An intelligent policy matching recommendation method

Abstract

The present application relates to the technical field of policy matching, and specifically relates to an intelligent policy matching recommendation method, comprising extracting policy conditions and policy rewards based on policy files, the policy conditions comprising flexible conditions and rigid conditions, obtaining target enterprise information, and comparing the information corresponding to the flexible conditions and the rigid conditions in the target enterprise information with the flexible conditions and the rigid conditions in each policy file one by one; when the number of non-compliant flexible conditions is greater than or equal to 1, calculating the ratio of the quantifiable indicators in the target enterprise information data and the corresponding non-compliant flexible conditions, and when the ratio is greater than 80%, pushing the corresponding policy file and the corresponding policy reward to the user. The present application recommends policies to users in a manner of matching flexible conditions and rigid conditions, and when the number of non-compliant flexible conditions is greater than or equal to 1, the corresponding policy can be continuously recommended, thereby solving the industry pain point of "near-standard enterprises being mis-screened".

Description

Technical Field

[0001] This invention relates to the field of policy matching technology, and in particular to an intelligent policy matching and recommendation method. Background Technology

[0002] In today's society, governments formulate and implement numerous policies to promote economic development, advance social progress, and safeguard people's well-being. These policies cover various fields such as the economy, science and technology, education, healthcare, and environmental protection, and are vast in number and complex in content. With the rapid development of information technology, although many government departments have established e-government systems to digitally store policy information, the efficiency of policy information utilization remains low. For enterprises, timely and accurate access to relevant policy information is crucial. However, due to the dispersed and complex nature of policy information, enterprises often struggle to find policies that meet their specific needs from the vast amount of available information. For example, a technology company may qualify for multiple science and technology innovation support policies, but due to a lack of effective policy matching tools, it may not fully understand these policies, thus missing opportunities to apply for funding, tax incentives, etc. While some policy matching methods exist to address this issue, they typically only match policies that fully meet the criteria, resulting in situations where "closely compliant companies are mistakenly screened," leading to low matching coverage. Summary of the Invention

[0003] In view of this, the purpose of this invention is to propose an intelligent policy matching and recommendation method to solve the problem that currently only policies that fully meet the requirements can be matched, and there is a situation where "enterprises that are close to meeting the requirements are mistakenly screened out", resulting in a low matching coverage.

[0004] To achieve the above objectives, the present invention provides an intelligent policy matching and recommendation method, comprising:

[0005] Based on policy documents, policy conditions and policy rewards are extracted. The policy conditions include flexible conditions and rigid conditions. The flexible conditions are quantifiable indicators, and the rigid conditions are non-quantifiable indicators.

[0006] Obtain target enterprise information, compare the information in the target enterprise information corresponding to the flexible and rigid conditions with the flexible and rigid conditions in each policy document, and obtain the flexible conditions that the target enterprise meets, the flexible conditions that it does not meet, the rigid conditions that it meets, and the rigid conditions that it does not meet in each policy document.

[0007] When the number of flexible conditions and the number of rigid conditions that are not met in the policy documents are both zero, the corresponding policy documents and corresponding policy rewards will be pushed to the user.

[0008] When the number of non-compliant rigid conditions in policy documents is zero, and the number of non-compliant flexible conditions is ≥1, the ratio of the target enterprise information data to the quantifiable indicators in the corresponding non-compliant flexible conditions is calculated. When the ratio is greater than 80%, the corresponding policy documents and corresponding policy rewards are pushed to the user.

[0009] Optionally, when the ratio is greater than 80%, the actual benefit is calculated based on the input cost and the resulting benefit of achieving the non-compliance elasticity condition, and the result is pushed to the user.

[0010] Optionally, when the number of non-compliant flexible conditions in the policy documents is zero and the number of non-compliant rigid conditions is 1, the actual benefit is calculated based on the input cost and the resulting benefit of achieving the non-compliant rigid conditions, and the result is pushed to the user.

[0011] Optionally, the method further includes:

[0012] Establish a database containing the matching results of multiple enterprises with various policy conditions, and mark the position of each enterprise in the industrial chain;

[0013] When a target company is unable to meet the target policy due to a flexibility ratio below 80% or more than one unmet rigid condition, the following steps shall be performed:

[0014] Supply chain node matching: The supply chain nodes of the target enterprise are compared with the supply chain nodes of enterprises in the database to filter out related enterprises located at upstream, same level or downstream nodes;

[0015] Data fusion verification: The data of the selected related enterprises is merged with the data of the target enterprise to generate a simulated merged dataset;

[0016] Policy compliance determination: If the simulated merged dataset meets all the conditions of the target policy, the corresponding related enterprise information in the database will be output.

[0017] Optionally, when multiple related companies that meet the requirements exist in the database, the following rules shall be applied to the output:

[0018] Dynamic weight setting: Allows users to set differentiated primary weight values ​​W for different nodes in the industry chain: upstream / same level / downstream. C ;

[0019] Priority Calculation: The score for each related enterprise is calculated based on the primary weight value. The calculation formula is as follows:

[0020] Score = 100 × W C ;

[0021] Sorting Output: Sort the related companies by overall score from highest to lowest and output the list of related companies.

[0022] When there are multiple qualified related companies, in order to further filter out more suitable related companies, users can set a first-level weight value for related companies at the upstream, peer, or downstream nodes according to their own needs. Then, the score of each related company is calculated based on the first-level weight value, and the related companies are sorted from high to low according to the comprehensive score and output as a list. This can further filter out related companies that better meet the user's expectations.

[0023] Optionally, when a target enterprise is unable to meet the target policy due to a flexibility ratio below 80%, the allocation of the primary weight value is dynamically generated through an intelligent weight allocation algorithm, including the following steps:

[0024] Calculate the weighted components of the synergistic effect:

[0025] Based on target company A and related company B i Topological distance d(L) in the industry chain map A ,L Bi The synergistic effect value S is calculated using the exponential decay model:

[0026]

[0027] d(L A ,L Bi ) represents target companies A and B i In the topological distance of the industry chain map, k represents the attenuation factor;

[0028] Calculate the policy gain coefficient P gain :

[0029]

[0030] Where R P The amount of the policy incentive is represented by Cost(C), the integration cost is represented by Cost(C), and the PPI is the policy compliance potential index.

[0031]

[0032] η represents the excess gain coefficient. For related company B i Under elastic condition e j Quantization value, For target company A under the elastic condition e j Quantization value, As required by policy thresholds, E1 represents the set of non-compliant elastic conditions, and E2 represents the number of non-compliant elastic conditions.

[0033] First-level weight value W C =α·S+β·P gain, where α and β are secondary weight values, and α+β=1.

[0034] The synergistic effect value of the industrial chain (S); the policy compliance potential index (PPI); and the policy gain coefficient (P) are used to measure the industrial chain synergistic effect value (S); the policy compliance potential index (PPI); and the policy gain coefficient (P). gain The three factors work together to drive optimal decision-making, which solves the problem of subjectivity in weighting compared to human weighting and significantly improves accuracy.

[0035] Optionally, the extraction of policy conditions and policy rewards based on policy documents, wherein the flexible conditions are quantifiable indicators and the rigid conditions are non-quantifiable indicators, specifically including:

[0036] Policy text parsing: The structured fields in policy documents are identified using a natural language processing model, including: policy name, extracted from the document title; policy rewards, extracted from the fields "reward content", "support standards", and "policy support" to extract the amount or resource description; and policy conditions, extracted from the fields "application requirements", "applicable objects", and "policy support" to extract conditional statements.

[0037] Condition classification processing: Quantifiable indicators containing numerical ranges, percentages, or amounts are marked as flexible conditions; non-quantifiable indicators containing qualifications, regions, or ownership are marked as rigid conditions.

[0038] Structured storage: The extracted policy names, policy rewards, flexible condition lists, and rigid condition lists are stored in the policy knowledge graph database.

[0039] Optionally, the step of comparing the target company information with the policy conditions one by one specifically includes:

[0040] Conditional traversal matching: Traverse each flexible and rigid condition in the policy document, and perform the following: Extract data with the same field name from the enterprise information database;

[0041] Differentiated verification logic: Flexible condition: Call the numerical comparator to verify whether the enterprise data meets the threshold; Rigid condition: Call the text matcher to verify whether the enterprise attributes contain keywords.

[0042] Optionally, the flexible conditions that the target enterprise meets, the flexible conditions that it does not meet, the rigid conditions that it meets, and the rigid conditions that it does not meet are as follows:

[0043] Eligible flexibility criteria: Entries with enterprise data ≥ the threshold;

[0044] Unqualified flexibility conditions: Entries where enterprise data is less than the threshold;

[0045] Rigid condition for compliance: Entries whose text matches successfully;

[0046] Rigid conditions that are not met: Entries for which text matching fails.

[0047] When using this method for policy matching and recommendation, the first step is to extract policy conditions and incentives based on policy documents. These policy conditions include flexible and rigid conditions. Flexible conditions are quantifiable indicators, while rigid conditions are non-quantifiable indicators. Next, target company information is obtained, which can be acquired through user input. The information corresponding to the flexible and rigid conditions in the target company information is compared one-to-one with the flexible and rigid conditions in each policy document. This yields the number of flexible conditions the target company meets, the number of flexible conditions it does not meet, the number of rigid conditions it meets, and the number of rigid conditions it does not meet in each policy document. When the number of non-compliant flexible conditions and the number of non-compliant rigid conditions in any policy document are zero, it indicates that the target company's creditworthiness is good. If the information meets the policy conditions in the policy document, the corresponding policy document and policy reward will be pushed to the user. When the number of non-compliant rigid conditions in the policy document is zero and the number of non-compliant flexible conditions is ≥1, the ratio of the target company's information data to the quantifiable indicators in the corresponding non-compliant flexible conditions will be calculated. When the ratio is greater than 80%, the corresponding policy document and policy reward will be pushed to the user. For example, if the target company's R&D investment is 4.2 million, and the corresponding policy flexible condition is R&D investment ≥5 million, then (4.2 million / 5 million) × 100% = 84%. This indicates that the target company is close to meeting the policy conditions and can meet the relevant requirements through appropriate improvements. At this time, the corresponding policy will be pushed to the user.

[0048] As described above, this invention recommends policies to users by matching flexible and rigid conditions. When both flexible and rigid conditions are met, it indicates that the target company's information meets the policy conditions in the policy document. At this point, the corresponding policy document and policy rewards are pushed to the user, achieving accurate policy recommendations. Meanwhile, when the number of non-compliant flexible conditions is ≥1, the corresponding policy can still be recommended according to the above requirements, thereby improving the matching coverage rate and solving the industry pain point of "closely compliant companies being mistakenly screened," providing users with more comprehensive information to help them make decisions. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart of the policy matching method according to an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0052] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0053] like Figure 1 As shown, an intelligent policy matching and recommendation method includes:

[0054] S1: Based on policy documents, extract policy conditions and policy rewards. The policy conditions include flexible conditions and rigid conditions. The flexible conditions are quantifiable indicators, and the rigid conditions are non-quantifiable indicators.

[0055] Optionally, this includes (a) policy text parsing: identifying structured fields in policy documents using a natural language processing model, where: policy name, extracted from the document title; policy rewards, extracted from the fields "reward content", "support standards", and "policy support" for amount or resource description; and policy conditions, extracted from the fields "application requirements", "applicable objects", and "policy support" for conditional statements.

[0056] (b) Condition classification processing: Quantifiable indicators that include numerical range, percentage or amount are marked as flexible conditions (e.g., "R&D investment ≥ 5 million yuan"); non-quantifiable indicators that include qualifications, region, ownership, etc. are marked as rigid conditions (e.g., "registered in the high-tech zone").

[0057] (c) Structured storage: The extracted policy names, policy rewards, flexible condition lists, and rigid condition lists are stored in the policy knowledge graph database;

[0058] S2: Obtain target enterprise information, compare the information in the target enterprise information corresponding to the flexible and rigid conditions with the flexible and rigid conditions in each policy document, and obtain the flexible conditions that the target enterprise meets, the flexible conditions that it does not meet, the rigid conditions that it meets, and the rigid conditions that it does not meet in each policy document.

[0059] Optional, specifically including (a) conditional traversal matching: traversing each flexible and rigid condition in the policy document and performing: extracting data with the same name field from the enterprise information database (such as the flexible condition "R&D investment" corresponding to "R&D expenses" in the enterprise's financial statements);

[0060] (b) Differentiated verification logic: Flexible conditions: call the numerical comparator to verify whether the enterprise data meets the threshold (e.g., enterprise R&D expenses ≥ 5 million yuan); Rigid conditions: call the text matcher to verify whether the enterprise attributes contain keywords (e.g., whether the enterprise qualification list contains "high-tech enterprise").

[0061] Eligible flexibility criteria: Entries with enterprise data ≥ the threshold;

[0062] Unqualified flexibility conditions: Entries where enterprise data is less than the threshold;

[0063] Rigid condition for compliance: Entries whose text matches successfully;

[0064] Rigid conditions that are not met: Entries where text matching fails;

[0065] S3: When the number of non-compliant flexible conditions and non-compliant rigid conditions in the policy documents is zero, the corresponding policy documents and corresponding policy rewards will be pushed to the user.

[0066] S4: When the number of non-compliant rigid conditions in the policy documents is zero, and the number of non-compliant flexible conditions is ≥1, calculate the ratio of the target enterprise information data to the quantifiable indicators in the corresponding non-compliant flexible conditions. When the ratio is greater than 80%, push the corresponding policy documents and corresponding policy rewards to the user.

[0067] When using this method for policy matching and recommendation, the first step is to extract policy conditions and incentives based on policy documents. These policy conditions include flexible and rigid conditions. Flexible conditions are quantifiable indicators, while rigid conditions are non-quantifiable indicators. Next, target company information is obtained, which can be acquired through user input. The information corresponding to the flexible and rigid conditions in the target company information is compared one-to-one with the flexible and rigid conditions in each policy document. This yields the number of flexible conditions the target company meets, the number of flexible conditions it does not meet, the number of rigid conditions it meets, and the number of rigid conditions it does not meet in each policy document. When the number of non-compliant flexible conditions and the number of non-compliant rigid conditions in any policy document are zero, it indicates that the target company's creditworthiness is good. If the information meets the policy conditions in the policy document, the corresponding policy document and policy reward will be pushed to the user. When the number of non-compliant rigid conditions in the policy document is zero and the number of non-compliant flexible conditions is ≥1, the ratio of the target company's information data to the quantifiable indicators in the corresponding non-compliant flexible conditions will be calculated. When the ratio is greater than 80%, the corresponding policy document and policy reward will be pushed to the user. For example, if the target company's R&D investment is 4.2 million, and the corresponding policy flexible condition is R&D investment ≥5 million, then (4.2 million / 5 million) × 100% = 84%. This indicates that the target company is close to meeting the policy conditions and can meet the relevant requirements through appropriate improvements. At this time, the corresponding policy will be pushed to the user.

[0068] As described above, this invention recommends policies to users by matching flexible and rigid conditions. When both flexible and rigid conditions are met, it indicates that the target company's information meets the policy conditions in the policy document. At this point, the corresponding policy document and policy rewards are pushed to the user, achieving accurate policy recommendations. Meanwhile, when the number of non-compliant flexible conditions is ≥1, the corresponding policy can still be recommended according to the above requirements, thereby improving the matching coverage rate and solving the industry pain point of "closely compliant companies being mistakenly screened," providing users with more comprehensive information to help them make decisions.

[0069] In some embodiments, when the ratio is greater than 80%, the actual benefit is calculated based on the input cost and the resulting benefit of achieving the non-compliance elasticity condition, and the result is pushed to the user. The input cost can be obtained through the difference between the elasticity condition and the target enterprise information, and the resulting benefit may include policy rewards, as well as expected benefits obtained through relevant historical data. By pushing the actual benefit to the user, more intuitive data can be provided for decision-making.

[0070] In some embodiments, when the number of non-compliant flexible conditions in the policy document is zero and the number of non-compliant rigid conditions is 1, the actual benefit is calculated based on the input cost and the resulting benefit of achieving the non-compliant rigid condition, and the result is pushed to the user. For example, when the place of registration is non-compliant, the actual benefit is calculated based on the input cost and the resulting benefit of migrating the place of registration. By pushing the actual benefit to the user, more intuitive data can be provided for the user's decision-making.

[0071] When a target enterprise is unable to meet the target policy due to a flexibility ratio below 80% or more than one unmet rigid condition, relying solely on the enterprise's own situation is insufficient. To improve policy matching coverage and provide further solutions for users, in some embodiments, the method further includes:

[0072] Establish a database containing the matching results of multiple enterprises with various policy conditions, and mark the position of each enterprise in the industrial chain;

[0073] When a target company is unable to meet the target policy due to a flexibility ratio below 80% or more than one unmet rigid condition, the following steps shall be performed:

[0074] Supply chain node matching: The supply chain nodes of the target enterprise are compared with the supply chain nodes of enterprises in the database to filter out related enterprises located at upstream, same level or downstream nodes;

[0075] Data fusion verification: The data of the selected related enterprises is merged with the data of the target enterprise to generate a simulated merged dataset;

[0076] Policy compliance determination: If the simulated merged dataset meets all the conditions of the target policy, the corresponding related enterprise information in the database will be output.

[0077] In this embodiment, when a target enterprise cannot meet the target policy due to a flexibility ratio below 80% or more than one unmet rigid condition, the target enterprise's industry chain nodes are compared with those of enterprises in the database. Related enterprises located upstream, at the same level, or downstream are selected. The selected related enterprise data is then merged with the target enterprise data to generate a simulated merge dataset. Finally, if the simulated merge dataset meets all the conditions of the target policy, the corresponding related enterprise information in the database is output. Through this information, users can communicate with related enterprises to determine if a merger or acquisition is possible, thus resolving policy-related issues and other operational problems for the enterprise. This improves the coverage of policy matching and provides further solutions for users. The policy matching dimension expands from a single enterprise to the entire industry chain, opening up a new path for resource integration.

[0078] In some embodiments, when multiple eligible related companies exist in the database, the following rules are applied:

[0079] Dynamic weight setting: Allows users to set differentiated primary weight values ​​W for different nodes in the industry chain (upstream / same level / downstream). C ;

[0080] Priority Calculation: The score for each related enterprise is calculated based on the primary weight value. The calculation formula is as follows:

[0081] Score = 100 × W C ;

[0082] Sorting Output: Sort the related companies by overall score from highest to lowest and output the list of related companies.

[0083] When there are multiple qualified related companies, in order to further filter out more suitable related companies, users can set a first-level weight value for related companies at the upstream, peer, or downstream nodes according to their own needs. Then, the score of each related company is calculated based on the first-level weight value, and the related companies are sorted from high to low according to the comprehensive score and output as a list. This can further filter out related companies that better meet the user's expectations.

[0084] In some embodiments, when the target enterprise is unable to meet the target policy due to a flexibility ratio below 80%, the primary weight value W... C The allocation is dynamically generated through an intelligent weight allocation algorithm, including the following steps:

[0085] (a) Calculate the weighted components of the synergistic effect:

[0086] Based on target company A and related company B i Topological distance d(L) in the industry chain map A ,L Bi The synergistic effect value S is calculated using the exponential decay model:

[0087]

[0088] d(L A ,L Bi ) represents target companies A and B i Topological distances in the industry chain map (e.g., upstream to downstream = 3 hops, same level = 1 hop);

[0089] k represents the decay factor (default 0.5, weight decays by 40% for every additional hop);

[0090] (b) Calculate the policy gain coefficient P gain :

[0091]

[0092] Where R P The amount is the policy incentive amount, Cost(C) is the integration cost, and PPI is the Policy Potential Index (PPI).

[0093]

[0094] η: Excess gain coefficient (default 0.3, adjustable by the user); For related company B i Under elastic condition e j Quantization value, For target company A under the elastic condition e j Quantization value, As required by policy thresholds, E1 represents the set of flexible conditions that do not meet the policy sensitivity weight (default 1, configurable: core condition weight increases), and E2 represents the number of flexible conditions that do not meet the policy sensitivity weight.

[0095] First-level weight value W C =α·S+β·P gain , where α and β are secondary weight values, and α+β=1.

[0096] For example: Target company A's defect set E1 = {R&D investment};

[0097] Policy conditions: R&D requirement: ≥1 million;

[0098] Target Company A's current R&D expenditure: 600,000;

[0099] Candidate combinations:

[0100] B1: R&D cost 500,000 → After integration, 1,100,000;

[0101] B2: R&D cost 700,000 → After integration, 1,300,000;

[0102] S1=e -0.5x2 =0.37;

[0103] S2=e -0.5x1 =0.61;

[0104] When η = 0.3, In the case of PPI B1 =1.036, PPI B2 =1.108;

[0105] Let R P = 2 million yuan; B1 combination cost: 800,000 yuan; B2 combination cost: 1 million yuan; (In actual operation, relevant data can be obtained from historical data of the same type);

[0106] P gainB1 =0.555; P gainB2 =0.507;

[0107] When α = 0.4 and β = 0.6,

[0108] W CB1 =0.4802, W CB2 =0.5468; 100×W CB2 >100×W CB1

[0109] Therefore, it is more recommended to consider related companies, such as B2, for further consideration.

[0110] As can be seen from the above, this embodiment objectively quantifies the collaborative cost through the industrial chain synergy effect value S; and dynamically assesses the excess value and policy gain coefficient P through the policy compliance potential index PPI. gain By suppressing resource misallocation and driving optimal decision-making through the synergy of these three factors, the problem of subjective weighting is solved compared to human weighting judgment, significantly improving accuracy.

[0111] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention (including the claims) is limited to these examples; within the framework of the invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in the details for the sake of brevity.

[0112] This invention is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An intelligent policy matching and recommendation method, characterized in that, include: Based on policy documents, policy conditions and policy rewards are extracted. The policy conditions include flexible conditions and rigid conditions. The flexible conditions are quantifiable indicators, and the rigid conditions are non-quantifiable indicators. Obtain target enterprise information, compare the information in the target enterprise information corresponding to the flexible and rigid conditions with the flexible and rigid conditions in each policy document, and obtain the flexible conditions that the target enterprise meets, the flexible conditions that it does not meet, the rigid conditions that it meets, and the rigid conditions that it does not meet in each policy document. When the number of flexible conditions and the number of rigid conditions that are not met in the policy documents are both zero, the corresponding policy documents and corresponding policy rewards will be pushed to the user. When the number of rigid conditions that are not met in the policy documents is zero, and the number of flexible conditions that are not met is greater than or equal to 1, the ratio of the target enterprise information data to the quantifiable indicators in the corresponding flexible conditions that are not met is calculated. When the ratio is greater than 80%, the corresponding policy documents and corresponding policy rewards are pushed to the user. The method further includes: Establish a database containing matching results of multiple enterprises with various policy conditions, and assign the position of each enterprise to its corresponding node in the industrial chain; when a target enterprise cannot meet the target policy due to a flexibility ratio of less than 80% or more than one unmet rigid condition, perform the following steps: Supply chain node matching: The supply chain nodes of the target enterprise are compared with the supply chain nodes of enterprises in the database to filter out related enterprises located at upstream, same level or downstream nodes; Data fusion verification: The data of the selected related enterprises is merged with the data of the target enterprise to generate a simulated merged dataset; Policy compliance determination: If the simulated merged dataset meets all the conditions of the target policy, the corresponding related enterprise information in the database will be output. When multiple related companies that meet the requirements exist in the database, the following rules will be applied: Dynamic weight setting: Allows users to set differentiated primary weight values ​​for different nodes in the industry chain: upstream / same level / downstream. W C ; Priority Calculation: The score for each related enterprise is calculated based on the primary weight value. The calculation formula is as follows: Score = 100 × W C ; Sorted Output: Sort the related companies by overall score from highest to lowest and output the list of related companies. When a target enterprise is unable to meet the target policy due to a flexibility ratio below 80%, the allocation of the primary weight values ​​is dynamically generated through an intelligent weight allocation algorithm, including the following steps: Calculating the synergistic effect weight components: based on the target enterprise A Related companies B i Topological distance in the industry chain map d(L A ,L Bi ) The synergistic effect value S is calculated using the exponential decay model: ； d(L A ,L Bi ) Indicates the target company A and B i Topological distance in the industry chain map k Indicates the attenuation factor; Calculate the policy gain coefficient P gain : ； in R P The amount of the policy reward, Cost ( C (This refers to integration costs) PPI As a potential index for policy compliance; ； η This represents the excess gain coefficient. Related companies B i Under elastic conditions e j Quantization value, For target companies A Under elastic conditions e j Quantization value, As required by policy thresholds, As a policy-sensitive weight, E 1 represents the set of elastic conditions that are not met. E 2 indicates the number of non-compliant elastic conditions; first-level weight value. ,in α , β These are secondary weight values. α + β =1.

2. The intelligent policy matching and recommendation method according to claim 1, characterized in that, When the ratio is greater than 80%, the actual benefit is calculated based on the input cost and the resulting benefit of achieving the non-compliance elasticity condition, and the result is pushed to the user.

3. The intelligent policy matching and recommendation method according to claim 1, characterized in that, When the number of non-compliant flexible conditions in policy documents is zero, and the number of non-compliant rigid conditions is 1, the actual benefit is calculated based on the input cost and the resulting benefit of achieving the non-compliant rigid conditions, and the result is pushed to the user.

4. The intelligent policy matching and recommendation method according to claim 1, characterized in that, The extraction of policy conditions and policy incentives based on policy documents, wherein the flexible conditions are quantifiable indicators and the rigid conditions are non-quantifiable indicators, specifically includes: Policy text parsing: The structured fields in policy documents are identified using a natural language processing model, including: policy name, extracted from the document title; policy rewards, extracted from the "reward content", "support standards", and "policy support" fields to extract the amount or resource description; and policy conditions, extracted from the "application requirements", "applicable objects", and "policy support" fields to extract conditional statements. Condition classification processing: Quantifiable indicators containing numerical ranges, percentages, or amounts are marked as flexible conditions; non-quantifiable indicators containing qualifications, regions, or ownership are marked as rigid conditions. Structured storage: The extracted policy names, policy rewards, flexible condition lists, and rigid condition lists are stored in the policy knowledge graph database.

5. The intelligent policy matching and recommendation method according to claim 1, characterized in that, The process of comparing target enterprise information with policy conditions one by one includes: condition traversal matching: traversing each flexible and rigid condition in the policy document; execution: extracting data with the same field name from the enterprise information database; and differential verification logic: flexible conditions: calling a numerical comparator to verify whether the enterprise data meets the threshold; rigid conditions: calling a text matcher to verify whether the enterprise attributes contain keywords.

6. The intelligent policy matching and recommendation method according to claim 1, characterized in that, The target enterprise meets the following flexible conditions, does not meet the following flexible conditions, meets the following rigid conditions, and does not meet the following rigid conditions: Meeting the flexible conditions: Entries with enterprise data ≥ the threshold; Not meeting the flexible conditions: Entries with enterprise data < the threshold; Meeting the rigid conditions: Entries with successful text matching; Not meeting the rigid conditions: Entries with unsuccessful text matching.

Images