectives and tool nodes (such as an 8x8 matrix); the color gradient of the matrix cells (from white to dark red) maps the normalized value (0~1) of the co-occurrence frequency , for example =0.5 shows moderate red, =0.9 shows dark red; the matrix cell numerical label is displayed, and clicking the cell can link to highlight the corresponding node pair edge in the collaborative network graph, realizing the bidirectional mapping of the matrix network.

[0063] A52, interactive operation unit: support filtering nodes by policy objectives (PR pollution reduction, CR carbon reduction, etc.) or tool types (government guidance, market incentives, etc.), such as displaying only CR carbon reduction related nodes and edges; mouse wheel zooms the network graph, double-clicking a node can focus on the node and expand its associated nodes.

[0064] Real-time information display: when the mouse hovers over a node, a pop-up window displays detailed information about the node, including: node name (such as GE green expansion), classification type (objective / tool); node degree value and ranking; top 3 list of co-occurrence frequency with other nodes. Clicking on the node can trigger the details panel to display the co-occurrence frequency , synergy and the list of policy documents belonging to the node.

[0065] Data support exports the visualization results to PNG, SVG format pictures, or interactive HTML files; it also provides a sharing link to support cross-device viewing of real-time updated collaborative visualization results (requires permission verification). In addition, it can retrieve , , , etc. from the collaborative data table in the data storage module.

[0066] A6, custom function module: used for customized processing of data; through open parameter adjustment, sub-network customization, and analysis dimension expansion interface, it supports users to customize analysis logic according to actual needs (such as different policy characteristics of different river basins, special evaluation objectives), improves the adaptability of the system to different scenarios, and provides support for personalized collaborative evaluation of green river basin policies. Specifically, it can include: A61, parameter adjustment unit: provides a visualization weight configuration interface, users can batch adjust weights according to policy effectiveness levels (such as national, provincial, and municipal levels) or file types (laws, regulations, and normative documents) .

[0067] For example: adjust the default weight of provincial regulations from 0.8 to 0.9, increase the influence weight of local policies in co-occurrence analysis; set a weight coefficient of 1.1 for specific field documents (such as Yangtze River Basin special policies) to highlight regional policy coordination analysis requirements.

[0068] A62, sub-network customization unit: for multi-dimensional filtering conditions and real-time recalculation. Multi-dimensional filtering includes selecting central policies, provincial policies, municipal policies or combined levels through a drop-down menu, and the system automatically filters policy documents of the corresponding effective level (A61) Filter by level), recalculate sub-network coordination .

[0069] After sub-network filtering, the filtering conditions and weight parameters are passed to the co-occurrence analysis unit, and the weight parameter passed is the latest value set by the user through the parameter adjustment unit , real-time update coordination data; specifically, the system automatically triggers the recalculation process of the data processing module: filter the policy text library and target tool classification table according to the filtering conditions; recalculate the co-occurrence frequency , co-occurrence frequency ; generate new node degree and coordination degree , synchronize the network graph and heat matrix of the visualization module.

[0070] A63, analysis dimension expansion unit: for time axis dynamic analysis and custom index addition. Support policy file dimension splitting by year, quarter, and month to generate time series coordination trend charts.

[0071] For example: select a time interval, the system calculates the global coordination degree of each period by quarter , draws a line chart to show the dynamic changes of policy system coordination; compare the sub-network coordination degrees of different years to analyze policy optimization effects. Time dimension calculation supports weight accumulation: when calculating by quarter, the weight of each quarter file remains independent to ensure the comparability of cross-period data.

[0072] This unit provides API interfaces for users to import custom analysis indicators (such as policy tool coverage and target achievement rate) for linkage analysis with existing coordination indicators.

[0073] For example: after importing the carbon trading policy coverage indicator, the system can analyze the correlation between this coverage and the CR carbon reduction-market incentive coordination degree, and generate a scatter plot to assist decision-making.

[0074] The module reads the metadata (release time, validity level, etc.) of the policy text library through the API of the data storage module for screening, and writes the user-defined weight parameters and sub-network configuration into the system configuration table.

[0075] In a second aspect, a green watershed policy coordination analysis method includes the following specific steps: S1, obtaining green watershed related policy texts, generating an original data set through multi-source acquisition and text screening; S2, performing semantic analysis on the original data set to generate a basic database; S3, performing correlation analysis on the basic database to generate coordination data; S4, visualizing the coordination data; S5, storing the original data set, basic database and coordination data using a hierarchical storage architecture.

[0076] In summary, the present application has the following advantages: (1) By constructing a three-dimensional coordination analysis framework, the data processing module realizes multi-dimensional quantitative analysis, forming a standardized quantitative system from a single node to a system level, and realizing scientific evaluation of policy coordination; (2) The text data module automatically maps policy texts to generate a structured basic database, and through the interface with the data storage module, a standardized data interaction process is constructed to improve the efficiency of manual coding; (3) The visualization interface improves the efficiency of policy coordination relationship presentation, and the dynamic weighting mechanism meets the differentiated analysis needs in different management scenarios; (4) The self-defined function module performs secondary processing on the policy data in the data storage module, and can flexibly adjust the analysis logic according to the characteristics of local policies, special evaluation needs and other scenarios, improving adaptability.

[0077] The above specific embodiments are only used to illustrate the technical solutions of the present application and not to limit it, although the present application has been described in detail with reference to examples, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A green watershed policy synergy analysis system, characterized in that, include: Text acquisition module: Used to acquire policy texts related to green watersheds, generate raw datasets through multi-source acquisition and text filtering, and output them to the text datafication module; Text datafication module: used to perform semantic parsing on the original dataset, generate a basic database, and store it in the data storage module; Data processing module: used to perform correlation analysis on the basic database, generate collaborative data storage to the data storage module and output it to the visualization analysis module; Visualization and analysis module: used to visualize the collaborative data; Data storage module: used to store the original dataset, basic database, and collaborative data using a hierarchical storage architecture; Custom function module: Used to write configuration parameters to the data storage module and trigger recalculation by the data processing module.

2. The green watershed policy synergy analysis system according to claim 1, characterized in that, The text acquisition module includes: Data acquisition unit: used to acquire multi-source policy texts and transmit them to the text filtering unit; Text filtering unit: used to deduplicate and filter the types of the multi-source policy texts, and output the filtered valid text data to the text data digitization module.

3. The green watershed policy synergy analysis system according to claim 1, characterized in that, The text data digitization module includes: Automatic Policy Target Classification Unit: This unit receives valid text output from the text filtering unit, performs target dimension mapping based on a preset classification framework, and transmits the mapping results to the tool partitioning unit. Automatic Policy Tool Segmentation Unit: Used to extract and classify tool types from text after target classification, and transmit the classification results to the data structure unit; Data structuring unit: Used to integrate the results of target classification and tool segmentation, generate a basic database containing effectiveness level weights, and store it in the data storage module.

4. The green watershed policy synergy analysis system according to claim 1, characterized in that, The data storage module includes: Storage structure design unit: used to build a hierarchical storage system, receive the basic database of data structure units and establish data association relationships; Data interaction mechanism unit: used to provide a write channel for the text acquisition module, a read / write channel for the data processing module, and a read channel for the visualization analysis module and custom function modules through standardized interfaces.

5. The green watershed policy synergy analysis system according to claim 1, characterized in that, The data processing module includes: Co-occurrence calculation unit: used to obtain the basic database from the data storage module, calculate the weighted co-occurrence frequency of policy node pairs based on the effectiveness weight, and transmit the results to the network topology analysis unit; Network topology analysis unit: used to receive the results of the co-occurrence calculation unit, analyze the policy system association structure, and transmit the calculated node degree and network average degree to the collaborative quantification unit; Co-occurrence metric unit: used to integrate co-occurrence and topology analysis results, generate synergy metrics for node pairs and network layers, and store them in the data storage module.

6. The green watershed policy synergy analysis system according to claim 5, characterized in that, The effectiveness weighting is calculated using the weighted co-occurrence frequency of policy node pairs, and the specific calculation formula is as follows: in, As a policy node and The number of times they co-occur. The total number of policy documents, The weight of the k-th policy document, For policy objectives and tool nodes; The formula for calculating the node degree is: in, For nodes The degree, This represents the total number of nodes for policy objectives and tools. For nodes and co-occurrence frequency, A higher value indicates a stronger centrality of the node within the policy network; The formula for calculating the average degree of the network is: in, The average degree of the network (the average degree of all nodes, reflecting the overall correlation density of the policy system); This represents the total number of nodes. Let be the degree of each node.

7. The green watershed policy synergy analysis system according to claim 5, characterized in that, The coordination index includes network coordination degree, and the specific calculation formula is as follows: in, For the overall coordination of the policy network, This represents the total number of nodes. For node pair collaboration, The number of all non-self node pairs; the formula for calculating the cooperation of the node pairs is: in, For co-occurrence frequencies, This represents the maximum co-occurrence frequency of all node pairs.

8. The green watershed policy synergy analysis system according to claim 5, characterized in that, The synergy index includes the synergy degree of different sub-network structures, and the specific calculation formula is as follows: in, For the degree of cooperation of the sub-network, The number of nodes in the subnetwork. This refers to the coordination between node pairs within a subnetwork.

9. The green watershed policy synergy analysis system according to claim 1, characterized in that, The visualization analysis module includes: The graphics generation unit is used to obtain collaborative indicators from the data storage module and generate visual graphics based on the force-directed graph and heat matrix algorithm. Interactive control unit: Used to receive user operation commands, drive the graphics generation unit to update the displayed content, and obtain the latest analysis results in real time through the data storage module.

10. A method for analyzing the synergy of green watershed policies, characterized in that, The specific steps include: S1. Obtain policy texts related to green watersheds, and generate the original dataset through multi-source acquisition and text filtering; S2. Perform semantic parsing on the original dataset to generate a basic database; S3. Perform correlation analysis on the basic database to generate collaborative data; S4. Visualize the collaborative data; S5. The original dataset, basic database, and collaborative data are stored using a hierarchical storage architecture.

Images