A financing scheme generation method and system based on supply chain data

By acquiring multi-source data and utilizing Bayesian inference and graph neural network models, key node enterprises in the supply chain are identified, solving the problem of inaccurate risk assessment in existing technologies, generating personalized financing solutions, and improving the efficiency and accuracy of supply chain financial services.

CN118656730BActive Publication Date: 2024-11-22JIANGSU WAFA INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410794924.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-19
Publication Date
2024-11-22
Estimated Expiration
2044-06-19

AI Technical Summary

Technical Problem

Existing financing schemes rely primarily on the credit assessment of individual companies, neglecting the complex relationships and dynamic changes within the supply chain. This results in inaccurate risk assessments and outdated data processing, making it difficult to fully explore and utilize the potential value of multi-source heterogeneous data.

Method used

By acquiring basic enterprise information, supply chain transaction data, logistics information, and financial institution data, and using a multi-source data fusion mathematical model based on Bayesian inference to integrate the data, combined with social network analysis and graph neural network models, key node enterprises are identified, credit assessments and risk predictions are conducted, and personalized financing solutions are generated.

Benefits of technology

It accurately reflects the complex relationships between supply chain companies, generates more precise financing solutions, and improves the accuracy of risk assessment and the rationality of financing solutions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118656730B_ABST
    Figure CN118656730B_ABST
Patent Text Reader

Abstract

The present invention provides a method and system for generating a financing plan based on supply chain data, which relates to the technical field of financial services, including obtaining first information and second information; evaluating and processing the first information and the second information to obtain a supply chain influence scoring result; performing data fusion processing on the second information to obtain a supply chain fusion data set; performing identification processing on the supply chain fusion data set to obtain a supply chain network model; performing data integration processing on enterprise basic information, supply chain influence scoring results and supply chain network models based on a graph neural network mathematical model to obtain a comprehensive data view; performing time series analysis processing based on the comprehensive data view to obtain an evaluation result; optimizing the evaluation result based on a deep learning mathematical model to obtain a personalized financing plan. The present invention is based on a multi-source data fusion mathematical model based on Bayesian reasoning, which can effectively process multi-source heterogeneous data in the supply chain and achieve efficient data fusion.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of financial services, and in particular to a method and system for generating a financing plan based on supply chain data. Background Art

[0002] With the continuous development of the global economy and the increasing complexity of supply chain management, the importance of financial services for supply chain-related enterprises in the modern economy has become increasingly prominent. Supply chain financial services aim to provide personalized financing solutions for enterprises in the supply chain through close cooperation between financial institutions and enterprises in various links of the supply chain, thereby improving capital liquidity and overall operational efficiency. However, current supply chain financial services face many challenges and problems in practical applications. First, the existing financing plan formulation model mainly relies on the credit assessment of a single enterprise. This method only analyzes the financial data and credit records provided by the enterprise, ignoring the complex relationships and dynamic changes of enterprises in the supply chain, making the risk assessment results inaccurate. Secondly, the existing technology is relatively backward in data processing. It usually uses simple statistical and linear regression models for analysis, lacks effective fusion and in-depth analysis of multi-source heterogeneous data, and is difficult to fully explore and utilize the potential value of big data, resulting in deviations in evaluation results and unreasonable financing plans.

[0003] Based on the above shortcomings of the prior art, there is an urgent need for a method and system for generating a financing plan based on supply chain data. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for generating a financing plan based on supply chain data to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present application provides a method for generating a financing solution based on supply chain data, comprising:

[0006] Acquire first information and second information, wherein the first information includes basic information of the enterprise applying for financial services, and the second information includes basic information of enterprises in the supply chain, supply chain transaction data, logistics information, and financial institution data;

[0007] Performing evaluation processing based on the first information and the second information, and obtaining a supply chain influence score result of the enterprise by evaluating the importance and substitutability of the enterprise in the supply chain;

[0008] Performing data fusion processing on the second information based on a multi-source data fusion mathematical model of Bayesian reasoning to obtain a supply chain fusion data set;

[0009] Performing identification processing based on the supply chain fusion data set, identifying key node enterprises in the supply chain through social network analysis and evaluating the centrality and connectivity of the key node enterprises in the entire supply chain to obtain a supply chain network model;

[0010] Based on a preset graph neural network mathematical model, data integration and association mining are performed on the basic information of the enterprise, the supply chain influence scoring results and the supply chain network model to obtain a comprehensive data view, which includes the network position and relationship strength of the enterprise applying for financial services in the supply chain;

[0011] Performing time series analysis and processing according to the comprehensive data view to obtain an evaluation result, wherein the evaluation result includes the credit evaluation result, risk prediction result and systemic risk score of the enterprise in the supply chain;

[0012] The evaluation results are subjected to multi-objective optimization processing based on a preset deep learning mathematical model to generate financing plans under at least two different scenarios, and the financing plans are selected and adjusted through a hybrid intelligent algorithm to obtain a final personalized financing plan.

[0013] In a second aspect, the present application also provides a financing solution generation system based on supply chain data, including:

[0014] An acquisition module, used to acquire first information and second information, wherein the first information includes basic information of enterprises applying for financial services, and the second information includes basic information of enterprises in the supply chain, supply chain transaction data, logistics information, and financial institution data;

[0015] An evaluation module, configured to perform evaluation processing based on the first information and the second information, and obtain a supply chain influence score of the enterprise by evaluating the importance and substitutability of the enterprise in the supply chain;

[0016] A fusion module, which performs data fusion processing on the second information based on a multi-source data fusion mathematical model of Bayesian reasoning to obtain a supply chain fusion data set;

[0017] An identification module is used to perform identification processing based on the supply chain fusion data set, identify key node enterprises in the supply chain through social network analysis, and evaluate the centrality and connectivity of the key node enterprises in the entire supply chain to obtain a supply chain network model;

[0018] An integration module performs data integration and association mining on the basic information of the enterprise, the supply chain influence scoring results and the supply chain network model based on a preset graph neural network mathematical model to obtain a comprehensive data view, wherein the comprehensive data view includes the network position and relationship strength of the enterprise applying for financial services in the supply chain;

[0019] An analysis module, configured to perform time series analysis processing according to the comprehensive data view to obtain an evaluation result, wherein the evaluation result includes a credit evaluation result, a risk prediction result, and a systemic risk score in the supply chain of the enterprise;

[0020] The optimization module performs multi-objective optimization processing on the evaluation results based on a preset deep learning mathematical model, generates financing plans under at least two different scenarios, and selects and adjusts the financing plans through a hybrid intelligent algorithm to obtain a final personalized financing plan.

[0021] The beneficial effects of the present invention are:

[0022] The present invention is based on the Bayesian reasoning of the multi-source data fusion mathematical model, which can effectively process the multi-source heterogeneous data in the supply chain and realize the efficient fusion of data. Through clustering processing, conditional generalization calculation and maximum a posteriori estimation, it accurately reflects the complex relationship between supply chain enterprises and obtains a more accurate supply chain fusion data set.

[0023] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 It is a schematic diagram of the flow of the method for generating a financing plan based on supply chain data described in an embodiment of the present invention;

[0026] Figure 2 Flowchart of the genetic algorithm described in the embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0028] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0029] Embodiment 1:

[0030] This embodiment provides a method for generating a financing plan based on supply chain data.

[0031] See also Figure 1 , the figure shows that the method includes step S100, step S200, step S300, step S400, step S500, step S600 and step S700.

[0032] Step S100, obtaining first information and second information, wherein the first information includes basic information of the enterprise applying for financial services, and the second information includes basic information of the enterprise in the supply chain, supply chain transaction data, logistics information and financial institution data;

[0033] It is understandable that the basic information of enterprises mainly includes the basic information of enterprises applying for financial services, such as enterprise name, registered capital, legal representative, main business, historical operating data, financial statements (balance sheet, income statement, cash flow statement), credit record, shareholder structure, etc. To obtain this data, enterprises usually need to submit it actively and verify and supplement it through the internal information system of the enterprise or a third-party data provider. The basic information of enterprises in the supply chain includes the basic information of upstream suppliers and downstream customers related to the applicant enterprise, such as enterprise name, contact person, business cooperation history, etc. Supply chain transaction data involves transaction records between enterprises and their upstream and downstream partners in the supply chain, including purchase orders, sales orders, invoice information, payment records, etc. These data reflect the transaction behavior and credit status of enterprises in the supply chain. Logistics information includes data such as the transportation route, transportation method, warehousing information, and logistics suppliers of goods. This information is used to analyze the liquidity and logistics efficiency of the supply chain. Financial institution data includes credit records, loan interest rates, credit approval standards, repayment records, etc. of financial institutions for enterprises in the supply chain. These data can be obtained from banks, financial service platforms, etc., reflecting the financing history and credit status of enterprises. During the data acquisition process, strictly abide by data privacy protection laws and regulations to ensure the security of sensitive corporate information and prevent data leakage and abuse.

[0034] Step S200: performing evaluation processing based on the first information and the second information, and obtaining a supply chain influence score result of the enterprise by evaluating the importance and substitutability of the enterprise in the supply chain;

[0035] It is understandable that by analyzing factors such as the transaction frequency, transaction amount, transaction stability, and logistics routes of enterprises in the supply chain, the position and importance of enterprises in the supply chain can be comprehensively evaluated; by analyzing information such as the transaction relationship, supply chain network structure, and logistics routes of enterprises, the degree of substitutability of enterprises can be evaluated. The supply chain influence score result is a comprehensive indicator that reflects the position and influence of enterprises in the supply chain. By evaluating the supply chain influence of enterprises, key nodes and cooperation opportunities in the supply chain can be discovered, cooperation and optimization of the supply chain can be promoted, and overall efficiency and competitiveness can be improved.

[0036] Step S300, performing data fusion processing on the second information based on a multi-source data fusion mathematical model of Bayesian reasoning to obtain a supply chain fusion data set;

[0037] It is understandable that the data sources involved in the supply chain usually come from different sources and different formats, and there may be inconsistent, incomplete and inaccurate data. Therefore, it is necessary to use data fusion technology to integrate and unify these data to eliminate conflicts and contradictions between data, so as to obtain a more reliable data set. Bayesian reasoning is a probabilistic inference method based on Bayes' theorem. By continuously updating the prior probability based on new evidence, the posterior probability is obtained. In data fusion, Bayesian reasoning is used to deal with the uncertainty and inconsistency between different data sources, so as to better fuse the data. In this step, a comprehensive supply chain fusion data set can be obtained by performing Bayesian reasoning and data fusion processing on the basic information of the enterprise, supply chain transaction data, logistics information and financial institution data. This data set will contain more comprehensive and accurate supply chain information, providing more reliable data support for subsequent analysis and decision-making.

[0038] Step S400: Perform identification processing based on the supply chain fusion data set, identify key node enterprises in the supply chain through social network analysis, and evaluate the centrality and connectivity of the key node enterprises in the entire supply chain to obtain a supply chain network model;

[0039] Furthermore, the constructed supply chain network model will more comprehensively and accurately reflect the relationship between various enterprises in the supply chain, which will help financial institutions to fully understand the structure and operation of the supply chain. Centrality is an indicator to measure the importance of a node in the network. Common centrality indicators include degree centrality, betweenness centrality and closeness centrality. By evaluating the centrality of key node enterprises, we can understand their influence and position in the supply chain, so as to better understand and manage the supply chain network. Based on the identification of key node enterprises and the results of centrality evaluation, a supply chain network model can be constructed. This model will show the relationship between various enterprises in the supply chain, helping financial institutions to fully understand the supply chain structure and the interactive relationship between enterprises.

[0040] Step S500: Based on a preset graph neural network mathematical model, data integration and association mining are performed on the basic information of the enterprise, the supply chain influence scoring results and the supply chain network model to obtain a comprehensive data view, which includes the network position and relationship strength of the enterprise applying for financial services in the supply chain;

[0041] Among them, the graph neural network model can make full use of the structural information and node characteristics of graph data to achieve effective representation and association mining of the graph structure, and can mine the potential connections and patterns in the data, which helps to discover potential cooperative relationships, influencing factors and possible risk points between enterprises, and provide financial institutions with deeper data insights.

[0042] Step S600: Perform time series analysis based on the comprehensive data view to obtain an evaluation result, which includes the enterprise's credit evaluation result, risk prediction result, and systemic risk score in the supply chain;

[0043] It is understandable that through time series analysis, we will analyze and compare the data of enterprises at different time points to reveal the evolution trend of their credit and risk, and make predictions and assessments for the future. Considering the complexity and interdependence of the supply chain, the risk assessment of enterprises is incorporated into the perspective of the entire supply chain, helping financial institutions to better understand and evaluate the stability and risk distribution of the entire supply chain network.

[0044] Step S700: Perform multi-objective optimization processing on the evaluation results based on a preset deep learning mathematical model to generate financing plans for at least two different scenarios, and select and adjust the financing plans through a hybrid intelligent algorithm to obtain a final personalized financing plan.

[0045] First, a deep learning mathematical model is used to perform multi-objective optimization for different objectives (such as financing costs, risk minimization, liquidity maximization, etc.) to ensure that the generated financing plan has advantages in many aspects. Based on the results of the multi-objective optimization process, at least two financing plans for different scenarios will be generated. These plans include different interest rates, terms, repayment methods, etc. to meet the needs and situations of different companies. The generated financing plans are selected and adjusted through hybrid intelligent algorithms (such as genetic algorithms, simulated annealing algorithms, etc.). These algorithms can find the optimal personalized financing plan while ensuring the diversity of the plan.

[0046] It should be noted that step S200 includes step S210, step S220, step S230 and step S240.

[0047] Step S210: perform importance evaluation based on enterprise basic information and supply chain transaction data, and obtain the importance score of the enterprise applying for financial services in the supply chain by constructing a transaction network diagram and using the PageRank algorithm to calculate the PageRank value of each enterprise;

[0048] First, a transaction network diagram is constructed based on the supply chain transaction data. In this diagram, each node represents an enterprise, and each edge represents the transaction relationship between two enterprises, which can intuitively show the transaction relationship between enterprises in the supply chain. Next, the PageRank algorithm is used to calculate the constructed transaction network diagram and apply it to the enterprise nodes in the supply chain to determine the importance of each enterprise in the supply chain. The PageRank algorithm can comprehensively consider the direct transaction relationship of the enterprise and the connection with other important enterprises, so it can more comprehensively evaluate the importance of the enterprise in the supply chain. Finally, through the calculation results of the PageRank algorithm, the PageRank value of each enterprise can be obtained, that is, its importance score in the supply chain. This score reflects the position and influence of the enterprise in the supply chain and can serve as an important basis for subsequent evaluation.

[0049] Step S220: Perform redundant network analysis based on the basic information of the enterprise and the logistics information, calculate the redundancy of each node enterprise in the supply chain and evaluate the position and influence of the enterprise applying for financial services in the logistics path to obtain a substitutability score;

[0050] Specifically, it is necessary to first collect comprehensive basic information on enterprises, including information such as company profile, operating history and key performance indicators. These data can provide insights into the structure and characteristics of each enterprise in the supply chain. Next, collect logistics information, which includes data related to the movement of goods and materials in the supply chain. This includes detailed information on transportation routes, delivery times, inventory levels, and potential bottlenecks or disruptions in the logistics network. The redundancy of each node enterprise is then assessed, which involves analyzing factors such as the number of alternative routes available, the degree of overlap in functions and capabilities between different enterprises, and the resilience of the supply chain network to disruptions. Based on this analysis, a redundancy score is assigned to each enterprise, indicating its degree of redundancy in the supply chain. In addition, the position and influence of the applicant enterprise in the logistics path must be assessed to determine its substitutability or other options when disruptions or changes occur in the supply chain.

[0051] Step S230, performing weighted comprehensive processing according to the importance score and the substitutability score to obtain a preliminary score result;

[0052] In this step, the importance score and the substitutability score are first weighted. This weighting may be based on the importance of different factors, such as the strategic position, market share or business relevance of each node enterprise in the supply chain. These weights can be adjusted and determined based on actual conditions and domain expertise. Next, the importance score and substitutability score of each enterprise are weighted and summed to obtain its overall score in the supply chain. The weighted summation method can be a simple weighted average or calculated according to a specific weighting function to better reflect the relationship between different factors. Finally, through this weighted comprehensive processing, the preliminary scoring results of each enterprise can be obtained. These preliminary scoring results will provide important references for subsequent analysis and decision-making, help better understand the position and influence of the applicant enterprise in the supply chain, and thus provide them with more accurate and personalized financial services.

[0053] Step S240: Adjust the preliminary scoring results and the historical credit records of the enterprise in the financial institution data, and calibrate the preliminary scoring results using a weighted linear regression model to obtain the final supply chain influence scoring results.

[0054] Specifically, we first use the historical credit records of enterprises in the data of financial institutions, including information such as the credit rating, loan records, and repayment records of enterprises. Then, we establish a weighted linear regression model, which takes the preliminary scoring results as independent variables and the historical credit records of enterprises in the data of financial institutions as dependent variables, and fits the relationship between the two through regression analysis. When establishing the model, we consider the weights of different credit records and possible nonlinear relationships. Then, we use the established weighted linear regression model to calibrate the preliminary scoring results to obtain more accurate and reliable supply chain influence scoring results. This calibration process can further improve the accuracy and objectivity of the scoring based on the preliminary scoring results and the historical credit records of enterprises in the data of financial institutions. The supply chain influence scoring results obtained through adjustment will provide important references for financial institutions, helping them to better understand the position and credit status of enterprises in the supply chain, so as to formulate financial service strategies more accurately, reduce risks, and improve efficiency.

[0055] It should be noted that step S300 includes step S310, step S320, step S330 and step S340.

[0056] Step S310: clustering the supply chain transaction data and logistics information, using a Gaussian mixture model to process the transaction volume, transaction frequency and logistics node information in the supply chain and fitting the multidimensional normal distribution of the data to obtain an initial multidimensional data distribution;

[0057] It is understandable that supply chain transaction data and logistics information include transaction volume and frequency between different enterprises, as well as the location and characteristics of logistics nodes. These data are usually multidimensional and contain information from different aspects. Cluster analysis is then used to process these data to identify potential patterns and structures. In order to better describe the distribution of the data, the Gaussian mixture model is used to fit the multidimensional normal distribution of the data, and a set of Gaussian distributions are used to approximate the distribution of the data, so as to better understand the characteristics and structure of the data, and provide an important basis for subsequent data fusion and processing.

[0058] Step S320: Perform conditional summary calculation based on the initial multidimensional data distribution and financial institution data, use a naive Bayes classifier to classify loan interest rates, credit approval standards, and repayment records, and calculate the posterior probability of each category to obtain a posterior probability set of all categories;

[0059] Specifically, financial institution data includes information such as loan interest rates, credit approval standards, and repayment records, which will be used to help understand and evaluate the loan risk and credit status of the enterprise. Next, the conditional summary calculation method will be used, which is one of the key steps of the naive Bayes classifier. Through conditional summary calculation, the conditional probability distribution of other variables under given conditions will be calculated based on known categories (such as loan interest rates, credit approval standards, and repayment records, etc.). Then, the naive Bayes classifier will be applied, which is a probabilistic classification algorithm used to deal with classification problems. In this embodiment, the naive Bayes classifier is used to divide the enterprise into different categories, such as "high risk", "medium risk" and "low risk", using factors such as the loan interest rate, credit approval standards, and repayment records of the enterprise as features. Finally, the posterior probability of each category will be calculated, which is to estimate the probability of different categories under the condition of given observation data. By calculating the posterior probability set, the probability distribution of the enterprise belonging to different risk categories can be obtained, so as to better evaluate the credit status and loan risk of the enterprise. This step classifies factors such as the company's loan interest rate, credit approval standards, and repayment record through conditional summary calculations and naive Bayes classifiers, and calculates the posterior probability of each category, thereby more accurately assessing the company's credit status and loan risk.

[0060] Step S330: Perform maximum a posteriori estimation processing according to the posterior probability set, and construct a causal relationship graph using a Bayesian network to obtain a fusion causal network that reflects the correlation between supply chain enterprises;

[0061] Specifically, the maximum a posteriori estimation processing method is a method for estimating parameters by considering prior knowledge and observed data. In this step, the posterior probability set is used as the observed data, and combined with the possible prior knowledge, the correlation between supply chain enterprises is estimated by maximum a posteriori estimation. Next, the Bayesian network will be used to construct a causal relationship graph. The Bayesian network is a probabilistic graphical model used to represent the dependency relationship between variables. In this embodiment, a causal relationship graph between supply chain enterprises will be established through a Bayesian network, in which nodes represent enterprises and edges represent relationships between enterprises, such as cooperative relationships, transaction relationships, etc. By constructing a Bayesian network, the correlation and mutual influence relationship between supply chain enterprises can be better understood and described. This fusion causal network can help identify key node enterprises in the supply chain, and thus better evaluate the stability, reliability and risk of the supply chain.

[0062] Step S340: sampling and inference processing are performed on the fusion causal network based on a preset Markov chain Monte Carlo mathematical model to obtain a supply chain fusion data set.

[0063] It can be understood that Markov Chain Monte Carlo (MCMC) is a statistical simulation method for random sampling from probability distribution. In this step, the Markov Chain Monte Carlo method is used to extract samples from the joint probability distribution of the fused causal network to obtain the correlation and interaction information between supply chain enterprises. Next, the sample data obtained by sampling is inferred. By performing statistical analysis and inference on the sampled samples, the probability distribution of various indicators between supply chain enterprises can be obtained, such as the strength of the relationship between enterprises, the degree of influence, etc. These inference results will help to gain a deeper understanding of the characteristics and structure of the supply chain. The final generated supply chain fusion data set contains the correlation and interaction information between supply chain enterprises, providing an important data basis for subsequent analysis and application.

[0064] It should be noted that step S400 includes step S410, step S420, step S430 and step S440.

[0065] Step S410, performing K-core decomposition processing according to the supply chain fusion data set, and obtaining a preliminary key node set by removing low-degree nodes;

[0066] It can be understood that the K-core decomposition algorithm is a graph theory algorithm used to identify key nodes in a graph. The algorithm gradually removes non-key nodes in the graph by continuously removing nodes with degrees lower than K and their related edges, and finally obtains a set of key nodes with high degrees and closely connected. In this step, the supply chain fusion data set is subjected to K-core decomposition according to the set K value. By continuously iteratively removing nodes with degrees lower than K, those nodes that have little impact on the overall structure of the supply chain are gradually eliminated, thereby screening out a set of preliminary key nodes. After the K-core decomposition process, a preliminary set of key nodes of the supply chain is obtained. These nodes have high degrees and close connectivity, play an important role in the entire supply chain network, and have an important impact on the stability and efficiency of the supply chain. These nodes serve as the basis for subsequent social network analysis and help identify key node enterprises in the supply chain.

[0067] Step S420, calculating the number of times each node serves as a supply chain intermediary node according to the preliminary key node set to obtain a betweenness centrality score;

[0068] It should be noted that betweenness centrality is a commonly used indicator in social network analysis to measure the degree of intermediary a node plays in a network. In a supply chain network, betweenness centrality can reflect the importance of a node in the path connecting different nodes. If a node appears in the shortest path between many pairs of nodes, then its betweenness centrality will be high. The calculation formula is:

[0069]

[0070] Among them, C B (v) is the betweenness centrality score of node v, s, t and v are all node names, σ st represents the total number of shortest paths from node s to node t, σ st (v) represents the number of paths that pass through node v among all the shortest paths from node s to node t. When calculating betweenness centrality, we will traverse all pairs of nodes s and t and calculate the number of shortest paths that pass through node v. Then, divide the number of shortest paths that pass through node v by the total number of shortest paths to obtain the betweenness centrality score of node v. This step can accurately evaluate the degree of intermediary of each node in the supply chain network by calculating the betweenness centrality. This helps to identify nodes that play a key intermediary role in the supply chain and provides an important basis for the subsequent construction and analysis of the supply chain network model.

[0071] Step S430, calculating the degree and distance of each node according to the supply chain fusion data set and the betweenness centrality score, and obtaining the degree centrality score and closeness centrality score of the node in the supply chain;

[0072] When calculating degree centrality and closeness centrality, it is necessary to consider the degree of direct connection of the node in the network and the distance from the node to other nodes. These indicators can help understand the importance and position of the node in the supply chain. Degree centrality reflects the direct connection of the node, while closeness centrality considers the distance relationship between the node and other nodes. By calculating degree centrality and closeness centrality, the importance and position of each node in the supply chain network can be more comprehensively evaluated. Among them, the degree centrality score calculation formula is:

[0073]

[0074] The formula for calculating the closeness centrality score is:

[0075]

[0076] Among them, v and u are node names, C D (v) represents the degree centrality score of node v, deg(v) represents the degree of node v, that is, the number of directly connected nodes, C C (v) The closeness centrality score of node v, N represents the total number of nodes in the network, d(v,u) represents the shortest path distance from node v to node u, and V represents the set of nodes in the network.

[0077] Step S440: Calculate the weighted clustering coefficient based on the betweenness centrality score, degree centrality score and closeness centrality score, obtain the final centrality and connectivity of key node enterprises by calculating the local network density of the nodes, and integrate the supply chain fusion data set to obtain the supply chain network model.

[0078] It is understandable that the betweenness centrality score can help determine which nodes play a key role in information dissemination and interaction, the degree centrality score reflects the direct connection of the node, and the proximity centrality score considers the distance between the node and other nodes. By comprehensively considering these indicators, key node enterprises in the supply chain can be more accurately identified. These enterprises play a core role in the entire supply chain network and play an important supporting role in the stability and operation of the entire supply chain. Compared with simple degree centrality or betweenness centrality evaluation, this weighted clustering coefficient calculation method can better reflect the actual situation of the supply chain network, thereby providing more reliable data support for subsequent risk assessment and financing plan formulation. In addition, combined with the supply chain fusion data set for integration, the supply chain network model can be closer to the actual situation, thereby improving the accuracy and credibility of the evaluation results.

[0079] It should be noted that step S500 includes step S510, step S520, step S530 and step S540.

[0080] Step S510: Based on the preset graph neural network mathematical model, the basic information of the enterprise, the supply chain influence scoring results and the supply chain network model are integrated and processed for association mining to obtain the association information representation;

[0081] The key to this step is to use the ability of graph neural networks to integrate different types of data and mine the associations between them. In this way, a more comprehensive understanding of the position, influence, and relationship between enterprises in the supply chain and other enterprises can be achieved. First, graph neural networks can process data at the node level and the graph level. In this embodiment, it is necessary to integrate and mine the basic information of the enterprise, the supply chain influence scoring results, and the supply chain network model, which means that the characteristics of individual enterprises and their position and influence in the entire supply chain network need to be considered at the same time. Graph neural networks can effectively capture the complex associations between these data and convert them into feature representations that can be used for subsequent processing. Secondly, graph neural networks can process different types of graph structure data, including directed graphs, undirected graphs, and graphs with node and edge attributes. Also in this embodiment, a complex supply chain network needs to be processed, which includes both the relationship between enterprises and their attribute information, such as enterprise basic information and supply chain transaction data. The flexibility of graph neural networks makes it possible to integrate these different types of data together to build a comprehensive graph model, so as to better understand the relationship and characteristics between each node enterprise in the supply chain. Finally, graph neural networks have the ability to learn and reason, and can learn complex patterns and rules from data, and use them for subsequent predictions and decisions. Graph neural networks can learn the relationship between enterprises at each node in the supply chain from the basic information of the enterprise, the supply chain influence score results, and the supply chain network model, and then predict the credit assessment results, risk prediction results, and systemic risk scores of the enterprise. This end-to-end learning method can better mine information from the data and provide more accurate and reliable decision support for supply chain financial services.

[0082] Step S520: Perform community division processing according to the association information representation, and obtain community division results by dividing the enterprises that are closely related to each other in the graph into the same community. The community division results include the network position of each enterprise in the supply chain;

[0083] It is understandable that there are complex relationships between enterprises in the supply chain, and community division can help group these enterprises according to their similarities and connections in the supply chain network. Such division helps to better understand the structure and organization of the supply chain, identify a collection of enterprises with similar functions or status, and thus provide a clearer perspective for subsequent analysis and decision-making. For example, the results of community division can show the core enterprise groups in the supply chain and the relationships between them, and can also identify some specific types of enterprise collections, such as manufacturers, distributors, etc. This information is very important for formulating targeted financing plans and risk management strategies. In addition, the results of community division can also provide network position information of each enterprise in the supply chain. Through community division, we can understand the community to which each enterprise belongs and its position in the community, so as to better understand its position and role in the entire supply chain. This information is of great significance for evaluating the supply chain influence, risk exposure and relationship between enterprises and other enterprises, and can provide more comprehensive and accurate data support for subsequent risk assessment and financing plan formulation.

[0084] Step S530: Calculate the relationship strength based on the association information representation and the community division result, and obtain the relationship strength between enterprises by calculating the similarity in the association information;

[0085] In this step, the strength of the relationship between enterprises is evaluated by using the attribute information, transaction data and other related information between enterprises through appropriate similarity calculation methods. For example, methods such as cosine similarity or Euclidean distance can be used to compare the attribute characteristics between enterprises to evaluate the similarity between them; or the closeness of the transaction relationship between enterprises can be evaluated based on the frequency and amount of transaction data. Through this calculation of relationship strength, the relationship between enterprises, including cooperative relationships, competitive relationships, etc., can be understood more accurately. This is very important for subsequent risk assessment and financing plan formulation, and can help better identify other enterprises closely related to the enterprise, so as to better grasp the risks and opportunities in the supply chain. In addition, by considering the results of community division, the strength of the relationship between enterprises in different communities can also be evaluated. This helps to have a deeper understanding of the connection and influence between different communities, and provides more comprehensive and accurate data support for the targeted formulation of financing plans and risk management strategies.

[0086] Step S540: Perform data visualization processing according to the network position and relationship strength to obtain a comprehensive data view.

[0087] The purpose of this step is to present complex data information in an intuitive and easy-to-understand way, providing users with a comprehensive and clear picture of the supply chain network. Specifically, the network position information of enterprises in the supply chain network is used to determine their node positions in the graph. This can be achieved by drawing nodes in two-dimensional or three-dimensional space and grouping or coloring them according to their different network positions. For example, core enterprises and secondary enterprises can be marked with different node colors or shapes to highlight their positions in the supply chain. Such visualization can intuitively show the position and status of enterprises in the supply chain and help users quickly understand the overall structure of the supply chain. Secondly, the strength of the relationship between enterprises calculated previously is used to determine the weight of the connection between nodes or the thickness of the edge. The strength of the relationship between enterprises is reflected by adjusting the thickness or color of the connection, so that the connection between nodes with strong relationship strength is more prominent, while the connection between nodes with weak relationship strength is thinner and weaker. Such visualization can intuitively show the closeness of the relationship between enterprises and help users quickly understand the connection and influence between enterprises in the supply chain. Through the visualization of this comprehensive data view, users can understand the structure of the supply chain network, the relationship between enterprises, and their position and influence in the entire supply chain at a glance. This helps users better understand the operation mechanism of the supply chain, identify potential risks and opportunities, and provide more accurate and comprehensive data support for subsequent decision-making.

[0088] It should be noted that step S600 includes step S610, step S620, step S630 and step S640.

[0089] Step S610: Modeling the time series data in the comprehensive data view based on a preset long short-term memory network mathematical model, and obtaining the dynamic credit score time series of the enterprise by capturing the long-term and short-term dependencies in the enterprise credit score data;

[0090] It can be understood that the LSTM network is a recursive neural network that is particularly suitable for processing and predicting time series data and can effectively capture long-term dependencies in time series. In this step, the LSTM network is used to model the credit score data of the enterprise to identify long-term trends and short-term fluctuations in the credit score time series. The LSTM network modeling can take into account the time dependency in the time series data and the correlation between the sequences, so as to more accurately predict the future credit score trend. The LSTM network can automatically learn the patterns and laws in the time series data and predict future development trends based on historical data, so it can better capture the dynamic changes of the enterprise credit score.

[0091] Step S620: perform hidden state estimation on the dynamic credit score time series, and predict the future credit trend of the enterprise by processing the randomness and trend in the time series through the state space model to obtain the future credit evaluation result of the enterprise;

[0092] It should be noted that hidden state estimation is a technique for inferring hidden states based on observation sequences, and is usually used in conjunction with state space models. The state space model is a probabilistic model that describes the evolution of a system. It divides the state of the system into observed states and hidden states. The hidden state is usually not directly observable, but is indirectly inferred from the observed state. In this step, the dynamic credit score time series of the enterprise is used as the observation sequence, and the hidden state, that is, the state that reflects the potential factors and laws behind the credit score, is inferred through the hidden state estimation technique. Then, the state space model is used to describe the evolution law between these hidden states, while taking into account the randomness and trend in the time series. Through the state space model, the dynamic change mechanism behind the credit score can be better understood, so as to more accurately predict the future credit change trend of the enterprise. Finally, by predicting the future credit change trend of the enterprise, the future credit assessment results of the enterprise can be obtained. This assessment result can provide important reference information for financial institutions and enterprises, helping them to better formulate risk management strategies and financing decisions. By timely predicting the credit change trend of enterprises, financial institutions can be helped to more effectively identify and manage potential credit risks, while also providing enterprises with better financing opportunities and conditions.

[0093] Step S630: Based on the comprehensive data view and the credit assessment result, a Bayesian dynamic linear model is used to perform Bayesian inference processing to obtain a risk prediction result, where the risk prediction result is the risk level of the enterprise at a specific point in time in the future;

[0094] It can be understood that the Bayesian dynamic linear model is a dynamic model based on Bayesian statistical theory, which can continuously update the estimation of unknown parameters by using observed data while considering prior knowledge. In this step, the comprehensive data view and historical credit assessment results of the enterprise are used as observation data, and the Bayesian dynamic linear model is used to infer the future risk level of the enterprise. Specifically, the comprehensive data view and credit assessment results of the enterprise are used as the input of the model, and the Bayesian inference method is used to predict the future risk level according to the parameters of the model, combined with historical data and current credit assessment results. Such risk prediction results can provide important reference information for financial institutions and enterprises, helping them to better formulate risk management strategies and financing decisions. Through the inference processing of the Bayesian dynamic linear model, the influence of uncertainty factors and historical data can be taken into account, and the risk level of the enterprise at a specific point in time in the future can be more accurately predicted. This helps financial institutions and enterprises to timely discover and respond to potential risks, and ensure the stable operation of the financial system and the sustainable development of enterprises.

[0095] Step S640: Based on the comprehensive data view and risk prediction results, the connectivity and contagion of the supply chain network are analyzed using a systemic risk model, and an evaluation systemic risk score is obtained by evaluating the impact of risks on the entire network when the enterprise encounters risks in the supply chain.

[0096] It can be understood that the systemic risk model is a model that comprehensively considers the supply chain network structure, enterprise correlation and risk propagation mechanism, which can help understand the risk propagation process and the formation mechanism of systemic risk in the supply chain system. In this step, the systemic risk model is used to analyze the connectivity and contagion of the supply chain network, so as to evaluate the degree of influence of the enterprise on the entire network when risks occur in the supply chain. Specifically, the comprehensive data view and risk prediction results are first used to determine the specific risk events and the scope of influence, and then the systemic risk model is used to simulate the risk propagation process in the supply chain network. By considering the correlation between enterprises, the structure of the supply chain network and the path of risk propagation, the degree of influence of the enterprise on the entire supply chain system when risks occur can be evaluated. Finally, according to the evaluation results, the degree of influence of the enterprise on the entire network when risks occur in the supply chain is obtained, and it is converted into a systemic risk score. This score can provide important reference information for financial institutions and enterprises, helping them to better evaluate and manage systemic risks in the supply chain system, thereby reducing risk exposure and ensuring the stable operation of the supply chain system.

[0097] It should be noted that step S700 includes step S710, step S720 and step S730.

[0098] Step S710: Based on the evaluation results and the preset Pareto frontier optimization mathematical model, modeling is performed for the optimization goals of financing cost, risk minimization and liquidity maximization to obtain at least two preliminary financing plans;

[0099] It can be understood that the Pareto frontier optimization mathematical model is a multi-objective optimization method, which can help find the best balance between multiple conflicting objectives, that is, maximize one objective without sacrificing another objective. In this step, the financing cost, risk and liquidity are regarded as the three key objectives in the enterprise financing decision, and they are uniformly modeled and processed using the Pareto frontier optimization mathematical model. First, the weights and importance of financing costs, risks and liquidity are determined according to the evaluation results, and then the Pareto frontier optimization mathematical model is used to reasonably weigh and optimize these objectives. Through mathematical modeling and optimization processing, at least two preliminary financing plans can be obtained, which optimize the financing benefits and risk management of the enterprise as much as possible while meeting multiple objectives. These preliminary financing plans will provide enterprises with a variety of choices and help them formulate the most suitable financing strategies according to actual conditions and needs. At the same time, by using the Pareto frontier optimization method, the mutual influence and trade-off relationship between different objectives can be fully considered, so as to better realize the scientificity and effectiveness of financing decisions.

[0100] Step S720: Perform simulation processing according to all financing schemes, and obtain the optimal financing scheme set by simulating different market environments and risk scenarios using a genetic algorithm and evaluating the performance of each preliminary financing scheme under various possible scenarios;

[0101] Specifically, Figure 2 As shown in the figure, all preliminary financing plans are first simulated and put into the population of the genetic algorithm as individuals. Then, by simulating operations such as natural selection, crossover and mutation, the individuals in the population are optimized and evolved generation by generation until the optimal solution that meets specific conditions is found. In the simulation process, various possibilities under different market environments and risk scenarios are considered to evaluate the performance of each preliminary financing plan under these scenarios. Through the iterative optimization of the genetic algorithm, a set of optimal financing plans can be obtained, which have good performance in various scenarios and can meet the financing needs and goals of the enterprise to the greatest extent.

[0102] Step S730: Based on the preset deep learning, the optimal financing plan set is processed, and the strategy selection and dynamic adjustment processing are performed by setting the reward mechanism and the policy gradient method to obtain the final personalized financing plan.

[0103] It should be noted that deep learning is an artificial intelligence technology that can automatically discover the laws and features in the data by learning a large amount of data, and make effective decisions and predictions. In this step, the preset deep learning model is used to process the optimal financing solution set to further improve the quality and personalization of the financing solution. Specifically, the optimal financing solution set is first used as the input data of the model, and then the deep learning model is used to learn and analyze these data to discover the potential laws and features therein. In the learning process, a corresponding reward mechanism is set to guide the model to evaluate and select different solutions, and the policy gradient method is used to dynamically adjust and optimize the model. Through continuous iterative learning and adjustment, the deep learning model can gradually improve the understanding and optimization capabilities of the financing solution, thereby generating a more personalized and enterprise-oriented financing solution. These personalized solutions can better meet the actual situation and goals of the enterprise, helping it to maximize the financing benefits and minimize the risks.

[0104] Embodiment 2:

[0105] This embodiment provides a financing solution generation system based on supply chain data, including:

[0106] An acquisition module, used to acquire first information and second information, the first information including basic information of enterprises applying for financial services, and the second information including basic information of enterprises in the supply chain, supply chain transaction data, logistics information and financial institution data;

[0107] An evaluation module, used to perform evaluation processing based on the first information and the second information, and obtain a supply chain influence score result of the enterprise by evaluating the importance and substitutability of the enterprise in the supply chain;

[0108] A fusion module, which performs data fusion processing on the second information based on a multi-source data fusion mathematical model of Bayesian reasoning to obtain a supply chain fusion data set;

[0109] The identification module is used to perform identification processing based on the supply chain fusion data set, identify key node enterprises in the supply chain through social network analysis, and evaluate the centrality and connectivity of key node enterprises in the entire supply chain to obtain a supply chain network model;

[0110] The integration module integrates and processes the basic information of enterprises, supply chain influence scoring results and supply chain network models based on the preset graph neural network mathematical model to obtain a comprehensive data view, which includes the network position and relationship strength of enterprises applying for financial services in the supply chain;

[0111] The analysis module is used to perform time series analysis based on the comprehensive data view to obtain evaluation results, including the enterprise's credit evaluation results, risk prediction results, and systemic risk scores in the supply chain;

[0112] The optimization module performs multi-objective optimization processing on the evaluation results based on the preset deep learning mathematical model, generates financing plans for at least two different scenarios, and selects and adjusts the financing plans through a hybrid intelligent algorithm to obtain the final personalized financing plan.

[0113] In this application, the assessment modules include:

[0114] The first evaluation unit is used to evaluate the importance of the enterprise based on the basic information of the enterprise and the supply chain transaction data. By constructing a transaction network diagram and using the PageRank algorithm to calculate the PageRank value of each enterprise, the importance score of the enterprise applying for financial services in the supply chain is obtained;

[0115] The first analysis unit is used to conduct redundant network analysis based on the basic data of enterprises and logistics information, and obtain the substitutability score by calculating the redundancy of each node enterprise in the supply chain and evaluating the position and influence of the enterprise applying for financial services in the logistics path;

[0116] The first processing unit is used to perform weighted comprehensive processing according to the importance score and the substitutability score to obtain a preliminary score result;

[0117] The first adjustment unit is used to make adjustments based on the preliminary scoring results and the historical credit records of the enterprise in the financial institution data, and to obtain the final supply chain influence scoring results by calibrating the preliminary scoring results using a weighted linear regression model.

[0118] In this application, the fusion module includes:

[0119] The first clustering unit is used to perform clustering processing based on the supply chain transaction data and logistics information, use the Gaussian mixture model to process the transaction volume, transaction frequency and logistics node information in the supply chain, and fit the multidimensional normal distribution of the data to obtain the initial multidimensional data distribution;

[0120] The first calculation unit is used to perform conditional summary calculation based on the initial multidimensional data distribution and financial institution data, classify the loan interest rate, credit approval standard and repayment record using a naive Bayes classifier, and calculate the posterior probability of each category to obtain a posterior probability set of all categories;

[0121] The second processing unit is used to perform maximum a posteriori estimation processing according to the posterior probability set, and obtain a fusion causal network reflecting the correlation between supply chain enterprises by constructing a causal relationship graph using a Bayesian network;

[0122] The first sampling unit samples and infers the fusion causal network based on a preset Markov chain Monte Carlo mathematical model to obtain a supply chain fusion data set.

[0123] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0124] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A financing solution generation method based on supply chain data, It is characterized in that include: Acquire first information and second information, wherein the first information includes basic information of the enterprise applying for financial services, and the second information includes basic information of enterprises in the supply chain, supply chain transaction data, logistics information, and financial institution data; Performing evaluation processing based on the first information and the second information, and obtaining a supply chain influence score result of the enterprise by evaluating the importance and substitutability of the enterprise in the supply chain; Performing data fusion processing on the second information based on a multi-source data fusion mathematical model of Bayesian reasoning to obtain a supply chain fusion data set; Performing identification processing based on the supply chain fusion data set, identifying key node enterprises in the supply chain through social network analysis and evaluating the centrality and connectivity of the key node enterprises in the entire supply chain to obtain a supply chain network model; Based on a preset graph neural network mathematical model, data integration and association mining are performed on the basic information of the enterprise, the supply chain influence scoring results and the supply chain network model to obtain a comprehensive data view, which includes the network position and relationship strength of the enterprise applying for financial services in the supply chain; Performing time series analysis and processing according to the comprehensive data view to obtain an evaluation result, wherein the evaluation result includes the credit evaluation result, risk prediction result and systemic risk score of the enterprise in the supply chain; Perform multi-objective optimization processing on the evaluation results based on a preset deep learning mathematical model to generate financing plans under at least two different scenarios, and select and adjust the financing plans through a hybrid intelligent algorithm to obtain a final personalized financing plan; The multi-source data fusion mathematical model based on Bayesian reasoning performs data fusion processing on the second information to obtain a supply chain fusion data set, including: Clustering is performed based on the supply chain transaction data and the logistics information, and a Gaussian mixture model is used to process the transaction volume, transaction frequency, and logistics node information in the supply chain and to fit the multidimensional normal distribution of the data to obtain an initial multidimensional data distribution; Performing conditional summary calculations based on the initial multidimensional data distribution and the financial institution data, using a naive Bayes classifier to classify loan interest rates, credit approval standards, and repayment records, and calculating the posterior probability of each category to obtain a posterior probability set of all categories; Performing maximum a posteriori estimation processing according to the posterior probability set, and constructing a causal relationship graph using a Bayesian network to obtain a fusion causal network that reflects the correlation between supply chain enterprises; Based on a preset Markov chain Monte Carlo mathematical model, the fusion causal network is sampled and inferred to obtain a supply chain fusion data set; The evaluation results are subjected to multi-objective optimization processing based on a preset deep learning mathematical model to generate financing plans under at least two different scenarios, and the financing plans are selected and adjusted through a hybrid intelligent algorithm to obtain a final personalized financing plan, including: Based on the evaluation results and the preset Pareto frontier optimization mathematical model, modeling is performed for the optimization goals of financing cost, risk minimization and liquidity maximization to obtain at least two preliminary financing plans; Perform simulation processing based on all the financing plans, and obtain the optimal financing plan set by simulating different market environments and risk scenarios using genetic algorithms and evaluating the performance of each of the preliminary financing plans under various possible scenarios; The optimal financing plan set is processed based on the preset deep learning, and the strategy selection and dynamic adjustment are performed by setting the reward mechanism and the strategy gradient method to obtain the final personalized financing plan.

2. The method for generating a financing plan based on supply chain data according to claim 1, It is characterized in that An evaluation is performed based on the first information and the second information, and the supply chain influence score of the enterprise is obtained by evaluating the importance and substitutability of the enterprise in the supply chain, including: An importance evaluation process is performed based on the basic information of the enterprise and the supply chain transaction data, and the importance score of the enterprise applying for financial services in the supply chain is obtained by constructing a transaction network diagram and calculating the PageRank value of each enterprise using the PageRank algorithm; Conduct redundant network analysis based on the basic information of the enterprise and the logistics information, calculate the redundancy of each node enterprise in the supply chain and evaluate the position and influence of the enterprise applying for financial services in the logistics path to obtain a substitutability score; A preliminary scoring result is obtained by performing weighted comprehensive processing according to the importance score and the substitutability score; Adjustments are made based on the preliminary scoring results and the historical credit records of the enterprise in the financial institution data, and the final supply chain influence scoring results are obtained by calibrating the preliminary scoring results using a weighted linear regression model.

3. The method for generating a financing plan based on supply chain data according to claim 1, It is characterized in that According to the supply chain fusion data set, identification processing is performed, key node enterprises in the supply chain are identified through social network analysis, and the centrality and connectivity of the key node enterprises in the entire supply chain are evaluated to obtain a supply chain network model, including: Perform K-core decomposition processing on the supply chain fusion data set, and obtain a preliminary key node set by removing low-degree nodes; Calculate the number of times each node serves as a supply chain intermediary node based on the preliminary key node set to obtain a betweenness centrality score; Calculating the degree and distance of each node according to the supply chain fusion data set and the betweenness centrality score to obtain the degree centrality score and closeness centrality score of the node in the supply chain; The weighted clustering coefficient is calculated according to the betweenness centrality score, the degree centrality score and the closeness centrality score, and the final centrality and connectivity of key node enterprises are obtained by calculating the local network closeness of the nodes, and the supply chain network model is obtained by integrating with the supply chain fusion data set.

4. The method for generating a financing plan based on supply chain data according to claim 1, It is characterized in that Based on the preset graph neural network mathematical model, data integration and association mining are performed on the basic information of the enterprise, the supply chain influence scoring results and the supply chain network model to obtain a comprehensive data view, including: Based on a preset graph neural network mathematical model, data integration and association mining are performed on the basic information of the enterprise, the supply chain influence scoring result and the supply chain network model to obtain an association information representation; Performing community division processing according to the association information representation, and obtaining a community division result by dividing the enterprises closely related to each other in the graph into the same community, wherein the community division result includes the network position of each enterprise in the supply chain; Calculate the relationship strength according to the association information representation and the community division result, and obtain the relationship strength between enterprises by calculating the similarity in the association information; Data visualization is performed according to the network positions and the relationship strengths to obtain a comprehensive data view.

5. The method for generating a financing plan based on supply chain data according to claim 1, It is characterized in that The evaluation results are obtained by performing timing analysis processing according to the comprehensive data view, including: Modeling the time series data in the comprehensive data view based on a preset long short-term memory network mathematical model, and obtaining the dynamic credit score time series of the enterprise by capturing the long-term and short-term dependencies in the enterprise credit score data; Perform hidden state estimation on the dynamic credit score time series, and predict the future credit trend of the enterprise by processing the randomness and trend in the time series through the state space model to obtain the future credit evaluation result of the enterprise; Based on the comprehensive data view and the credit assessment result, a Bayesian inference process is performed using a Bayesian dynamic linear model to obtain a risk prediction result, wherein the risk prediction result is the risk level of the enterprise at a specific point in time in the future; Based on the comprehensive data view and the risk prediction results, the connectivity and contagion of the supply chain network are analyzed using a systemic risk model, and an evaluation systemic risk score is obtained by evaluating the impact of a company on the entire network when risks occur in the supply chain.

6. A financing solution generation system based on supply chain data, It is characterized in that include: An acquisition module, used to acquire first information and second information, wherein the first information includes basic information of enterprises applying for financial services, and the second information includes basic information of enterprises in the supply chain, supply chain transaction data, logistics information, and financial institution data; An evaluation module, configured to perform evaluation processing based on the first information and the second information, and obtain a supply chain influence score of the enterprise by evaluating the importance and substitutability of the enterprise in the supply chain; A fusion module, which performs data fusion processing on the second information based on a multi-source data fusion mathematical model of Bayesian reasoning to obtain a supply chain fusion data set; An identification module is used to perform identification processing based on the supply chain fusion data set, identify key node enterprises in the supply chain through social network analysis, and evaluate the centrality and connectivity of the key node enterprises in the entire supply chain to obtain a supply chain network model; An integration module performs data integration and association mining on the basic information of the enterprise, the supply chain influence scoring results and the supply chain network model based on a preset graph neural network mathematical model to obtain a comprehensive data view, wherein the comprehensive data view includes the network position and relationship strength of the enterprise applying for financial services in the supply chain; An analysis module, configured to perform time series analysis processing according to the comprehensive data view to obtain an evaluation result, wherein the evaluation result includes a credit evaluation result, a risk prediction result, and a systemic risk score in the supply chain of the enterprise; An optimization module performs multi-objective optimization processing on the evaluation results based on a preset deep learning mathematical model, generates financing plans under at least two different scenarios, and selects and adjusts the financing plans through a hybrid intelligent algorithm to obtain a final personalized financing plan; Wherein, the fusion module includes: A first clustering unit is used to perform clustering processing according to the supply chain transaction data and the logistics information, use a Gaussian mixture model to process the transaction volume, transaction frequency and logistics node information in the supply chain and fit the multidimensional normal distribution of the data to obtain an initial multidimensional data distribution; A first calculation unit is used to perform conditional summary calculation according to the initial multidimensional data distribution and the financial institution data, classify the loan interest rate, credit approval standard and repayment record using a naive Bayes classifier and calculate the posterior probability of each category to obtain a posterior probability set of all categories; A second processing unit is used to perform maximum a posteriori estimation processing according to the posterior probability set, and obtain a fusion causal network reflecting the correlation between supply chain enterprises by constructing a causal relationship graph using a Bayesian network; A first sampling unit, which performs sampling and inference processing on the fusion causal network based on a preset Markov chain Monte Carlo mathematical model to obtain a supply chain fusion data set; Wherein, the optimization module includes: A first modeling unit is used to perform modeling processing for optimization goals of financing cost, risk minimization and liquidity maximization according to the evaluation results and a preset Pareto frontier optimization mathematical model to obtain at least two preliminary financing plans; A first simulation unit is used to perform simulation processing according to all the financing schemes, and obtain an optimal financing scheme set by simulating different market environments and risk scenarios using a genetic algorithm and evaluating the performance of each of the preliminary financing schemes under various possible scenarios; The first generating unit processes the optimal financing plan set based on preset deep learning, performs strategy selection and dynamic adjustment processing by setting a reward mechanism and a policy gradient method, and obtains a final personalized financing plan.

7. A financing solution generation system based on supply chain data according to claim 6, It is characterized in that The evaluation module includes: The first evaluation unit is used to perform importance evaluation processing based on the basic information of the enterprise and the supply chain transaction data, and obtain the importance score of the enterprise applying for financial services in the supply chain by constructing a transaction network diagram and calculating the PageRank value of each enterprise using the PageRank algorithm; The first analysis unit is used to perform redundant network analysis based on the basic information of the enterprise and the logistics information, and obtain an alternative score by calculating the redundancy of each node enterprise in the supply chain and evaluating the position and influence of the enterprise applying for financial services in the logistics path; A first processing unit, configured to perform weighted comprehensive processing according to the importance score and the substitutability score to obtain a preliminary score result; The first adjustment unit is used to perform adjustment processing according to the preliminary scoring result and the historical credit record of the enterprise in the financial institution data, and obtain the final supply chain influence scoring result by calibrating the preliminary scoring result using a weighted linear regression model.