Blockchain-based customer credit rating method, device, medium and product

Abstract

This application provides a blockchain-based customer credit rating method, apparatus, medium, and product, relating to the fintech field. The method includes: acquiring customer transaction data from the blockchain, and determining the customer's transaction behavior characteristics and transaction network characteristics based on the transaction data; inputting the transaction behavior characteristics and transaction network characteristics into a risk scoring model to obtain a risk node assessment result corresponding to the customer; and inputting the transaction data and risk node assessment result into a credit rating model to generate a credit rating result. This solution leverages the immutability of blockchain data to directly acquire customer transaction data from the blockchain for credit rating, ensuring the authenticity and accuracy of the data; it first assesses risk nodes based on the transaction data, and then obtains a credit rating result based on the risk node assessment result and the transaction data. This dual assessment ensures the accuracy of the customer credit rating result.

Description

Technical Field

[0001] This application relates to the field of financial technology, and in particular to a blockchain-based customer credit rating method, device, medium, and product. Background Technology

[0002] The demand for liquidity in industrial and supply chains has surged, but supply chain financing is heavily reliant on the credit of core enterprises. Financing for upstream and downstream customers requires confirmation of ownership by core enterprises, which limits the financing channels for SMEs.

[0003] Currently, SMEs must apply for financing from core enterprises before seeking further funding. The core enterprise conducts a credit assessment based on the application materials provided by the SME to confirm its creditworthiness and enhance the SME's creditworthiness. Banks then review and disburse loans based on this confirmation document. This reliance on the core enterprise for assessment and confirmation makes SMEs dependent on it for financing, resulting in inefficiency. Furthermore, banks cannot verify the accuracy of the core enterprise's assessment results or the authenticity of the business transactions, leading to risks in the financing transactions.

[0004] Therefore, this method cannot guarantee the accuracy of customer credit assessment results. Summary of the Invention

[0005] This application provides a blockchain-based customer credit rating method, apparatus, medium, and product to address the technical problem of difficulty in ensuring the accuracy of customer credit assessment results.

[0006] In a first aspect, this application provides a blockchain-based customer credit rating method, comprising: obtaining customer transaction data from the blockchain, and determining the customer's transaction behavior characteristics and transaction network characteristics based on the customer's transaction data; inputting the transaction behavior characteristics and transaction network characteristics into a risk scoring model to obtain risk node assessment results corresponding to the customer; wherein, the risk node assessment results are used to indicate the risk scores of multiple nodes that are related to the customer; inputting the customer's transaction data and risk node assessment results into a credit rating model to generate a credit rating result; wherein, the credit rating result is used to indicate the customer's credit rating.

[0007] In one possible implementation, historical transaction data of different customers is acquired, and the historical transaction data is labeled according to a predefined data type to obtain labeled transaction data; the labeled transaction data is then processed by feature alignment using a hybrid federated learning technique to obtain feature data; and a credit rating model is trained based on the feature data and the actual credit rating results to obtain a credit rating model.

[0008] In one possible implementation, determining customer transaction behavior characteristics and transaction network characteristics based on customer transaction data includes: performing feature analysis on customer transaction data; obtaining multiple transaction behavior characteristics corresponding to customer transaction data; wherein, transaction behavior characteristics include order behavior characteristics, fulfillment behavior characteristics, fund behavior characteristics, and association behavior characteristics; determining multiple nodes and multiple edges associated with the customer from the transaction network graph based on the customer transaction data; wherein, the transaction network graph is used to indicate the relationship between transaction data, customers associated with transaction data, and transaction resources; determining transaction network characteristics based on customer transaction data, multiple nodes, and multiple edges; wherein, transaction network characteristics are used to indicate the centrality of corresponding nodes and the connection pattern of corresponding edges.

[0009] In one possible implementation, transaction behavior characteristics and transaction network characteristics are input into a risk scoring model to obtain risk node assessment results corresponding to the customer, including: for any node that is related to the customer, determining the transaction data corresponding to the node from the customer's transaction data; determining the customer's transaction behavior characteristics and transaction network characteristics based on the transaction data corresponding to the node; and inputting the transaction behavior characteristics and transaction network characteristics into a risk scoring model to obtain the risk score of the node.

[0010] In one possible implementation, in response to the customer's business needs, a credit rating standard corresponding to the business needs is determined; based on the credit rating results and the credit rating standard, it is determined whether the customer has the business authority; wherein, the business needs include financing needs.

[0011] In one possible implementation, determining whether a customer has business authority based on credit rating results and credit rating standards includes: determining whether the credit rating meets the credit rating standards; if the credit rating meets the credit rating standards, determining whether the risk scores of multiple nodes are less than a preset score; if the risk scores of multiple nodes are all less than the preset score, determining that the customer has business authority; if the risk score of any node among the multiple nodes is not less than the preset score, determining that the customer does not have business authority.

[0012] In one possible implementation, customer types are defined; a graph neural network model is used to analyze the relationships between different customers based on their transaction data; multi-level relationship information is obtained; the credit rating results, customer types, and multi-level relationship information are used to update the knowledge graph-based supply chain panorama view; the supply chain panorama view is used to indicate the relationships between multiple customers in the supply chain and the credit rating of customers.

[0013] Secondly, this application provides a blockchain-based customer credit rating device, comprising: an acquisition module for acquiring customer transaction data from the blockchain and determining customer transaction behavior characteristics and transaction network characteristics based on the customer transaction data; a first generation module for inputting the transaction behavior characteristics and transaction network characteristics into a risk scoring model to obtain risk node assessment results corresponding to the customer; wherein the risk node assessment results are used to indicate the risk scores of multiple nodes related to the customer; and a second generation module for inputting the customer transaction data and risk node assessment results into a credit rating model to generate a credit rating result; wherein the credit rating result is used to indicate the customer's credit rating.

[0014] In one possible implementation, the acquisition module is further configured to: acquire historical transaction data of different customers, label the historical transaction data according to a predefined data type to obtain labeled transaction data; perform feature alignment processing on the labeled transaction data using hybrid federated learning technology to obtain feature data; and train a model based on the feature data and the real credit rating results to obtain a credit rating model.

[0015] In one possible implementation, the acquisition module is further configured to: perform feature analysis on the customer's transaction data; obtain multiple transaction behavior features corresponding to the customer's transaction data; wherein the transaction behavior features include order behavior features, fulfillment behavior features, fund behavior features, and association behavior features; based on the customer's transaction data, determine multiple nodes and multiple edges associated with the customer from the transaction network graph; wherein the transaction network graph is used to indicate the association between transaction data, customers associated with the transaction data, and transaction resources; determine transaction network features based on the customer's transaction data, multiple nodes, and multiple edges; wherein the transaction network features are used to indicate the centrality of the corresponding nodes and the connection pattern of the corresponding edges.

[0016] In one possible implementation, the first generation module is further configured to: for any node that is associated with a customer, determine the transaction data corresponding to the node from the customer's transaction data; determine the customer's transaction behavior characteristics and transaction network characteristics based on the transaction data corresponding to the node; and input the transaction behavior characteristics and transaction network characteristics into a risk scoring model to obtain the node's risk score.

[0017] In one possible implementation, the processing module is used to: determine the credit rating standard corresponding to the customer's business needs in response to the customer's business needs; and determine whether the customer has the business authority based on the credit rating result and the credit rating standard; wherein the business needs include financing needs.

[0018] In one possible implementation, the processing module is further configured to: determine whether the credit rating meets the credit rating standard; if the credit rating meets the credit rating standard, determine whether the risk scores of multiple nodes are less than a preset score; if the risk scores of multiple nodes are all less than the preset score, determine that the customer has business authorization; if the risk score of any node among the multiple nodes is not less than the preset score, determine that the customer does not have business authorization.

[0019] In one possible implementation, the processing module is further configured to: define customer types; use a graph neural network model to analyze the relationships between different customers based on their transaction data; obtain multi-level relationship information; update the knowledge graph-based supply chain panorama view with credit rating results, customer types, and multi-level relationship information; the supply chain panorama view is used to indicate the relationships between multiple customers in the supply chain and the credit rating of customers.

[0020] Thirdly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the aforementioned method.

[0021] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the aforementioned method.

[0022] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0023] This application provides a blockchain-based customer credit rating method, apparatus, medium, and product. It obtains customer transaction data from the blockchain and determines the customer's transaction behavior characteristics and transaction network characteristics based on this data. The transaction behavior characteristics and transaction network characteristics are input into a risk scoring model to obtain a risk node assessment result corresponding to the customer. The transaction data and risk node assessment result are then input into a credit rating model to generate a credit rating result. This solution leverages the immutability of blockchain data to directly obtain customer transaction data from the blockchain for credit rating, ensuring the authenticity and accuracy of the data. By first assessing risk nodes based on the transaction data, and then obtaining the credit rating result based on the risk node assessment result and the transaction data, this dual assessment ensures the accuracy of the customer credit rating result. Attached Figure Description

[0024] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0025] Figure 1 A flowchart illustrating a blockchain-based customer credit rating method provided in this application. Figure 1 ;

[0026] Figure 2 A flowchart illustrating a blockchain-based customer credit rating method provided in this application. Figure 2 ;

[0027] Figure 3 A flowchart illustrating a blockchain-based customer credit rating method provided in this application. Figure 3 ;

[0028] Figure 4 This is a multi-level relationship diagram among customers;

[0029] Figure 5 A schematic diagram of a blockchain-based customer credit rating device provided in this application;

[0030] Figure 6 This is a schematic diagram of the structure of an electronic device provided in this application.

[0031] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0032] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0033] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of the relevant data all comply with the relevant laws, regulations, and standards of the relevant countries and regions, have taken necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation access points for users to choose to authorize or refuse.

[0034] Furthermore, the technical solution involved in this application, which involves big data analysis of user information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.) and the use of artificial intelligence technology for automated decision-making, and makes decisions that have a significant impact on personal rights based on the results of automated decision-making, provides users with corresponding operation entry points for users to choose to agree to or reject the results of automated decision-making; if the user chooses to reject, the process will proceed to the expert decision-making process.

[0035] It should be noted that the blockchain-based customer credit rating method, device, medium and product provided in this application can be used in the fintech field, or in any field other than fintech. The application field of the blockchain-based customer credit rating method, device, medium and product in this application is not limited.

[0036] Currently, the global industrial and supply chains are experiencing unprecedented complexity and volatility. Factors such as sharp fluctuations in raw material prices, rising logistics costs, and cyclical fluctuations in order demand are intertwined, leading to a surge in the demand for liquidity among SMEs in the upstream and downstream of the industrial chain. However, supply chain financing heavily relies on the credit of core enterprises, and financing for upstream and downstream customers requires confirmation of ownership by core enterprises, which limits the financing channels for SMEs.

[0037] Currently, SMEs must apply for financing from a core enterprise before seeking further funding. The core enterprise conducts a credit assessment and confirms the SME's creditworthiness based on the application materials provided, thus enhancing the SME's creditworthiness. Banks then review and approve loans based on this confirmation document. Specifically, when an SME requests financing, it must first submit detailed application materials to the cooperating core enterprise, including order documents and historical performance records. The core enterprise then conducts a credit assessment and issues a confirmation document, meaning that the core enterprise implicitly guarantees the SME's financing activity with its own credit. Only then will banks and other financial institutions simplify their independent risk control review of the SME based on the core enterprise's confirmation result, and subsequently issue loans.

[0038] However, this approach restricts the financing efficiency and channel expansion of SMEs, and puts banks and other financial institutions in a passive position due to information asymmetry, making it difficult to control financing risks; it also leads to a significant reduction in the accuracy of credit assessment results; therefore, it is difficult to guarantee the accuracy of customer credit assessment results.

[0039] To address the difficulty in ensuring the accuracy of customer credit assessment results in existing technologies, this application provides a blockchain-based customer credit rating method, apparatus, medium, and product. First, customer transaction data is obtained from the blockchain, and based on this data, the customer's transaction behavior characteristics and transaction network characteristics are determined. These characteristics are then input into a risk scoring model to obtain a risk node assessment result corresponding to the customer. Finally, the transaction data and risk node assessment results are input into a credit rating model to generate a credit rating result. This solution leverages the immutability of blockchain data to directly obtain customer transaction data from the blockchain for credit rating, ensuring the authenticity and accuracy of the data. By first assessing risk nodes based on the transaction data, and then obtaining the credit rating result based on both the risk node assessment result and the transaction data, this dual assessment ensures the accuracy of the customer credit assessment results.

[0040] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0041] Example 1

[0042] Figure 1 A flowchart illustrating a blockchain-based customer credit rating method provided in this application. Figure 1 ,like Figure 1 The blockchain-based customer credit rating method provided in this application includes:

[0043] S101 obtains customer transaction data from the blockchain and determines customer transaction behavior characteristics and transaction network characteristics based on the customer transaction data;

[0044] Blockchain is a decentralized distributed ledger technology. Its core features are that it uses cryptography, consensus mechanisms and distributed storage to achieve data immutability, traceability, transparency and collective maintenance, and can ensure the security and credibility of transactions or information records without relying on centralized institutions.

[0045] By using smart contracts on the blockchain to automatically capture transaction data such as purchase orders, transaction vouchers, and transaction records, the efficiency of data acquisition is greatly improved. A smart contract is a piece of automatically executable code stored on the blockchain; when preset conditions are triggered, the contract terms are automatically executed.

[0046] S102, Input the transaction behavior characteristics and transaction network characteristics into the risk scoring model to obtain the risk node assessment results corresponding to the customer; wherein, the risk node assessment results are used to indicate the risk scores of multiple nodes that are related to the customer.

[0047] The nodes include customer nodes and resource nodes. Customer nodes include core enterprise customers and supplier customers from Tier 1 to Tier N. Resource nodes include logistics routes, warehousing centers, etc. The higher the tier number, the smaller the supplier enterprise. For example, Tier 1 supplier customers are large enterprises, Tier 2 supplier customers are medium-sized enterprises, and Tier 3 supplier customers are small enterprises.

[0048] Specifically, when assessing a customer's credit, it is necessary to identify all factors related to the customer. For example, when assessing the credit of customer 1, customer 1's transaction data is obtained, and the customer's transaction behavior characteristics and transaction network characteristics are determined from the transaction data. Among these, the nodes with risk associations with customer 1 include customer 2, customer 3, and the logistics route. For example, by inputting the transaction behavior characteristics and transaction network characteristics into the risk scoring model, the risk score between customer 1 and customer 2 can be obtained as 45 points, the risk score between customer 1 and customer 3 as 25 points, and the risk score between customer 1 and the logistics route as 30 points.

[0049] S103, input the customer's transaction data and risk node assessment results into the credit rating model to generate a credit rating result; the credit rating result is used to indicate the customer's credit level.

[0050] By inputting the risk score between Customer 1 and Customer 2 (45 points), the risk score between Customer 1 and Customer 3 (25 points), the risk score between Customer 1 and the logistics route (30 points), and Customer 1's transaction data into the credit rating model, Customer 1's credit rating is obtained as A-.

[0051] In practical applications, if core enterprises collude with SMEs to fabricate transactions such as false orders, the confirmation documents may become invalid, and financial institutions will still face the risk of bad debts. In addition, fluctuations in the core enterprise's own operations, such as tight cash flow, may also lead to the inability to fulfill the payment commitments made after the confirmation of rights.

[0052] This embodiment provides a blockchain-based customer credit rating method. It obtains customer transaction data from the blockchain and determines the customer's transaction behavior characteristics and transaction network characteristics based on this data. The transaction behavior characteristics and transaction network characteristics are input into a risk scoring model to obtain a risk node assessment result corresponding to the customer. The transaction data and risk node assessment result are then input into a credit rating model to generate a credit rating result. This solution leverages the immutability of blockchain data to directly obtain customer transaction data from the blockchain for credit rating, ensuring the authenticity and accuracy of the data. By first assessing risk nodes based on the transaction data, and then obtaining the credit rating result based on the risk node assessment result and the transaction data, this dual assessment ensures the accuracy of the customer credit rating result.

[0053] In one possible implementation, historical transaction data of different customers is obtained, and the historical transaction data is labeled according to a predefined data type to obtain labeled transaction data.

[0054] Feature data is obtained by performing feature alignment on labeled transaction data using a hybrid federated learning technique.

[0055] A credit rating model is obtained by training the model based on feature data and real credit rating results.

[0056] The predefined data types include trade data types, credit data types, financial data types, and basic data types. Trade data includes information such as upstream and downstream orders and accounts receivable in the supply chain; credit data includes information such as customer credit and customs information; financial data includes customer trade transactions and borrowing information; and basic data includes information such as customer age and marital status.

[0057] Hybrid Federated Learning is an advanced form of Federated Learning that combines the advantages of centralized and distributed federated learning. While protecting data privacy, it incorporates homomorphic encryption and secure multi-party computation to achieve cross-enterprise joint modeling and knowledge sharing while protecting the local data privacy of each participant.

[0058] Specifically, feature alignment processing includes feature definition alignment, feature format alignment, and feature distribution alignment. In practical applications, even data of the same type can differ in the data definitions of different customers or entities. Feature definition differences: Features with the same meaning may have different names; for example, "payment delay days" is recorded as "accounting period deviation" in the core enterprise customer system and as "overdue duration" in the bank system. Feature format differences: The units for numerical features differ; for example, transaction amounts are measured in "ten thousand yuan" for core enterprise customers and in "yuan" for Tier 1 suppliers. The classification standards for categorical features differ; for example, "customer level" is divided into three levels (A / B / C) for core enterprise customers and four levels (Excellent / Good / Medium / Poor) for banks. Feature distribution differences: The same feature is distributed differently in the data of different entities; for example, transaction amounts for core enterprise customers are generally higher, while transaction amounts for Tier N suppliers are concentrated in the lower range.

[0059] Specifically, historical transaction data of different customers are obtained, such as historical transaction data of Tier 1, Tier 2, and Tier 3 supplier customers. The historical transaction data is labeled according to a predefined data type to obtain labeled transaction data. The labeled transaction data is then processed for feature alignment using hybrid federated learning technology to obtain feature data. The model is trained based on the feature data and the real credit rating results to obtain the credit rating model.

[0060] By using hybrid federated learning techniques to align the features of labeled transaction data, feature data is obtained. This allows for a unified standard of transaction data features among different participants, such as core enterprises, SMEs, and banks, while maintaining data confidentiality. The accuracy of the credit rating model is improved by training the model using this feature data.

[0061] In one possible implementation, determining a customer's transaction behavior characteristics and transaction network characteristics based on the customer's transaction data includes:

[0062] Perform feature analysis on customer transaction data to obtain multiple transaction behavior features corresponding to the customer's transaction data; among which, transaction behavior features include order behavior features, fulfillment behavior features, fund behavior features, and related behavior features;

[0063] Based on customer transaction data, multiple nodes and edges associated with the customer are identified from the transaction network graph; the transaction network graph is used to indicate the relationships between transaction data, customers associated with the transaction data, and transaction resources.

[0064] The transaction network characteristics are determined based on the customer's transaction data, multiple nodes, and multiple edges; among them, the transaction network characteristics are used to indicate the centrality of the corresponding nodes and the connection pattern of the corresponding edges.

[0065] Among them, transaction behavior characteristics refer to the dynamic behavior of the transaction participants; transaction behavior characteristics include order behavior characteristics, performance behavior characteristics, fund behavior characteristics, and related behavior characteristics, as shown in Table 1 below.

[0066] Table 1 lists the specific characteristics of a type of transaction behavior.

[0067]

[0068] The transaction network features include the structural attributes of nodes and edges; specifically, in addition to node centrality and edge connection patterns, the transaction network features also include topological evolution features, as shown in Table 2 below.

[0069] Table 2 lists the specific characteristics of the transaction network.

[0070]

[0071] In practical applications, feature analysis is performed on customer transaction data to obtain multiple transaction behavior features corresponding to the customer's transaction data. Based on the customer's transaction data, multiple nodes and edges associated with the customer are identified from the transaction network graph. The transaction network characteristics are then determined based on the customer's transaction data, multiple nodes, and multiple edges. By performing feature analysis on customer transaction data, extracting transaction behavior features, and then combining this with the transaction network graph to mine associated nodes and edges and determine transaction network characteristics, the problem of inaccurate credit assessments leading to financing difficulties for small and medium-sized enterprises due to insufficient data can be solved.

[0072] In one possible implementation, transaction behavior characteristics and transaction network characteristics are input into a risk scoring model to obtain risk node assessment results corresponding to the customer, including:

[0073] For any node that is associated with a customer, determine the corresponding transaction data of the node from the customer's transaction data;

[0074] Based on the transaction data corresponding to the node, the customer's transaction behavior characteristics and transaction network characteristics are determined; the transaction behavior characteristics and transaction network characteristics are input into the risk scoring model to obtain the risk score of the node.

[0075] Based on the above example, let's take the cooperation between Customer 1, a "core enterprise customer," and Customer 2, a "tier 1 supplier," in the supply chain as an example; accurately locate all transaction records related to Customer 2 from Customer 1's transaction data, such as purchase orders within 3 years, payment vouchers (e.g., the amount and payment time of each order), and performance records (e.g., Customer 2's on-time delivery rate).

[0076] Based on the corresponding transaction data, the following transaction behavior characteristics were extracted: From a time perspective, the transaction frequency between Customer 1 and Customer 2 showed quarterly fluctuations, with the transaction frequency in the first and second quarters accounting for 60% of the annual total, corresponding to the peak production season; From a monetary perspective, the cumulative transaction amount reached 23 million yuan, and the standard deviation of the single transaction amount was 85,000 yuan, indicating relatively small fluctuations and strong cooperation stability. However, the payment for the most recent transaction was delayed by 15 days, marking the first time the payment was overdue and causing the on-time payment rate to drop from 100% to 96%.

[0077] By mining the characteristics of the transaction network graph: Customer 2 belongs to supplier 1 in the network. The "weight" of the edge directly connected to Customer 1, that is, the transaction amount accounts for 18% of Customer 1's total purchase amount, which is at a medium level, indicating that Customer 1's dependence on it is moderate. However, Customer 2 has a high centrality "supplied to 3 sister companies of Customer 1 at the same time (through 3 related edges)". If Customer 2 has an anomaly, it may affect the overall procurement chain of Customer 1 and its related companies.

[0078] After integrating transaction behavior characteristics and network characteristics, the data is input into the risk scoring model, and the final output risk score between Customer 1 and Customer 2 is 45 points. This score reflects both the minor anomalies in recent transactions, i.e., the first overdue payment, and also considers the transmission influence of Customer 2 in the network. By obtaining transaction data of customers associated with each customer, transaction behavior characteristics and transaction network characteristics are determined and risk scores are generated, thereby improving the accuracy of risk scoring.

[0079] In one possible implementation, Figure 2 A flowchart illustrating a blockchain-based customer credit rating method provided in this application. Figure 2 ,like Figure 2 As shown, it includes:

[0080] S201, in response to the customer's business needs, determines the credit rating standards corresponding to the business needs;

[0081] The credit rating standards are set according to demand; for example, for customers' financing needs, the credit rating standard is set as Class A rating, including A+, A, and A-.

[0082] S202 determines whether a client has the necessary business authorization based on credit rating results and credit rating standards; business needs include financing needs.

[0083] Based on the above example, the calculated credit rating result for each customer is compared with the corresponding credit rating standard to determine whether the customer has the necessary business authority. By responding to the customer's business needs and determining the corresponding credit rating standard, the system judges whether the customer has the necessary business authority based on the credit rating result and the credit rating standard. Quantifying the standards reduces human intervention, making the assessment results more accurate and truthful.

[0084] In practice, core enterprises are typically only willing to grant credit to their directly cooperating Tier 1 suppliers. Due to the indirect nature of transactions and lack of transparency with Tier 2 and N suppliers, they rarely provide credit granting support, making it difficult for SMEs at the end of the supply chain to access financing. This solution uses blockchain data for credit assessment and determines whether to grant financing based on the assessment results, thus avoiding the limitations imposed on SME financing by core enterprise credit granting.

[0085] In one possible implementation, determining whether a customer has the necessary business authority based on credit rating results and credit rating standards includes:

[0086] Determine whether the credit rating meets the credit rating standards; if the credit rating meets the credit rating standards, determine whether the risk scores of multiple nodes are lower than the preset scores.

[0087] If the risk scores of multiple nodes are all lower than the preset scores, then it is determined that the customer has the business permissions.

[0088] If the risk score of any of the multiple nodes is not less than the preset score, then it is determined that the customer does not have the business authorization.

[0089] The preset score can be set according to the needs, for example, 55 points.

[0090] Building on the previous example, in response to Customer 1's financing needs, the credit rating standard for the financing transaction is determined to be Category A. The risk scores for Customer 1's risk points are as follows: 45 points for the risk between Customer 1 and Customer 2, 25 points for the risk between Customer 1 and Customer 3, and 30 points for the risk between Customer 1 and the logistics route. Customer 1's credit rating is A-. Comparing Customer 1's credit rating with the credit rating standard for the financing transaction shows that Customer 1's credit rating meets the corresponding standard. Judging the risk scores of the nodes associated with Customer 1, the risk scores between Customer 1 and Customer 2, Customer 1 and Customer 3, and Customer 1 and the logistics route are all lower than the preset scores, thus determining that Customer 1 meets the financing conditions. When Customer 1 meets the financing conditions, the bank will calculate a weighted risk score based on Customer 1's multiple risk scores, and combine this with the estimated financing amount to calculate the final financing amount.

[0091] In practical applications, if a customer's credit rating does not meet the corresponding credit rating standards, the customer is determined not to have business authority; if a customer's credit rating meets the corresponding credit rating standards, but any one of their risk scores is not lower than a preset score, the customer is determined not to have business authority.

[0092] After the credit rating is verified, the risk score is further verified, forming a dual risk control barrier of "qualitative rating + quantitative scoring", thereby reducing financing risk.

[0093] In one possible implementation, Figure 3 A flowchart illustrating a blockchain-based customer credit rating method provided in this application. Figure 3 ,like Figure 3 As shown, it includes:

[0094] S301, Defines the customer type;

[0095] The defined customer types include core enterprises and suppliers / distributors from level 1 to level N; in the supply chain, a trade chain of "core enterprise → level 1 → level 2 → ... → level N" will be formed.

[0096] Specifically, Tier 1 entities are partners who directly transact with the core enterprise, including Tier 1 suppliers and Tier 1 distributors. Tier 1 suppliers directly provide raw materials and components to the core enterprise; Tier 1 distributors directly purchase finished products from the core enterprise for resale. For example, a car manufacturer's Tier 1 supplier is the engine plant, and its Tier 1 distributor is the regional 4S stores. Both have frequent and large-scale trade dealings with the core enterprise, typically involving long-term framework agreements and standardized transaction processes.

[0097] Tier 2 Entities: Entities that directly transact with Tier 1 entities but indirectly serve the core enterprise. For example, a Tier 1 supplier, an engine manufacturer, might have a Tier 2 supplier that is a piston ring manufacturer, which supplies parts to the engine manufacturer; a Tier 1 dealer, a 4S store, might have a Tier 2 dealer that is a district / county-level outlet, which purchases goods from the 4S store. Trade between Tier 2 and Tier 1 entities revolves around the production / sales needs of the Tier 1 entity, and the transaction scale is usually smaller than that of Tier 1 entities.

[0098] N-level entities: These are the smallest entities at the end of the chain, acting as suppliers / distributors to N-1 level entities, and are N layers away from the core enterprise. N-level entities have low transaction frequency and small transaction amounts, but they undertake the most basic raw material supply or end-user distribution functions of the chain.

[0099] S302, using a graph neural network model to analyze the relationships between different customers based on customer transaction data; to obtain multi-level relationship information;

[0100] There are multiple layers of transaction relationships in the supply chain. Core enterprises may have transactions with multiple levels of suppliers, and these suppliers have their own downstream suppliers, forming a multi-layered structure.

[0101] For example, a graph neural network model can be used to analyze the relationships between different customers based on the transaction data of customer 1; multi-level relationship information can be obtained; specifically, customer 1 is a level 1 supplier, customer 2 is a level 2 supplier, customer 3 is a level 3 supplier, customer 4 is a level 1 distributor, and customer 5 is a level 2 distributor. Figure 4 A multi-level relationship diagram among customers, such as Figure 4 As shown; Customer 1 can supply goods to core enterprise customers, Customer 2 can supply goods to Customer 1, Customer 3 can supply goods to Customer 1, Customer 2 and core enterprise customers; Customer 4 can pick up goods from core enterprise customers for sale, and Customer 5 can pick up goods from core enterprise customers and Customer 4 for sale.

[0102] S303 updates the knowledge graph-based supply chain panorama view with credit rating results, customer types, and multi-level relationship information; the supply chain panorama view is used to indicate the relationships between multiple customers in the supply chain and the credit rating of customers.

[0103] Traditional supply chain management often only focuses on the directly cooperating Level 1 entities, lacking awareness of Level 2 and higher (N levels) entities. For example, it may be unclear whether Level 2 manufacturers supplying Level 1 suppliers are reliable. Knowledge graphs, through hierarchical connections, can intuitively display the complete chain of "core enterprise → Level 1 → Level 2 → ... → Level N," and can even trace the entire lifecycle of a finished product, such as "a car tire → produced by a Level 1 supplier → its rubber raw material comes from a Level 2 supplier → the rubber is produced from a Level 3 plantation." Furthermore, it allows users to query the credit ratings of customers on the chain.

[0104] Based on the above example, Customer 1's credit rating is A-, Customer 1 is a Tier 1 supplier, Customer 1 directly supplies raw materials to the core enterprise, Customers 2 and 3 supply raw materials to Customer 1, and Customer 3 also directly supplies raw materials to the core enterprise; the credit rating results, customer types, and multi-level relationship information are updated in the knowledge graph-based supply chain panorama view to ensure the timeliness and accuracy of customer credit assessment and information.

[0105] Example 2

[0106] Figure 5 A schematic diagram of a blockchain-based customer credit rating device provided for this application is shown below. Figure 5 As shown, the blockchain-based customer credit rating device 500 provided in this application includes:

[0107] The acquisition module 501 is used to acquire customer transaction data from the blockchain and determine customer transaction behavior characteristics and transaction network characteristics based on the customer transaction data;

[0108] The first generation module 502 is used to input transaction behavior characteristics and transaction network characteristics into the risk scoring model to obtain risk node assessment results corresponding to the customer; wherein, the risk node assessment results are used to indicate the risk scores of multiple nodes that are related to the customer.

[0109] The second generation module 503 is used to input transaction data and risk node assessment results into the credit rating model to generate credit rating results; wherein, the credit rating results are used to indicate the customer's credit rating.

[0110] The blockchain-based customer credit rating device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0111] Figure 6 This is a schematic diagram of the structure of an electronic device provided in this application. Figure 6 As shown, the electronic device 60 provided in this embodiment includes at least one processor 601 and a memory 602. Optionally, the device 60 further includes a communication component 603. The processor 601, memory 602, and communication component 603 are connected via a bus 604.

[0112] In a specific implementation, at least one processor 601 executes computer execution instructions stored in memory 602, causing at least one processor 601 to perform the above-described method.

[0113] The specific implementation process of processor 601 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0114] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method; its implementation principle and technical effect are similar and will not be described in detail here.

[0115] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the technical solution of the above method embodiments. Its implementation principle and technical effects are similar, and will not be repeated here.

[0116] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0117] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0118] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0119] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.

[0120] When integrated units / modules are implemented in hardware, the hardware can be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor can be any suitable hardware processor, such as a CPU, GPU, FPGA, DSP, and ASIC, etc. Unless otherwise specified, the storage unit can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc.

[0121] If the integrated unit / module is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0122] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0123] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0124] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A blockchain-based customer credit rating method, characterized in that, include: Obtain customer transaction data from the blockchain, and determine customer transaction behavior characteristics and transaction network characteristics based on the customer transaction data; The transaction behavior characteristics and the transaction network characteristics are input into the risk scoring model to obtain the risk node assessment results corresponding to the customer; wherein, the risk node assessment results are used to indicate the risk scores of multiple nodes that are related to the customer. The customer's transaction data and the risk node assessment results are input into the credit rating model to generate a credit rating result; wherein, the credit rating result is used to indicate the customer's credit level.

2. The method according to claim 1, characterized in that, The method further includes: Obtain historical transaction data from different customers, and annotate the historical transaction data according to a predefined data type to obtain annotated transaction data; The labeled transaction data is processed by feature alignment using a hybrid federated learning technique to obtain feature data; The credit rating model is obtained by training the model based on the feature data and the actual credit rating results.

3. The method according to claim 1, characterized in that, The process of determining the customer's transaction behavior characteristics and transaction network characteristics based on the customer's transaction data includes: The transaction data of the customer is subjected to feature analysis to obtain multiple transaction behavior features corresponding to the transaction data of the customer; wherein, the transaction behavior features include order behavior features, fulfillment behavior features, fund behavior features, and association behavior features; Based on the customer's transaction data, multiple nodes and edges associated with the customer are identified from the transaction network graph; wherein, the transaction network graph is used to indicate the relationship between transaction data, customers associated with the transaction data, and transaction resources; The transaction network characteristics are determined based on the customer's transaction data, the multiple nodes, and the multiple edges; wherein, the transaction network characteristics are used to indicate the centrality of the corresponding nodes and the connection pattern of the corresponding edges.

4. The method according to claim 1, characterized in that, The step of inputting the transaction behavior characteristics and the transaction network characteristics into the risk scoring model to obtain the risk node assessment results corresponding to the customer includes: For any node that is associated with the customer, determine the transaction data corresponding to the node from the customer's transaction data; Based on the transaction data corresponding to the node, the customer's transaction behavior characteristics and transaction network characteristics are determined; the transaction behavior characteristics and transaction network characteristics are input into the risk scoring model to obtain the risk score of the node.

5. The method according to claim 1, characterized in that, The method further includes: In response to the customer's business needs, determine the corresponding credit rating standards for those business needs; The credit rating results and credit rating standards are used to determine whether the customer has the necessary business authorization; wherein, the business needs include financing needs.

6. The method according to claim 5, characterized in that, The determination of whether a customer has the necessary business authorization based on the credit rating result and the credit rating criteria includes: Determine whether the credit rating meets the credit rating standard; if the credit rating meets the credit rating standard, determine whether the risk scores of the multiple nodes are less than a preset score; If the risk scores of all the nodes are less than the preset score, then it is determined that the customer has the business permissions. If the risk score of any of the multiple nodes is not less than a preset score, then it is determined that the customer does not have business permissions.

7. The method according to claim 2, characterized in that, The method further includes: Define customer types; Using a graph neural network model, the relationships between different customers are analyzed based on the customer's transaction data to obtain multi-level relationship information. The credit rating results, customer types, and multi-level relationship information are used to update the knowledge graph-based supply chain panorama view; the supply chain panorama view is used to indicate the relationships between multiple customers in the supply chain and the credit rating of customers.

8. A blockchain-based customer credit rating device, characterized in that, include: The acquisition module is used to acquire customer transaction data from the blockchain and determine customer transaction behavior characteristics and transaction network characteristics based on the customer transaction data; The first generation module is used to input the transaction behavior characteristics and the transaction network characteristics into the risk scoring model to obtain the risk node assessment results corresponding to the customer; wherein, the risk node assessment results are used to indicate the risk scores of multiple nodes that are related to the customer. The second generation module is used to input the customer's transaction data and the risk node assessment results into the credit rating model to generate a credit rating result; wherein the credit rating result is used to indicate the customer's credit level.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

Images