Supply chain financing risk processing method and device based on 5G message

Abstract

The invention provides a supply chain financing risk processing method and device based on a 5G message, and relates to the field of computer data processing technology and financial science and technology, and the method comprises the steps: obtaining a financing business request through a 5G message processing node; according to the financing service request, obtaining identity information, financing service materials and a mobile phone number of the first customer, and identity information and a mobile phone number of a second customer; if the mobile phone number is an untrusted mobile phone number, identity verification is carried out on client identity information through a client resource library; if the mobile phone number is the trusted mobile phone number, inquiring account information in a bank background database according to the trusted mobile phone number of the first client and the trusted mobile phone number of the second client under the condition that the first client and the second client authorize inquiry; determining a supply chain feature according to the financing business material, inputting the account information and the supply chain feature into a financing risk identification model, and determining a supply chain financing risk identification result; and performing risk control by adopting a corresponding risk control strategy according to the supply chain financing risk identification result.

Description

Technical Field

[0001] The present invention relates to computer data processing technology, the field of fintech, and particularly to a method and device for processing supply chain financing risks based on 5G messages. Background Art

[0002] This section aims to provide background or context for the embodiments of the present invention described in the claims. The description herein is not admitted to be prior art merely by including it in this section.

[0003] Supply chain financing is a financing model that takes the core enterprises in the supply chain and their related upstream and downstream supporting enterprises as a whole, and formulates an overall financial solution based on the control of goods rights and cash flow according to the transaction relationships and industry characteristics of the enterprises in the supply chain. Supply chain financing solves the problems of difficult financing and difficult guarantee for upstream and downstream enterprises. Moreover, by breaking through the financing bottlenecks of upstream and downstream, it can also reduce the financing cost of the supply chain and improve the competitiveness of core enterprises and supporting enterprises. Supply chain financing is a financing model that formulates an overall financial solution based on the control of goods rights and cash flow according to the transaction relationships and industry characteristics of the enterprises in the supply chain, taking the core enterprises in the supply chain and their related upstream and downstream supporting enterprises as a whole. Small, medium and micro enterprises in the supply chain can provide an important force for the sustainable development of the supply chain. Therefore, it is particularly important to analyze the supply chain financing risks for the development of enterprises, the sustainable and stable development of the supply chain, and the capital security of capital providers.

[0004] However, currently in the supply chain financing system, the situations of corporate legal person customers are complex and changeable. They can play different financing roles in different chains. However, due to the lack of communication of risk information of customers in different chains, there are differences in the risk levels of a customer in different chains; and the customer data is messy and disorderly, and the impact of each feature on the financing risk cannot be accurately evaluated. The efficiency of the financial data acquisition method and identity verification method of customers is low, and the financing risk status of customers cannot be analyzed timely and accurately, and reasonable and effective risk control cannot be carried out on the financing risk.

[0005] In summary, there is an urgent need for a technical solution that can overcome the above defects, improve the data acquisition and identity verification methods, and improve the financing risk analysis and identification methods. Summary of the Invention

[0006] To solve the problems existing in the prior art, the present invention proposes a method and device for processing supply chain financing risks based on 5G messaging. The present invention uses 5G messaging to transmit data between customers and banks, realizes the judgment of trusted mobile phone numbers based on 5G messaging nodes, provides technical support for customer identity recognition, and further identifies risks for financing business materials and customer information through a financing risk identification model, thereby effectively controlling supply chain financing risks. The overall solution of the present invention realizes digital, automated, and intelligent full-process real-time, dynamic, and automatic risk control.

[0007] In the first aspect of the embodiments of the present invention, a method for processing supply chain financing risks based on 5G messaging is proposed. The method includes:

[0008] Obtain a financing business request through a 5G messaging processing node; wherein, the financing business request is sent by a first customer through a 5G mobile terminal to the 5G messaging processing node;

[0009] Obtain the identity information, financing business materials, and mobile phone number of the first customer, and the identity information and mobile phone number of the second customer according to the financing business request; wherein, the first customer is the financing requestor, and the second customer is the fund provider; the 5G messaging processing node determines whether the mobile phone number is a trusted mobile phone number based on a trusted mobile phone number database;

[0010] If the mobile phone number of the first customer or the second customer is an untrusted mobile phone number, authenticate the identity information of the customer through a customer resource library; after successful authentication, determine that the mobile phone number is a trusted mobile phone number;

[0011] Batch send the newly added trusted mobile phone numbers to the trusted mobile phone number database of the 5G messaging processing node according to a preset time point;

[0012] If the mobile phone numbers of the first customer and the second customer are trusted mobile phone numbers, and with the authorization of the first customer and the second customer to query, query the account information in the bank's back-end database according to the trusted mobile phone number of the first customer and the trusted mobile phone number of the second customer;

[0013] Determine the supply chain characteristics according to the financing business materials, input the account information and the supply chain characteristics into a financing risk identification model, and determine the supply chain financing risk identification result; wherein, the financing risk identification model is generated through machine learning training based on the mapping relationship between the account information and the supply chain characteristics and the supply chain financing risk characteristics;

[0014] According to the supply chain financing risk identification result, adopt corresponding risk control strategies to control risks.

[0015] In the second aspect of the embodiments of the present invention, a device for processing supply chain financing risks based on 5G messaging is proposed. The device includes:

[0016] a request acquisition module, configured to acquire a financing service request through a 5G message processing node; wherein the financing service request is sent by a first customer to the 5G message processing node through a 5G mobile terminal;

[0017] An information acquisition module is configured to obtain the identity information, financing business materials, and mobile phone number of the first customer, as well as the identity information and mobile phone number of the second customer, based on the financing business request; wherein the first customer is the financing requester and the second customer is the funding provider; the 5G message processing node determines whether the mobile phone number is a trusted mobile phone number based on the trusted mobile phone number database;

[0018] An identity authentication module is used to authenticate the identity information of the customer through the customer resource library if the mobile phone number of the first customer or the second customer is an untrusted mobile phone number; after successful verification, determine that the mobile phone number is a trusted mobile phone number;

[0019] A batch sending module is used to send the newly added trusted mobile phone numbers in batches to the trusted mobile phone number database of the 5G message processing node according to the preset time point;

[0020] An information query module is configured to query the bank's backend database for account information based on the first customer's trusted mobile phone number and the second customer's trusted mobile phone number if the first customer and the second customer's mobile phone numbers are trusted mobile phone numbers and the first customer and the second customer authorize the query;

[0021] a risk identification module, configured to determine supply chain characteristics based on the financing business materials, input the account information and supply chain characteristics into a financing risk identification model, and determine a supply chain financing risk identification result; wherein the financing risk identification model is generated by machine learning training based on the mapping relationship between the account information and supply chain characteristics and the supply chain financing risk characteristics;

[0022] The risk control module is used to perform risk control using corresponding risk control strategies based on the supply chain financing risk identification results.

[0023] In a third aspect of an embodiment of the present invention, a computer device is proposed, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements a supply chain financing risk handling method based on 5G messages when executing the computer program.

[0024] In a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is proposed, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, a supply chain financing risk handling method based on 5G messages is implemented.

[0025] In a fifth aspect of the embodiments of the present invention, a computer program product is provided. The computer program product includes a computer program which, when executed by a processor, implements a method for processing supply chain financing risks based on 5G messages.

[0026] The supply chain financing risk handling method and device based on 5G messages proposed in the present invention obtains a financing business request through a 5G message processing node; wherein, the financing business request is sent by a first customer to the 5G message processing node through a 5G mobile terminal; according to the financing business request, the identity information, financing business materials and mobile phone number of the first customer, as well as the identity information and mobile phone number of the second customer are obtained; wherein, the first customer is the financing requester, and the second customer is the funding party; the 5G message processing node determines whether the mobile phone number is a trusted mobile phone number based on the trusted mobile phone number database; if the mobile phone number of the first customer or the second customer is an untrusted mobile phone number, the identity information of the customer is authenticated through the customer resource library; after successful verification, the mobile phone number is determined to be a trusted mobile phone number; according to the preset time The node sends the newly added trusted mobile phone numbers in batches to the trusted mobile phone number database of the 5G message processing node; if the mobile phone numbers of the first customer and the second customer are trusted mobile phone numbers, the account information is queried in the bank's back-end database based on the trusted mobile phone numbers of the first customer and the second customer when the first customer and the second customer authorize the query; the supply chain characteristics are determined based on the financing business materials, and the account information and supply chain characteristics are input into the financing risk identification model to determine the supply chain financing risk identification result; wherein, the financing risk identification model is generated by machine learning training through the mapping relationship between the account information and supply chain characteristics and the supply chain financing risk characteristics; according to the supply chain financing risk identification result, the corresponding risk control strategy is adopted to perform risk control. The overall solution allows users to upload information through the terminal using 5G messages, realize the rapid transmission of messages and data between customers and banks, realize the judgment of trusted mobile phone numbers based on 5G message nodes, provide technical support for customer identity identification, and improve the efficiency of subsequent identity verification by financial institutions; sort out financing business materials to obtain reasonable, orderly, and financing business-related supply chain characteristics; query account information based on the customer's trusted mobile phone number, use supply chain characteristics and customer information as samples for risk identification, and perform risk identification through the financing risk identification model, thereby effectively controlling supply chain financing risks. The financing risk identification model uses multi-dimensional data for mapping training, which can effectively improve the accuracy of model identification and the effect of evaluation. The present invention uses 5G messages to realize rapid data transmission of information, and judges trusted mobile phone numbers through operators to realize identity verification, improve the identity verification method in financial scenarios, and improve identity verification efficiency; use financing risk identification models and multi-dimensional data for risk identification, effectively control supply chain financing risks, provide strong protection for enterprise development and the property safety of funding parties, and provide strong technical support for financial technology scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0028] Figure 1 It is a schematic flowchart of a method for processing supply chain financing risks based on 5G messages according to an embodiment of the present invention.

[0029] Figure 2 It is a schematic diagram of an exemplary scenario according to an embodiment of the present invention.

[0030] Figure 3 It is a schematic flowchart of a process for processing trusted mobile phone numbers according to an embodiment of the present invention.

[0031] Figure 4 It is a schematic flowchart of a process for authenticating the identity information of a customer through a customer resource library according to an embodiment of the present invention.

[0032] Figure 5 It is a schematic flowchart of a process for determining supply chain characteristics based on financing business materials according to an embodiment of the present invention.

[0033] Figure 6 It is a schematic flowchart of a process for training a feature extraction model according to an embodiment of the present invention.

[0034] Figure 7 It is a schematic flowchart of a process for screening supply chain characteristics according to an embodiment of the present invention.

[0035] Figure 8 It is a schematic flowchart of a process for training a financing risk identification model according to an embodiment of the present invention.

[0036] Figure 9 It is a schematic flowchart of a process for financing risk identification according to an embodiment of the present invention.

[0037] Figure 10 It is a schematic flowchart of a process for risk control according to an embodiment of the present invention.

[0038] Figure 11 It is a schematic flowchart of a process for processing supply chain financing risks based on 5G messages according to a specific embodiment of the present invention.

[0039] Figure 12 It is a schematic diagram of the architecture of a device for processing supply chain financing risks based on 5G messages according to an embodiment of the present invention.

[0040] Figure 13 It is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. Detailed implementation manners

[0041] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided solely to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. Rather, these embodiments are provided to make this disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0042] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present disclosure may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software.

[0043] According to an embodiment of the present invention, a method and apparatus for managing supply chain financing risks based on 5G messaging are proposed, relating to computer data processing technology and the field of financial technology. This invention utilizes 5G messaging to transmit data between customers and banks, and uses 5G messaging nodes to determine trusted mobile phone numbers, providing technical support for customer identity verification. Furthermore, a financing risk identification model is used to identify risks in financing business data and customer information, thereby effectively controlling supply chain financing risks. The overall solution of this invention achieves real-time, dynamic, and automated risk control throughout the entire process, which is digital, automated, and intelligent.

[0044] The principles and spirit of the present invention are explained in detail below with reference to several representative embodiments of the present invention.

[0045] Figure 1 This is a flow chart of a supply chain financing risk management method based on 5G messaging according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0046] S101, obtaining a financing service request through a 5G message processing node; wherein the financing service request is sent by a first customer to the 5G message processing node through a 5G mobile terminal;

[0047] S102: Obtaining, based on the financing business request, the identity information, financing business materials, and mobile phone number of the first customer, as well as the identity information and mobile phone number of the second customer; wherein the first customer is the financing requester and the second customer is the funding provider; and determining, by the 5G message processing node, whether the mobile phone number is a trusted mobile phone number based on a trusted mobile phone number database;

[0048] S103, if the mobile phone number of the first customer or the second customer is an untrusted mobile phone number, authenticate the customer's identity information through the customer resource library; after successful authentication, determine that the mobile phone number is a trusted mobile phone number;

[0049] S104, Batch send the newly added trusted mobile phone numbers to the trusted mobile phone number database of the 5G message processing node according to the preset time points;

[0050] S105, If the mobile phone numbers of the first customer and the second customer are trusted mobile phone numbers, and under the authorization of the first customer and the second customer to query, query the account information in the bank back-end database according to the trusted mobile phone number of the first customer and the trusted mobile phone number of the second customer;

[0051] S106, Determine the supply chain characteristics according to the financing business materials, input the account information and the supply chain characteristics into the financing risk identification model, and determine the supply chain financing risk identification result; wherein, the financing risk identification model is generated by machine learning training through the mapping relationship between the account information and the supply chain characteristics and the supply chain financing risk characteristics;

[0052] S107, According to the supply chain financing risk identification result, adopt the corresponding risk control strategy to control the risk.

[0053] In the actual application scenario, the supply chain financing risk processing method based on 5G messages proposed by the present invention enables users to upload materials through the terminal using 5G messages, realizes the rapid transmission of messages and data between customers and banks, and judges the trusted mobile phone numbers based on the 5G message nodes, providing technical support for customer identity recognition and improving the efficiency of subsequent identity verification by financial institutions; sort out the financing business materials to obtain reasonable, orderly, and financing business-related supply chain characteristics; query the account information based on the trusted mobile phone numbers of customers, use the supply chain characteristics and customer information as samples for risk identification, and perform risk identification through the financing risk identification model, so as to effectively control the supply chain financing risk. The financing risk identification model uses multi-dimensional data for mapping training, which can effectively improve the accuracy of model identification and the evaluation effect. The present invention uses 5G messages to realize the rapid transmission of information data, judges the trusted mobile phone numbers through the operator to realize identity verification, improves the identity verification method in the financial scenario, and improves the identity verification efficiency; adopts the financing risk identification model and multi-dimensional data for risk identification, effectively controls the supply chain financing risk, provides strong guarantee for the development of enterprises and the property safety of capital providers, and provides strong technical support for the financial technology scenario.

[0054] In order to explain the above-mentioned supply chain financing risk processing method based on 5G messages more clearly, the following will be described in detail in combination with each step.

[0055] In one embodiment, refer to Figure 2 , which is an exemplary scenario schematic diagram of an embodiment of the present invention. As Figure 2As shown, the exemplary scenario includes a client 100, an operator 200, and a banking system 300. A communication connection is maintained between the client 100 and the operator 200, and the operator 200 is communicatively connected to the banking system 300. The client 100 can be a 5G mobile terminal. A customer can send a financing business request through the client 100, verify the customer's identity, process financial business, etc. The operator 200 is a 5G message processing node, which sets up a trusted mobile phone number database to verify the customer's mobile phone number and determine whether it is a trusted mobile phone number. In an actual application scenario, after being authenticated by the operator 200, the banking system 300 is a bank-end node, which performs information query based on the trusted mobile phone number (it should be noted that in the present invention, the storage and reading of customer information require customer authorization), performs risk identification through a financing risk identification model, and processes the financing business based on the financing risk identification result.

[0056] The present invention analyzes financing business materials through a 5G message processing node and a bank-end node to identify financing risks. While ensuring the control of financing risks, it improves the financing business processing flow, provides reasonable financing business processing services for enterprises, and reduces financing risks. The overall solution realizes the management of full-digital, automated, and intelligent full-process real-time, dynamic, and automatic risk control and warning, thereby improving service efficiency and service breadth, and providing overall services for both the financing side and the capital side.

[0057] In one embodiment, S101, obtain a financing business request through a 5G message processing node; wherein, the financing business request is sent by a first customer through a 5G mobile terminal to the 5G message processing node.

[0058] Specifically, the first customer is usually the financing requestor; the opposite of the financing requestor is the capital provider (the second customer), and the capital provider provides funds to the first customer.

[0059] In one embodiment, S102, obtain the identity information, financing business materials, and mobile phone number of the first customer, and the identity information and mobile phone number of the second customer according to the financing business request; wherein, the first customer is the financing requestor, and the second customer is the capital provider; the 5G message processing node determines whether the mobile phone number is a trusted mobile phone number based on the trusted mobile phone number database.

[0060] For trusted mobile phone numbers, the present invention uses a multi-factor authentication method to determine the trusted mobile phone numbers of customers, and uses the trusted mobile phone number database to store the trusted mobile phone numbers to realize the judgment of whether the mobile phone number is a trusted mobile phone number. Specifically, refer to Figure 3 , the method for verifying the customer's identity and determining the trusted mobile phone number is:

[0061] S301, construct a trusted mobile phone number database;

[0062] S302. Perform real-name authentication on the customer's mobile phone number based on the customer's identity document and biometric characteristics. After successful authentication, determine the customer's mobile phone number as a trusted mobile phone number; otherwise, determine the customer's mobile phone number as an untrusted mobile phone number.

[0063] S303. Add the trusted mobile phone number to the trusted mobile phone number database.

[0064] Real-name authentication can adopt a dual authentication method of identity documents and biometric characteristics. Among them, when verifying the identity document, the customer is required to provide a valid identity document, such as a second-generation ID card. The information on the identity document (such as name, ID number, etc.) will be recorded by the system and compared with the records in the database to verify its authenticity and validity. For biometric verification, the customer's biometric information can be collected, such as fingerprints, facial recognition, iris scanning, or voiceprint recognition. The collection of biometric characteristics needs to be carried out through secure sensors and devices to ensure the accuracy and security of the data. The collected biometric data will be matched with the information on the identity document to ensure that the biometric characteristics belong to the document holder and prevent identity theft and fraud.

[0065] In actual application scenarios, liveness detection can also be performed to improve security, and the authentication process will also include liveness detection technology. This technology can identify whether it is a real biometric characteristic, rather than a photo, video, or other forged biometric samples. Liveness detection can be achieved in various ways, such as requiring the user to perform specific actions or expressions (for facial recognition), or testing certain natural reactions of biometric characteristics (such as pulse or body temperature). It should be noted that during the entire authentication process, the customer's personal information and biometric data will be strictly protected. This includes measures such as data encryption, secure storage, and access control to prevent data leakage, abuse, or unauthorized access. When the identity document verification, biometric matching, and liveness detection are all passed, the customer's mobile phone number will be successfully real-name authenticated and determined as a trusted mobile phone number. If any inconsistencies or suspicious situations are found during the authentication process, the system will prompt further verification steps or require the customer to provide additional information for verification.

[0066] In one embodiment, S103. If the mobile phone number of the first customer or the second customer is an untrusted mobile phone number, verify the customer's identity information through the customer resource library; after successful verification, determine the mobile phone number as a trusted mobile phone number.

[0067] Among them, the customer resource library is the database of peer banks;

[0068] Specifically, refer to Figure 4 , verifying the customer's identity information through the customer resource library includes:

[0069] S401. Query whether there is customer information corresponding to the untrusted mobile phone number in the customer resource library;

[0070] S402. If it exists, perform identity comparison on the customer according to the customer information corresponding to the untrusted mobile phone number queried. If the comparison is successful, determine that the identity verification is successful;

[0071] S403. If it does not exist, call the identity authentication interface to perform real-name identity verification on the customer, and add the customer information with successful identity verification to the peer bank database.

[0072] In this embodiment, the identity verification is connected to the customer resource library, which can be the peer bank database of the bank node; for the untrusted mobile phone number, query the customer information corresponding to the untrusted mobile phone number based on the customer resource library to check the customer identity.

[0073] In the actual application scenario, the customer information corresponding to the untrusted mobile phone number may not be found in the customer resource library, then the identity verification method mentioned in S302 can be used to verify the customer identity.

[0074] Specifically, adopt the joint verification technology, combine the customer's identity document information and biometric data to perform real-name identity authentication on the customer's mobile phone number. This method can improve the accuracy and security of the authentication. Verify through the identity document information provided by the customer, including name, ID number, etc. Verify through the customer's biometric data, such as fingerprints, facial recognition, etc. Jointly verify the identity document information and biometric data to ensure the consistency and accuracy of both. Add the verification result to the peer bank database, and at the same time bind it to the untrusted mobile phone number, and this untrusted mobile phone number can be determined as a trusted mobile phone number.

[0075] In one embodiment, S104. Batch send the newly added trusted mobile phone numbers to the trusted mobile phone number database of the 5G message processing node according to the preset time point.

[0076] For the mobile phone number with successful identity verification, use it as a trusted mobile phone number.

[0077] In one embodiment, S105. If the mobile phone numbers of the first customer and the second customer are trusted mobile phone numbers, under the condition of the authorization query of the first customer and the second customer, query the account information in the bank back-end database according to the trusted mobile phone numbers of the first customer and the second customer.

[0078] The user uploads data through the terminal using 5G messages to achieve the rapid transmission of messages and data between the customer and the bank. Based on the 5G message node, judge the trusted mobile phone number, provide technical support for customer identity recognition, and improve the efficiency of subsequent identity verification by financial institutions.

[0079] In one embodiment, in S106, supply chain characteristics are determined according to the financing business materials, the account information and the supply chain characteristics are input into a financing risk identification model, and a supply chain financing risk identification result is determined; wherein, the financing risk identification model is generated through machine learning training based on the mapping relationship between the account information and the supply chain characteristics and the supply chain financing risk characteristics.

[0080] Reference Figure 5 , the detailed process of determining the supply chain characteristics according to the financing business materials is as follows:

[0081] In S501, the financing business materials of the first customer are identified through an OCR algorithm, and the recognition result is structurally converted to obtain financing business text content; the financing business text content includes registered basic data, credit investigation data, and transaction data;

[0082] Among them, for the first customer and the second customer, they can be corporate customers, and can be supply chain financing information corresponding to multiple supply chain chains obtained by the supply chain financing system through docking with the enterprise resource system. In the supply chain financing information, the registered basic data may include the enterprise organization code, enterprise social credit code, unit type, registration date, enterprise scale, industry, registered address, registered capital, administrative division, and whether it is listed; the credit investigation data may include the number of loan overdue periods, overdue amount, asset disposal information, etc.; the transaction data is mainly the relevant data of the enterprise's transactions with upstream and downstream in the supply chain.

[0083] Specifically, for the financing business materials, they are usually collected by scanning; the content of the scanned picture is recognized by a Chinese OCR algorithm to convert the picture content into text content; for common problems such as blurred, tilted, and distorted images, image preprocessing, automatic positioning, tilt correction, noise elimination are performed in sequence, and the interference of text line breaks and seal stamping is optimized to minimize the impact of interference and convert the picture content into text content.

[0084] In S502, the financing business text content is preprocessed to obtain preprocessed supply chain financing information;

[0085] In S503, feature extraction is performed on the preprocessed supply chain financing information by using feature engineering to obtain the supply chain feature set;

[0086] In S504, the supply chain feature set is screened through a pre-trained feature extraction model to obtain the supply chain features associated with the financing business.

[0087] Since there may be invalid information in the text content of the financing business, such as missing or incorrect information, it is necessary to preprocess the obtained supply chain financing information. Specifically, in S502, preprocess the text content of the financing business to obtain the preprocessed supply chain financing information, including:

[0088] Classify, clean, or balance the dataset of the text content of the financing business, eliminate incorrect data, and fill in missing data. In actual application scenarios, various data preprocessing methods can be adopted. Through data processing, the data quality can be improved to enhance the accuracy of the feature model screening results. Supply chain features can also be used as model training samples. The supply chain features after data preprocessing can enhance the accuracy of model training and improve the model performance.

[0089] Furthermore, for the preprocessed supply chain financing information, feature engineering is used to extract features to obtain the set of supply chain features. Since the set of supply chain features contains the content of each supply chain of the first customer, there may be irrelevant features to the determination of financing risks. If the financing risks are determined based on all the obtained supply chain features at this time, the irrelevant features will not only interfere with the determination of financing risks, but also affect the determination efficiency of financing risks due to the excessive amount of data. Therefore, the present invention uses a feature extraction model to screen the set of supply chain features to determine the supply chain features associated with the financing business.

[0090] For the feature extraction model, refer to Figure 6 , the detailed process of training the feature extraction model is as follows:

[0091] S601, obtain the supply chain features of the sample customer, and the supply chain features of the sample customer are composed of the supply chain training set and the supply chain test set of the sample customer;

[0092] S602, use the first genetic algorithm to train the supply chain training set of the sample customer to obtain an initial model;

[0093] S603, use the supply chain test set of the sample customer to test the initial model, and use the initial model that passes the test as the feature extraction model.

[0094] Specifically, the genetic algorithm is a search heuristic algorithm that simulates natural selection and genetic mechanisms and is used to solve optimization and search problems. In the field of machine learning, genetic algorithms can be used to train classification models, especially in aspects such as feature selection, model parameter optimization, and structure design. In a genetic algorithm, first, an encoding scheme needs to be defined to convert the parameters or structure of the classification model into chromosomes. For example, for a decision tree classifier, the chromosome can be represented as a sequence of feature selections and an arrangement of decision nodes. When initializing the population, a set of initial solutions is randomly generated as the starting point of the population. Each solution is a potential classification model, and its performance can be measured through cross-validation or other evaluation methods. Evaluate the fitness, and evaluate the fitness of each individual (classification model) in the population. The fitness function is usually based on the performance of the model on the training set, such as accuracy, recall, F1 score, etc. The higher the fitness of an individual, the greater the likelihood of being selected for reproduction. According to the results of the fitness evaluation, select excellent individuals for reproduction. Generate a new population according to selection, crossover, and mutation operations. The new population will be used for the next round of fitness evaluation and evolution process. Repeat the above process until a certain termination condition is met, such as reaching a predetermined number of iterations, the fitness reaching a certain threshold, or the performance no longer improving significantly. Finally, decode the individual with the highest fitness in the final population into a specific classification model to obtain the optimal classifier trained by the genetic algorithm.

[0095] Specifically, refer to Figure 7 In S504, the detailed process of screening the supply chain feature set through the pre-trained feature extraction model to obtain the supply chain features associated with the financing business is as follows:

[0096] S701, encode the supply chain feature set corresponding to the first customer to obtain a chromosome;

[0097] Chromosomes represent potential solutions, but in practical applications, they need to be decoded into executable operations. In a feature extraction task, a chromosome may encode the results of feature selection, the parameters of a network structure, or other information related to feature extraction. The decoding process is to convert these encodings into a format that the feature extraction model can understand and apply. In each iteration of the genetic algorithm, a new population of chromosomes is generated through selection, crossover, and mutation operations. Each newly generated chromosome will go through the above-mentioned feature extraction and evaluation processes to calculate its fitness. This process will continue until a predetermined number of iterations or other stopping criteria are reached. After the iteration ends, select the chromosome with the highest fitness from all the generated chromosomes. This chromosome represents the best feature extraction scheme found in the current search space.

[0098] S702. Input each of the said chromosomes into the pre-trained feature extraction model, and obtain the best chromosome corresponding to the optimal fitness when the number of iterations is reached;

[0099] S703. Decode the said best chromosome to obtain the supply chain features associated with the financing business.

[0100] The advantages of the genetic algorithm adopted by the present invention in training the classification model lie in its global search ability and the ability to explore complex search spaces.

[0101] By sorting and screening the financing business materials, reasonable, orderly supply chain features related to the financing business can be obtained, avoiding problems such as the risk information of a customer in different chains not being interconnected, resulting in different risk levels of a customer in different chains, and the customer's data being messy and disorderly. The supply chain features obtained after being processed by the present invention can ensure the accuracy of the risk identification results obtained by the subsequent financing risk identification model.

[0102] Use it as an input sample for the financing risk identification model to conduct supply chain financing risk identification.

[0103] In one embodiment, the financing risk identification model is generated through machine learning training based on the mapping relationship between the said account information and supply chain features and supply chain financing risk features. Refer to Figure 8 , and the specific training process is as follows:

[0104] S801. Obtain the supply chain features of the sample customers and the account information of the sample customers, and add risk labels to the supply chain features of the sample customers and the account information of the sample customers to obtain classification samples, where the risk labels include risk levels and the supply chain financing risk features corresponding to each risk level;

[0105] S802. Train the financing risk identification model according to the classification samples, where the financing risk identification model includes the mapping relationship between the account information and supply chain features and supply chain financing risk features.

[0106] Refer to Figure 9 , and the detailed process of inputting the account information and supply chain features into the financing risk identification model to determine the supply chain financing risk identification result is as follows:

[0107] S901. Input the supply chain features corresponding to the first customer, the account information of the first customer, and the account information of the second customer into the financing risk identification model to determine the classification sample with the highest correlation with the supply chain features corresponding to the first customer, the account information of the first customer, and the account information of the second customer;

[0108] S902. Determine the supply chain financing risk characteristics based on the risk label corresponding to the classification sample with the highest relevance, and use the supply chain financing risk characteristics as the supply chain financing risk identification result.

[0109] Taking enterprise customer A as an example, after obtaining the data set A1 of the supply chain characteristics and account information of enterprise customer A (closely related to the financing business), the financing risk identification model is also used to identify the supply chain characteristics and account information to obtain the financing risk result, where the financing risk result includes high risk or low risk. Before specific identification, the supply chain characteristics and account information of sample users need to be used for training to obtain the financing risk identification model.

[0110] The specific training process can be to add risk labels to the supply chain characteristics and account information of sample users. The risk labels can be different risk levels (such as high risk or low risk), so as to train the model according to the classified samples with added risk labels to obtain the financing risk identification model. The financing risk identification model includes the mapping relationship between the supply chain characteristics and account information and the classification label. After the data set A1 of the supply chain characteristics and account information of enterprise customer A is input into the financing risk identification model, the classification sample 2 with the highest similarity to the data set A2 of the supply chain characteristics and account information will be identified. The data set A2 of the supply chain characteristics and account information and the risk label A2 - high risk are included in the classification sample 2. At this time, it is determined that the data set A1 has the highest similarity to the data set A2. Since the data set A2 corresponds to high risk, it can be determined that the financing risk result corresponding to the data set A1 is high risk. The high risk and low risk mentioned in this embodiment can be compared with a set risk score or can be risk level identifiers set based on the business type.

[0111] In one embodiment, S107. According to the supply chain financing risk identification result, adopt the corresponding risk control strategy for risk control.

[0112] For the supply chain financing risk identification result, different risk control strategies are adopted to process the financing business. For example, if the risk is relatively high, the financing business is rejected or evaluated and analyzed by bank staff. If the risk is relatively low, it is determined that the financing business can be carried out, and the two customers (the first customer and the second customer) are notified by 5G message for online or offline signing processing, and the bank conducts business review. Refer to Figure 10 The detailed process of risk control is as follows:

[0113] S1001. According to the supply chain financing risk identification result, if the risk level corresponding to the supply chain financing risk characteristics is greater than the preset risk level threshold, or the supply chain financing risk characteristics belong to the scope of supply chain finance risk characteristics, initiate a risk alarm and transfer it to manual review;

[0114] S1002. If the risk level corresponding to the supply chain financing risk characteristics is less than or equal to the preset risk level threshold, and the supply chain financing risk characteristics do not belong to the scope of supply chain finance risk characteristics, it is determined that the risk analysis of this financing business passes. The first customer and the second customer are associated, and the banking business personnel are notified to assist in handling this financing business, and a 5G notification message is sent to the 5G mobile terminals of the first customer and the second customer through the 5G message processing node.

[0115] In the financing business scenario, different risk control strategies can be adopted to manage risks. Through risk assessment and monitoring, regularly assess and monitor the risk level of the financing project, timely discover and respond to potential risks, and effectively reduce the risk level of the financing business.

[0116] The following is illustrated with a specific application scenario. Refer to Figure 11 , which is a schematic flowchart of the supply chain financing risk processing based on 5G messages in a specific embodiment of the present invention. As Figure 11 shown, the customer initiates a financing business request. The bank identifies and analyzes the materials to obtain supply chain characteristics. The operator receives the customer information and relevant signing agreements, analyzes the customer's trusted mobile phone number, completes identity verification and the identification of the trusted mobile phone number. Further, the account information is queried through the trusted mobile phone number, and the account information and supply chain characteristics are analyzed using the financing risk identification model to obtain the risk identification result, and risk control is carried out based on this result.

[0117] In the actual application scenario, the present invention can adopt a cooperative approach with logistics, warehousing institutions, etc. to jointly monitor the cash flow and logistics status in the operation of enterprises, and explore financial technical means from it, thereby reducing financing risks. In terms of risk identification and management, mainly the following measures are included in view of the monitoring results: storing the inventories of upstream and downstream enterprises in the supply chain in the warehousing institutions designated by the bank and monitoring them in real time, so as to ensure that the credit risk is controllable; with the help of big data mining technology, that is, information such as the merchant's historical transaction records and order records, which greatly saves the time consumed in the approval process; the guarantee institution conducts on-site inspections of the office, business premises, and project sites of the borrowing enterprise, seriously analyzes the financial status of the enterprise, and signs a proxy purchase and repurchase agreement with the enterprise for new orders; judging whether the dealer's operation is normal through the information obtained from the brand manufacturer's dealer management platform, and discovering abnormal situations in advance; integrating the background ERP systems of large and medium-sized dealers to make the supervision of the business more timely and accurate; highly integrating and automating, so that the data provided by the dealer through the platform is real, and reasonably monitoring and risk analyzing the dealer's sales payment collection.

[0118] It should be noted that although the operations of the method of the present invention are described in a specific order in the above embodiments and accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution.

[0119] After introducing the method of the exemplary embodiment of the present invention, next, with reference to Figure 12 an apparatus for processing supply chain financing risks based on 5G messages of the exemplary embodiment of the present invention will be introduced.

[0120] The implementation of the apparatus for processing supply chain financing risks based on 5G messages can refer to the implementation of the above method, and the repeated parts will not be described again. The terms "module" or "unit" used hereinafter may be a combination of software and / or hardware that implements a predetermined function. Although the apparatuses described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0121] Based on the same inventive concept, the present invention also proposes an apparatus for processing supply chain financing risks based on 5G messages, as Figure 12 shown, the apparatus includes:

[0122] A request acquisition module 1210, configured to acquire a financing service request through a 5G message processing node; wherein, the financing service request is sent by a first customer through a 5G mobile terminal to the 5G message processing node;

[0123] An information acquisition module 1220, configured to acquire the identity information, financing service materials, and mobile phone number of the first customer, and the identity information and mobile phone number of the second customer according to the financing service request; wherein, the first customer is the financing requestor, and the second customer is the fund provider; the 5G message processing node determines whether the mobile phone number is a trusted mobile phone number based on a trusted mobile phone number database;

[0124] An identity verification module 1230, configured to, if the mobile phone number of the first customer or the second customer is an untrusted mobile phone number, verify the identity information of the customer through a customer resource library; after successful verification, determine that the mobile phone number is a trusted mobile phone number;

[0125] A batch sending module 1240, configured to batch send the newly added trusted mobile phone numbers to the trusted mobile phone number database of the 5G message processing node according to a preset time point;

[0126] An information query module 1250, which is used to query account information in the bank's back-end database according to the trusted mobile phone numbers of the first customer and the second customer if the mobile phone numbers of the first customer and the second customer are trusted mobile phone numbers and under the condition that the first customer and the second customer authorize the query;

[0127] A risk identification module 1260, which is used to determine the supply chain characteristics according to the financing business materials, input the account information and the supply chain characteristics into a financing risk identification model, and determine the supply chain financing risk identification result; wherein, the financing risk identification model is generated through machine learning training based on the mapping relationship between the account information and the supply chain characteristics and the supply chain financing risk characteristics;

[0128] A risk control module 1270, which is used to control risks by adopting corresponding risk control strategies according to the supply chain financing risk identification result.

[0129] In one embodiment, for the 5G message processing node, the processing flow for determining the trusted mobile phone number is as follows:

[0130] Construct a trusted mobile phone number database;

[0131] According to the customer's identity document and biometric features, conduct real-name authentication on the customer's mobile phone number. After the authentication is passed, determine that the customer's mobile phone number is a trusted mobile phone number; otherwise, determine that the customer's mobile phone number is an untrusted mobile phone number;

[0132] Add the trusted mobile phone number to the trusted mobile phone number database.

[0133] In one embodiment, the customer resource library is a peer bank database;

[0134] An identity verification module 1230 verifies the customer's identity information through the customer resource library, including:

[0135] Query whether there is customer information corresponding to the untrusted mobile phone number in the customer resource library;

[0136] If it exists, conduct an identity comparison on the customer according to the customer information corresponding to the untrusted mobile phone number queried. If the comparison is successful, determine that the identity verification is successful;

[0137] If it does not exist, call the identity authentication interface to conduct real-name authentication on the customer, and add the customer information with successful identity verification to the peer bank database.

[0138] In one embodiment, the risk identification module 1260 determines the supply chain characteristics according to the financing business materials, including:

[0139] Recognize the first customer's financing business materials using an OCR algorithm, and convert the recognition results into a structured financing business text; the financing business text includes basic registration data, credit data, and transaction data;

[0140] Preprocessing the financing business text content to obtain preprocessed supply chain financing information;

[0141] The pre-processed supply chain financing information is subjected to feature extraction using feature engineering to obtain the supply chain feature set;

[0142] The supply chain feature set is screened through a pre-trained feature extraction model to obtain supply chain features associated with the financing business.

[0143] In one embodiment, the risk identification module 1260 pre-processes the financing business text content to obtain pre-processed supply chain financing information, including:

[0144] Classify, clean or balance the text content of the financing business, eliminate erroneous data, and fill in accurate data.

[0145] In one embodiment, the feature extraction model is determined by the following method:

[0146] Obtaining supply chain characteristics of sample customers, where the supply chain characteristics of the sample customers are composed of a supply chain training set of the sample customers and a supply chain test set of the sample customers;

[0147] Using a first genetic algorithm to train the supply chain training set of the sample customers to obtain an initial model;

[0148] The initial model is tested using the supply chain test set of the sample customers, and the initial model that passes the test is used as the feature extraction model.

[0149] In one embodiment, the risk identification module 1260 filters the supply chain feature set using a pre-trained feature extraction model to obtain supply chain features associated with the financing business, including:

[0150] Encoding the supply chain feature set corresponding to the first customer to obtain a chromosome;

[0151] Inputting each of the chromosomes into the pre-trained feature extraction model, and obtaining the best chromosome corresponding to the optimal fitness when the number of iterations is reached;

[0152] The optimal chromosome is decoded to obtain the supply chain characteristics associated with the financing business.

[0153] In one embodiment, the financing risk identification model is determined by the following method:

[0154] Obtain the supply chain characteristics of sample customers and the account information of sample customers, and add risk labels to the supply chain characteristics of the sample customers and the account information of the sample customers to obtain classified samples, where the risk labels include risk levels and supply chain financing risk characteristics corresponding to each risk level;

[0155] Train the financing risk identification model according to the classified samples, where the financing risk identification model includes the mapping relationship between the account information, supply chain characteristics and supply chain financing risk characteristics.

[0156] In one embodiment, the risk identification module 1260 inputs the account information and supply chain characteristics into the financing risk identification model to determine the supply chain financing risk identification result, including:

[0157] Input the supply chain characteristics corresponding to the first customer, the account information of the first customer and the account information of the second customer into the financing risk identification model to determine the classified sample with the highest correlation with the supply chain characteristics corresponding to the first customer, the account information of the first customer and the account information of the second customer;

[0158] Determine the supply chain financing risk characteristics according to the risk labels corresponding to the classified sample with the highest correlation, and use the supply chain financing risk characteristics as the supply chain financing risk identification result.

[0159] In one embodiment, the risk control module 1270 performs risk control according to the supply chain financing risk identification result, including:

[0160] According to the supply chain financing risk identification result, if the risk level corresponding to the supply chain financing risk characteristics is greater than the preset risk level threshold, or the supply chain financing risk characteristics belong to the scope of supply chain finance risk characteristics, initiate a risk warning and transfer it to manual review;

[0161] If the risk level corresponding to the supply chain financing risk characteristics is less than or equal to the preset risk level threshold, and the supply chain financing risk characteristics do not belong to the scope of supply chain finance risk characteristics, determine that the risk analysis of this financing business passes, associate the first customer with the second customer, notify the banking business personnel to assist in handling this financing business, and send a 5G notification message to the 5G mobile terminals of the first customer and the second customer through the 5G message processing node.

[0162] It should be noted that although several modules of the 5G message-based supply chain financing risk processing device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present invention, the features and functions of two or more of the above-described modules can be embodied in one module. Conversely, the features and functions of one module described above can be further divided and embodied by multiple modules.

[0163] Based on the foregoing inventive concept, as Figure 13 shown, the present invention also provides a computer device 1300, including a memory 1310, a processor 1320, and a computer program 1330 stored in the memory 1310 and executable on the processor 1320. When the processor 1320 executes the computer program 1330, the foregoing 5G message-based supply chain financing risk processing method is implemented.

[0164] Based on the foregoing inventive concept, the present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the foregoing 5G message-based supply chain financing risk processing method is implemented.

[0165] Based on the foregoing inventive concept, the present invention provides a computer program product including a computer program, and when the computer program is executed by a processor, the 5G message-based supply chain financing risk processing method is implemented.

[0166] The supply chain financing risk handling method and device proposed by the present invention obtain a financing business request through a 5G message processing node; wherein, the financing business request is sent by a first customer to the 5G message processing node through a 5G mobile terminal; obtain the identity information, financing business materials and mobile phone number of the first customer, as well as the identity information and mobile phone number of the second customer according to the financing business request; wherein, the first customer is the financing requester and the second customer is the fund provider; the 5G message processing node determines whether the mobile phone number is a trusted mobile phone number based on a trusted mobile phone number database; if the mobile phone number of the first customer or the second customer is an untrusted mobile phone number, the identity information of the customer is authenticated through a customer resource library; after successful authentication, determine that the mobile phone number is a trusted mobile phone number; batch send the newly added trusted mobile phone numbers to the trusted mobile phone number database of the 5G message processing node according to a preset time point; if the mobile phone numbers of the first customer and the second customer are trusted mobile phone numbers, under the condition of authorization query by the first customer and the second customer, query account information in the bank background database according to the trusted mobile phone number of the first customer and the trusted mobile phone number of the second customer; determine the supply chain characteristics according to the financing business materials, input the account information and the supply chain characteristics into a financing risk identification model, and determine the supply chain financing risk identification result; wherein, the financing risk identification model is generated through machine learning training based on the mapping relationship between the account information and the supply chain characteristics and the supply chain financing risk characteristics; according to the supply chain financing risk identification result, adopt corresponding risk control strategies for risk control. The overall solution enables users to upload materials through the terminal using 5G messages, realizes the rapid transmission of messages and data between customers and banks, judges trusted mobile phone numbers based on 5G message nodes, provides technical support for customer identity recognition, and improves the efficiency of subsequent identity verification by financial institutions; sorts out the financing business materials to obtain reasonable, orderly, and financing business-related supply chain characteristics; queries account information based on the trusted mobile phone numbers of customers, uses supply chain characteristics and customer information as samples for risk identification, and conducts risk identification through a financing risk identification model, thereby effectively controlling supply chain financing risks. The financing risk identification model uses multi-dimensional data for mapping training, which can effectively improve the accuracy of model identification and the evaluation effect. The present invention uses 5G messages to achieve rapid data transmission of information, realizes identity verification by the operator's judgment of trusted mobile phone numbers, improves the identity verification method in the financial scenario, and improves the identity verification efficiency; adopts a financing risk identification model and multi-dimensional data for risk identification, effectively controls supply chain financing risks, provides strong guarantees for enterprise development and the property safety of fund providers, and provides strong technical support for the financial technology scenario.

[0167] In the technical solution of the present invention, the acquisition, storage, use, processing, etc. of data all comply with the relevant regulations of laws and regulations.

[0168] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, an apparatus, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0169] The present invention is described with reference to the flowcharts and / or block diagrams of methods and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or the combination of multiple flows and / or blocks.

[0170] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or the combination of multiple flows and / or blocks.

[0171] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or the combination of multiple flows and / or blocks.

[0172] Finally, it should be noted that the above-described embodiments are only specific embodiments of the present invention, which are used to illustrate the technical solutions of the present invention, rather than limiting it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions recorded in the foregoing embodiments, or can easily think of changes, or make equivalent replacements for some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for processing supply chain financing risks based on 5G messaging, characterized in that, The method includes: Obtaining a financing business request through a 5G message processing node; wherein, the financing business request is sent by a first customer to the 5G message processing node through a 5G mobile terminal; Obtaining the identity information, financing business materials and mobile phone number of the first customer, as well as the identity information and mobile phone number of the second customer according to the financing business request; wherein, the first customer is the financing requestor and the second customer is the fund provider; the 5G message processing node determines whether the mobile phone number is a trustworthy mobile phone number based on a trustworthy mobile phone number database; If the mobile phone number of the first customer or the second customer is an untrustworthy mobile phone number, authenticate the identity information of the customer through a customer resource library; after successful authentication, determine that the mobile phone number is a trustworthy mobile phone number; Batch send the newly added trustworthy mobile phone numbers to the trustworthy mobile phone number database of the 5G message processing node according to a preset time point; If the mobile phone numbers of the first customer and the second customer are trustworthy mobile phone numbers, query the account information in the bank back-end database according to the trustworthy mobile phone number of the first customer and the trustworthy mobile phone number of the second customer under the authorization of the first customer and the second customer to query; Determine the supply chain characteristics according to the financing business materials, input the account information and supply chain characteristics into a financing risk identification model, and determine the supply chain financing risk identification result; wherein, the financing risk identification model is generated through machine learning training based on the mapping relationship between the account information and supply chain characteristics and the supply chain financing risk characteristics; According to the supply chain financing risk identification result, adopt corresponding risk control strategies for risk control.

2. The method according to claim 1, wherein The method further includes: Constructing a trustworthy mobile phone number database; Authenticate the real-name identity of the customer's mobile phone number according to the customer's identity document and biometric characteristics. After successful authentication, determine that the customer's mobile phone number is a trustworthy mobile phone number, otherwise determine that the customer's mobile phone number is an untrustworthy mobile phone number; Add the trustworthy mobile phone number to the trustworthy mobile phone number database.

3. The method according to claim 1, wherein The customer resource library is a peer bank database; Authenticating the identity information of the customer through the customer resource library includes: Querying whether there is customer information corresponding to the untrustworthy mobile phone number in the customer resource library; If it exists, compare the identity of the customer according to the customer information corresponding to the untrustworthy mobile phone number queried. If the comparison is successful, determine that the identity authentication is successful; If it does not exist, call the identity authentication interface to authenticate the real-name identity of the customer, and add the customer information with successful identity authentication to the peer bank database.

4. The method according to claim 1, characterized in that, Determining the supply chain characteristics according to the financing business materials includes: Identifying the financing business materials of the first customer through an OCR algorithm, and structurally converting the identification result to obtain the financing business text content; the financing business text content includes registration basic data, credit investigation data and transaction data; Preprocess the financing business text content to obtain preprocessed supply chain financing information; Use feature engineering to extract features from the preprocessed supply chain financing information to obtain the supply chain feature set; Screen the supply chain feature set through a pre-trained feature extraction model to obtain supply chain features associated with the financing business.

5. The method according to claim 1, wherein Preprocess the content of the financing business text to obtain preprocessed supply chain financing information, including: Classify, clean, or balance the dataset of the financing business text content, remove incorrect data, and fill in missing data.

6. The method according to claim 4, characterized in that, The feature extraction model is determined by the following method: Obtain the supply chain features of sample customers, where the supply chain features of sample customers consist of a supply chain training set and a supply chain test set of sample customers; Use a first genetic algorithm to train the supply chain training set of the sample customers to obtain an initial model; Use the supply chain test set of the sample customers to test the initial model, and use the initial model that passes the test as the feature extraction model.

7. The method according to claim 6, wherein Screen the supply chain feature set through a pre-trained feature extraction model to obtain supply chain features associated with the financing business, including: Encode the supply chain feature set corresponding to the first customer to obtain chromosomes; Input each of the chromosomes into the pre-trained feature extraction model, and obtain the best chromosome corresponding to the optimal fitness when the iteration count is reached; Decode the best chromosome to obtain the supply chain features associated with the financing business.

8. The method according to claim 1, characterized in that The financing risk identification model is determined by the following method: Obtain the supply chain features of sample customers and the account information of sample customers, and add risk labels to the supply chain features of sample customers and the account information of sample customers to obtain classified samples, where the risk labels include risk levels and supply chain financing risk features corresponding to each risk level; Train the financing risk identification model according to the classified samples, where the financing risk identification model contains the mapping relationship between the account information and supply chain features and the supply chain financing risk features.

9. The method according to claim 8, wherein Input the account information and supply chain features into the financing risk identification model to determine the supply chain financing risk identification result, including: Input the supply chain features corresponding to the first customer, the account information of the first customer, and the account information of the second customer into the financing risk identification model to determine the classified sample with the highest correlation with the supply chain features corresponding to the first customer, the account information of the first customer, and the account information of the second customer; Determine the supply chain financing risk features according to the risk labels corresponding to the classified sample with the highest correlation, and use the supply chain financing risk features as the supply chain financing risk identification result.

10. The method according to claim 9, characterized in that, According to the supply chain financing risk identification result, adopt corresponding risk control strategies for risk control, including: According to the supply chain financing risk identification result, if the risk level corresponding to the supply chain financing risk features is greater than the preset risk level threshold, or the supply chain financing risk features belong to the scope of supply chain finance risk features, initiate a risk warning and transfer it for manual review; If the risk level corresponding to the supply chain financing risk characteristics is less than or equal to the preset risk level threshold, and the supply chain financing risk characteristics do not belong to the scope of supply chain finance risk characteristics, it is determined that the risk analysis of this financing business passes. The first customer and the second customer are associated, and the banking business personnel are notified to assist in handling this financing business. A 5G notification message is sent to the 5G mobile terminals of the first customer and the second customer through the 5G message processing node.

11. A supply chain financing risk handling device based on 5G messaging, characterized in that The device includes: A request acquisition module, configured to acquire a financing business request through a 5G message processing node; wherein, the financing business request is sent by a first customer to the 5G message processing node through a 5G mobile terminal; An information acquisition module, configured to acquire the identity information, financing business materials, and mobile phone number of the first customer, as well as the identity information and mobile phone number of the second customer according to the financing business request; wherein, the first customer is the financing requester, and the second customer is the fund provider; the 5G message processing node determines whether the mobile phone number is a trusted mobile phone number based on the trusted mobile phone number database; An identity verification module, configured to, if the mobile phone number of the first customer or the second customer is an untrusted mobile phone number, verify the identity information of the customer through the customer resource library; after successful verification, determine that the mobile phone number is a trusted mobile phone number; A batch sending module, configured to batch send the newly added trusted mobile phone numbers to the trusted mobile phone number database of the 5G message processing node according to a preset time point; An information query module, configured to, if the mobile phone numbers of the first customer and the second customer are trusted mobile phone numbers, query the account information in the bank back-end database according to the trusted mobile phone number of the first customer and the trusted mobile phone number of the second customer under the authorization of the first customer and the second customer; A risk identification module, configured to determine the supply chain characteristics according to the financing business materials, input the account information and the supply chain characteristics into a financing risk identification model, and determine the supply chain financing risk identification result; wherein, the financing risk identification model is generated through machine learning training based on the mapping relationship between the account information and the supply chain characteristics and the supply chain financing risk characteristics; A risk control module, configured to perform risk control by adopting corresponding risk control strategies according to the supply chain financing risk identification result.

12. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the method according to any one of claims 1 to 10.

14. A computer program product, characterized in that, The computer program product includes a computer program, and when the computer program is executed by the processor, it implements the method according to any one of claims 1 to 10.

Images