Electronic warrant full-process management and risk management and control system based on block chain

Abstract

The invention relates to the technical field of electronic warrants, and particularly discloses an electronic warrant full-process management and risk management and control system based on a block chain, and the system comprises the steps: calling a user credit record and a user insurance record according to the identity information of a user; and data splitting and semantic coding are carried out by using a natural language analysis and processing algorithm based on deep learning, causal association semantic dynamic migration is carried out on coded record features, and credit risk multi-mode joint representation is generated, so that automatic judgment on whether an insurance request is within a credit risk security level is realized. In this way, complex semantic information in the user credit record and the user insurance record can be effectively analyzed and utilized, and the credit risk judgment basis is more comprehensive and accurate. Meanwhile, the model can be continuously updated and adjusted according to new data and is not limited by a fixed formula and a preset coefficient factor any more, and the accuracy and effectiveness of risk judgment in different scenes are improved.

Description

Technical Field

[0001] The present application relates to the technical field of electronic letters of guarantee, and more specifically, to a blockchain-based full-process management and risk control system for electronic letters of guarantee. Background Art

[0002] An electronic letter of guarantee is an electronic guarantee certificate, which is issued by the guarantor to the beneficiary at the request of the insured, and is used to fulfill the payment or compensation liability when the insured defaults. However, there are many problems with the existing system: the lack of effective credit investigation and risk prevention measures makes it difficult to deal with counterfeiting and credit risks in the supply chain; the guarantor needs to make additional investments to verify the authenticity of the transaction, which increases the cost of risk control, and the data information is of low credibility and easy to tamper with.

[0003] Based on this, the existing patent CN114155093A proposes a blockchain-based electronic letter of guarantee full-process management and risk control system, which uses blockchain technology to build an electronic letter of guarantee management platform. Users submit insurance requests through smart contracts, and the platform receives and processes the requests, verifies the user's identity and credit status, and calculates the credit score (XY) to determine the risk level. It then conducts risk factor analysis and reviews and approves it. Finally, the issued electronic letter of guarantee is encrypted and distributed to the designated storage address to effectively prevent the electronic letter of guarantee from being tampered with and improve data security.

[0004] In this patent, the determination of whether the insurance request is within the credit risk safety level is achieved by calculating the credit score based on the overdue data and default data, and then comparing it with the preset threshold. However, this method only relies on numerical data such as the number of overdue times and amounts, and is unable to parse and utilize the complex semantic information in the text data, which may cause important background and context factors to be ignored when making judgments. Moreover, this method uses fixed formulas and preset coefficient factors. These parameters are generally set based on historical data, which makes it difficult to adapt to new market changes or special circumstances, and lacks flexibility and dynamic adjustment capabilities, thus affecting the accuracy and effectiveness of the judgment.

[0005] Therefore, an optimized blockchain-based electronic letter of guarantee full-process management and risk control solution is desired. Summary of the invention

[0006] This application provides a blockchain-based electronic guarantee full-process management and risk control system, which can effectively parse and utilize the complex semantic information in the user's credit record and the user's insurance record, making the basis for credit risk judgment more comprehensive and accurate. At the same time, the model can be continuously updated and adjusted according to new data, no longer constrained by fixed formulas and preset coefficient factors, and improve the accuracy and effectiveness of risk judgment in different scenarios.

[0007] According to one aspect of the present application, a blockchain-based electronic guarantee full-process management and risk control system is provided, including: a platform construction module, used to build an electronic guarantee management platform using blockchain technology; a request module, used to initiate an insurance request to the electronic guarantee management platform; a verification module, used to verify the user identity in response to the electronic guarantee management platform receiving the insurance request to determine whether the insurance request is within the credit risk security level; a guarantee letter distribution module, used to approve the insurance request in response to the insurance request being within the credit risk security level, and issue an electronic guarantee for the approved insurance request and encrypt and store it; Among them, the verification module includes: a retrieval unit, which is used to retrieve the user credit record and the user insurance record according to the user identity information; a record splitting and encoding unit, which is used to perform data splitting and semantic encoding on the user credit record and the user insurance record to obtain the time series distribution of the user credit record semantic embedding coding vector and the time series distribution of the user insurance record semantic embedding coding vector; a record temporal semantic walking unit, which is used to perform causal association-driven record semantic dynamic walking on the time series distribution of the user credit record semantic embedding coding vector and the time series distribution of the user insurance record semantic embedding coding vector to obtain the user credit record temporal semantic walking coding vector and the user insurance record temporal semantic walking coding vector; a record multimodal joint unit, which is used to fuse the user credit record temporal semantic walking coding vector and the user insurance record temporal semantic walking coding vector to obtain the user credit risk multimodal joint coding vector; a credit risk judgment unit, which is used to obtain an identification result indicating whether the insurance request is within the credit risk security level based on the user credit risk multimodal joint coding vector.

[0008] In the above-mentioned blockchain-based electronic letter of guarantee full-process management and risk control system, the record splitting coding unit includes: A data splitting subunit, used to split the user credit record and the user insurance record according to the time dimension to obtain a list of user credit record items and a list of user insurance items; A user credit record semantic encoding subunit, configured to semantically encode each user credit record item in the list of user credit record items to obtain a time series distribution of the user credit record semantic embedding encoding vector; The user insurance record semantic coding subunit is used to semantically code each user insurance item in the list of user insurance items to obtain a time series distribution of the semantic embedded coding vector of the user insurance record.

[0009] In the above-mentioned blockchain-based electronic letter of guarantee full-process management and risk control system, the user credit record semantic encoding subunit is used to: use a semantic encoder including RNN to semantically encode each user credit record item in the list of user credit record items to obtain a time series distribution of the user credit record semantic embedded coding vector.

[0010] In the above-mentioned blockchain-based electronic letter of guarantee full-process management and risk control system, the user insurance record semantic coding subunit is used to: use the semantic encoder containing RNN to semantically encode each user insurance item in the list of user insurance items to obtain the time series distribution of the semantic embedded coding vector of the user insurance record.

[0011] In the above-mentioned blockchain-based electronic letter of guarantee full-process management and risk control system, the record time sequence semantic walking unit includes: A user credit record semantic implicit feature mining subunit is used to perform implicit feature mining on each user credit record semantic embedding coding vector in the time series distribution of the user credit record semantic embedding coding vector to obtain a time series distribution of the user credit record semantic embedding depth implicit feature coding vector; A user credit record semantic causal feature construction subunit, for constructing a user credit record semantic causal association topological feature matrix based on the time series distribution of the user credit record semantic embedding deep implicit feature coding vector; A user credit record semantic surface context dynamic walking subunit, used to perform dynamic graph convolution-driven feature dynamic walking on the user credit record semantic causal association topological feature matrix and the time series distribution of the user credit record semantic embedding coding vector to obtain a user credit record semantic surface context dynamic walking semantic coding vector; A user credit record semantic hidden layer feature extraction subunit is used to perform the dynamic graph convolution-driven feature dynamic walk on the user credit record semantic causal association topological feature matrix and the time series distribution of the user credit record semantic embedding depth implicit feature coding vector to obtain the user credit record semantic hidden layer context dynamic walk semantic coding vector; The user credit record feature fusion subunit is used to fuse the user credit record semantic surface context dynamic wandering semantic coding vector and the user credit record semantic hidden context dynamic wandering semantic coding vector to obtain the user credit record time series semantic wandering coding vector.

[0012] In the above-mentioned blockchain-based electronic letter of guarantee full-process management and risk control system, the user credit record semantic causal feature construction subunit includes: A user credit record semantic causal factor calculation secondary subunit is used to calculate the semantic causal association factor between any two user credit record semantic embedding depth implicit feature coding vectors in the time series distribution of the user credit record semantic embedding depth implicit feature coding vector to obtain a user credit record semantic causal association topological matrix composed of multiple user credit record semantic causal association factors; The user credit record semantic causal triggering secondary sub-unit is used to perform causal triggering of the gated activation function on the user credit record semantic causal association topology matrix to obtain the user credit record semantic causal association topology feature matrix.

[0013] In the above-mentioned blockchain-based electronic letter of guarantee full-process management and risk control system, the user credit record semantic causal factor calculation secondary sub-unit includes: The semantic association matrix calculation tertiary subunit is used to calculate the association matrix between any two user credit record semantic embedding deep implicit feature encoding vectors to obtain the user credit record semantic association matrix; The third-level sub-unit for causal factor calculation based on statistical features is used to perform causal factor calculation based on statistical features on the user credit record semantic association matrix to obtain the user credit record semantic causal association factor corresponding to the user credit record semantic association matrix, wherein the user credit record semantic causal association factor is related to the mean, variance, maximum value and causal bias value of the user credit record semantic association matrix.

[0014] In the above-mentioned blockchain-based electronic letter of guarantee full-process management and risk control system, the three-level sub-unit for causal factor calculation based on statistical characteristics is used to: In response to the variance of the user credit record semantic association matrix being greater than or equal to a predetermined threshold, taking the weighted average of the distances between any two user credit record semantic embedding deep implicit feature encoding vectors in the time series distribution of the user credit record semantic embedding deep implicit feature encoding vectors as the causal bias value; In response to the variance of the user credit record semantic association matrix being less than the predetermined threshold, a weighted value of the mean of the user credit record semantic association matrix is ​​used as the causal bias value.

[0015] In the above-mentioned blockchain-based electronic letter of guarantee full-process management and risk control system, the credit risk judgment unit is used to: input the user credit risk multimodal joint encoding vector into the classifier-based credit risk identifier to obtain the identification result, and the identification result is used to indicate whether the insurance request is within the credit risk safety level.

[0016] The present application provides a blockchain-based electronic letter of guarantee full-process management and risk control system, which retrieves user credit records and user insurance records according to user identity information, uses natural language analysis and processing algorithms based on deep learning to perform data splitting and semantic encoding on the user credit records and the user insurance records, and then performs causal semantic dynamic walking on the encoded semantic embeddings of each user credit record and each user insurance record semantic embedding feature, so as to realize automatic judgment of whether the insurance request is within the credit risk security level based on the credit risk multimodal joint representation between the temporal semantic walking features of the user credit record and the temporal semantic walking features of the user insurance record obtained after association. This application can effectively parse and utilize the complex semantic information in the user credit record and the user insurance record, making the basis for credit risk judgment more comprehensive and accurate. At the same time, the model can be continuously updated and adjusted according to new data, no longer constrained by fixed formulas and preset coefficient factors, and improve the accuracy and effectiveness of risk judgment in different scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0018] Figure 1 This is a schematic block diagram of the blockchain-based full-process management and risk control system of electronic letters of guarantee in an embodiment of the present application.

[0019] Figure 2 This is a schematic block diagram of the verification module in the blockchain-based electronic letter of guarantee full-process management and risk control system in an embodiment of the present application.

[0020] Figure 3 This is a schematic diagram of the data flow of the verification module in the blockchain-based electronic letter of guarantee full-process management and risk control system in an embodiment of the present application.

[0021] Figure 4 This is a schematic block diagram of a record splitting coding unit in the blockchain-based electronic letter of guarantee full-process management and risk control system of an embodiment of the present application.

[0022] Figure 5 This is a schematic block diagram of a unit for recording time-series semantic walks in the blockchain-based electronic letter of guarantee full-process management and risk control system of an embodiment of the present application.

[0023] Figure 6 This is a schematic block diagram of a semantic causal feature construction subunit for user credit records in the blockchain-based electronic letter of guarantee full-process management and risk control system in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0025] Based on the existing patent CN114155093A, this application proposes a blockchain-based electronic letter of guarantee full-process management and risk control system. Figure 1 This is a schematic block diagram of the blockchain-based electronic guarantee full-process management and risk control system of the embodiment of this application. Figure 1 As shown, the blockchain-based electronic letter of guarantee full-process management and risk control system includes: a platform construction module 10, which is used to build an electronic letter of guarantee management platform using blockchain technology; a request module 20, which is used to initiate an insurance request to the electronic letter of guarantee management platform; a verification module 30, which is used to verify the user identity in response to the electronic letter of guarantee management platform receiving the insurance request to determine whether the insurance request is within the credit risk safety level; a letter of guarantee signature distribution module 40, which is used to approve the insurance request in response to the insurance request being within the credit risk safety level, and issue an electronic letter of guarantee for the approved insurance request and encrypt and store it.

[0026] Exemplarily, in the platform construction module 10, it is used to construct an electronic letter of guarantee management platform using blockchain technology. It should be understood that the reason for constructing an electronic letter of guarantee management platform using blockchain technology is to solve the problems of imperfect credit investigation, insufficient risk control and easy data tampering in the existing electronic letter of guarantee system. The traditional electronic letter of guarantee system lacks an effective credit investigation system and perfect risk prevention measures, and it is difficult to deal with counterfeit and shoddy products in the supply chain and the credit risks caused by them. In addition, banks, insurance companies or other guarantee institutions need to invest a lot of manpower and material resources to verify the authenticity of transactions, which increases additional risk control costs. At the same time, the existing electronic letter of guarantee data information has low credibility and is easy to be tampered with, which further increases the difficulty of risk management. By using blockchain technology to construct an electronic letter of guarantee management platform, the security and reliability of the system can be effectively improved. The core advantage of blockchain technology lies in its distributed ledger characteristics, which ensures the immutability and transparency of data.

[0027] Exemplarily, in the request module 20, it is used to initiate an insurance request to the electronic letter of guarantee management platform. It should be understood that the reason for initiating an insurance request to the electronic letter of guarantee management platform is that such a request is a necessary step to obtain the electronic letter of guarantee issued by the guarantee agency to the beneficiary. As an electronic guarantee certificate, the electronic letter of guarantee is issued to the beneficiary by a bank, insurance company, guarantee company or other guarantor at the request of the insured, and is used for the guarantor to perform a certain payment or economic compensation liability within a certain amount and a certain time limit on behalf of the applicant when the applicant fails to perform its responsibilities or obligations according to the agreement between the two parties. Therefore, the insurance request is not only a key action to start the entire process, but also an important link to ensure the security and reliability of the transaction. Specifically, when a user wants to obtain an electronic letter of guarantee through a guarantee agency, he needs to submit an insurance request to the electronic letter of guarantee management platform. This request contains detailed performance resource information, performance date, transferor information, guarantee agency information, and beneficiary information. These information together constitute a complete insurance request, which provides necessary data support for subsequent risk assessment and review. In order to initiate an insurance request to the electronic letter of guarantee management platform, the user first needs to interact with the platform through a blockchain smart contract. A blockchain smart contract is an automatically executed contract agreement that is written directly on the blockchain and automatically triggers corresponding operations based on preset conditions. Users send insurance requests through blockchain smart contracts, which means that all relevant information will be recorded on the blockchain, ensuring the authenticity and immutability of the information.

[0028] Exemplarily, in the verification module 30, in response to the electronic guarantee management platform receiving the insurance request, the user identity is verified to determine whether the insurance request is within the credit risk security level. It should be understood that the reason for verifying the user identity to determine whether the insurance request is within the credit risk security level in response to the electronic guarantee management platform receiving the insurance request is to ensure the security and reliability of the entire electronic guarantee system. Specifically, as an important financial tool, the electronic guarantee is used to protect the rights and interests of both parties to the transaction, especially to provide financial compensation when one party fails to perform its contractual obligations. Therefore, it is very important to accurately assess the credit risk of the insured. First, by verifying the user's identity, the authenticity of the insurance request can be confirmed to prevent the occurrence of fraud. Since the amount of funds involved in the electronic guarantee is usually large and related to the interests of multiple parties, it is necessary to ensure that the identities of all participants are authentic and reliable. If the user's identity cannot be effectively verified, false insurance requests may enter the system, thereby bringing unnecessary risks and losses to the guarantee agency.

[0029] Exemplarily, in the letter of guarantee signature distribution module 40, in response to the insurance request being within the credit risk security level, the insurance request is approved, and an electronic letter of guarantee is issued and encrypted for the insurance request after approval. It should be understood that the electronic letter of guarantee, as an electronic guarantee certificate, is issued to the beneficiary by a bank, insurance company, guarantee company or other guarantor at the request of the insured, and is used for the guarantor to perform a certain payment or economic compensation liability within a certain amount and a certain time limit on behalf of the applicant when the applicant fails to perform his or her responsibilities or obligations as agreed by both parties. Therefore, the insurance request is not only a key action to start the entire process, but also an important link to ensure the security and reliability of the transaction. By submitting an insurance request, the user can clarify the guarantee service he or she needs and provide necessary information for subsequent risk assessment and review. Once the electronic letter of guarantee management platform receives the insurance request and confirms that the request is within the credit risk security level, the system will approve the insurance request. Specifically, when the user's insurance request is considered to be within the credit risk security level, it indicates that his or her credit status is good and has a low risk of default. At this point, the system will enter the next approval process. First, the letter of guarantee issuance module will conduct a detailed risk factor analysis on the insurance request received, including but not limited to waiting time, access data and other multi-dimensional information. These analysis results will help determine whether to approve the insurance request. For the insurance request that passes the review, the system will issue the corresponding electronic letter of guarantee. The issuance process not only involves generating the electronic letter of guarantee itself, but also needs to ensure its security. To this end, the system uses encryption algorithms to encrypt the electronic letter of guarantee to prevent unauthorized access and tampering. The encrypted electronic letter of guarantee will be stored in a designated secure storage address to ensure that only authorized parties can access and use these files.

[0030] Accordingly, in the verification module, the technical concept of the present application is to retrieve the user credit record and the user insurance record according to the user identity information, and use the natural language analysis and processing algorithm based on deep learning to perform data splitting and semantic encoding on the user credit record and the user insurance record. Then, the encoded semantic embedding of each user credit record and the semantic embedding features of each user insurance record are respectively subjected to causal association semantic dynamic walking, so as to realize automatic judgment of whether the insurance request is within the credit risk security level according to the multimodal joint representation of credit risk between the temporal semantic walking features of the user credit record and the temporal semantic walking features of the user insurance record obtained after the association. The present application can effectively parse and utilize the complex semantic information in the user credit record and the user insurance record, making the basis for credit risk judgment more comprehensive and accurate. At the same time, the model can be continuously updated and adjusted according to new data, no longer constrained by fixed formulas and preset coefficient factors, and improve the accuracy and effectiveness of risk judgment in different scenarios.

[0031] In one embodiment, Figure 2 and Figure 3 As shown, the verification module 30 includes: a retrieval unit 31, which is used to retrieve the user credit record and the user insurance record according to the user identity information; a record splitting and encoding unit 32, which is used to perform data splitting and semantic encoding on the user credit record and the user insurance record to obtain the time series distribution of the user credit record semantic embedding coding vector and the time series distribution of the user insurance record semantic embedding coding vector; a record time series semantic walking unit 33, which is used to perform causal association driven record semantic dynamic walking on the time series distribution of the user credit record semantic embedding coding vector and the time series distribution of the user insurance record semantic embedding coding vector to obtain the user credit record time series semantic walking coding vector and the user insurance record time series semantic walking coding vector; a record multimodal joint unit 34, which is used to fuse the user credit record time series semantic walking coding vector and the user insurance record time series semantic walking coding vector to obtain the user credit risk multimodal joint coding vector; a credit risk judgment unit 35, which is used to obtain an identification result indicating whether the insurance request is within the credit risk security level according to the user credit risk multimodal joint coding vector.

[0032] Exemplarily, in the retrieval unit 31, the user's credit record and the user's insurance record are retrieved according to the user's identity information. It should be understood that when the user submits an insurance request to the electronic letter of guarantee management platform through a blockchain smart contract, the system will automatically obtain and verify the user's identity information. This process usually includes basic information such as the user's name, ID number, contact information, etc., and may be combined with biometric technology such as fingerprint or facial recognition to further confirm the user's identity. Once the user's identity is confirmed, the system will interact with external credit agencies and internal databases based on these identity information to retrieve the user's credit record and insurance record. Specifically, the system will establish a connection with a credit assessment agency (such as the central bank's credit reporting center or other third-party credit reporting service providers) through a secure data interface and send a query request containing the user's identity information. These credit agencies have detailed user credit history data, including repayment records, default situations, credit scores and other information. Through standardized data exchange protocols (such as APIs), the system can quickly obtain the latest credit records. At the same time, the system will also access the user's insurance records stored internally. These records usually include all the user's past insurance behaviors, such as the time, amount, guarantor, and whether claims have occurred. In order to ensure the consistency and integrity of the data, the system may use distributed databases or blockchain technology to store these records, which not only improves the security of the data, but also ensures that the data cannot be tampered with.

[0033] In one embodiment, Figure 4As shown, the record splitting and encoding unit 32 includes: a data splitting subunit 321, which is used to perform data splitting on the user credit record and the user insurance record according to the time dimension to obtain a list of user credit record items and a list of user insurance items; a user credit record semantic encoding subunit 322, which is used to perform semantic encoding on each user credit record item in the list of user credit record items to obtain a time series distribution of the user credit record semantic embedding encoding vector; a user insurance record semantic encoding subunit 323, which is used to perform semantic encoding on each user insurance item in the list of user insurance items to obtain a time series distribution of the user insurance record semantic embedding encoding vector.

[0034] Exemplarily, in the data splitting subunit 321, it is considered that the user's credit investigation and insurance behavior are essentially processes that change dynamically over time. For example, the repayment behavior in the credit record, the time of occurrence of the credit business, and the insurance time and claim settlement time in the insurance record all have an obvious time sequence. Therefore, in order to be able to understand and analyze the inherent characteristics between the data more accurately and clearly, in the technical solution of the present application, the user credit record and the user insurance record are data split according to the time dimension to obtain a list of user credit record items and a list of user insurance items. In this way, the changes in the user's credit behavior in different periods can be clearly observed. For example, checking the fluctuations of the user's overdue repayment situation over time and judging whether their credit status is gradually improving or deteriorating will help predict future credit risk trends.

[0035] Exemplarily, in the user credit record semantic encoding subunit 322, considering that each user credit record item contains its own semantic information and meaning, therefore, in the technical solution of the present application, each user credit record item in the list of user credit record items is semantically encoded to capture the deep-level features in the text, and obtain the time series distribution of the semantic embedding encoding vector of the user credit record. In particular, in a specific example of the present application, the user credit record semantic encoding subunit is used to: use a semantic encoder containing RNN to semantically encode each user credit record item in the list of user credit record items to obtain the time series distribution of the semantic embedding encoding vector of the user credit record. By using a semantic encoder containing RNN to semantically encode each user credit record item in the list of user credit record items to obtain the time series distribution of the semantic embedding encoding vector of the user credit record. It should be understood that the meaning of each credit record item often depends on the record items before and after it, that is, the context information. When processing a sequence, RNN can pass the information of the previous time step to the current time step, so that the model can understand the semantics of each record item based on the context and retain the characteristics of the time series, thereby more accurately representing the semantic information of each record item.

[0036] Exemplarily, in the user insurance record semantic coding subunit 323, considering that each user insurance item expresses insurance information at different time points, each information contains semantic meanings between contexts. Based on this, in the technical solution of the present application, each user insurance item in the list of user insurance items is semantically encoded to obtain a time series distribution of the user insurance record semantic embedding coding vector. In particular, in an example of the present application, the user insurance record semantic coding subunit is used to: use the semantic encoder containing RNN to semantically encode each user insurance item in the list of user insurance items to obtain a time series distribution of the user insurance record semantic embedding coding vector. It is also possible to semantically encode each user insurance item in the list of user insurance items by using the semantic encoder containing RNN to accurately capture the contextual semantic information of each insurance item in the entire sequence, thereby obtaining a time series distribution of the user insurance record semantic embedding coding vector.

[0037] Exemplarily, in the record time-series semantic walking unit 33, the time series distribution of the semantic embedding coding vector of the user credit record and the time series distribution of the semantic embedding coding vector of the user insurance record are respectively subjected to causal association-driven record semantic dynamic walking to obtain the user credit record time-series semantic walking coding vector and the user insurance record time-series semantic walking coding vector. It should be understood that, considering that the user's credit and insurance behavior change over time, the changes in these behaviors in different time periods and their causal associations are affected. For example, a recent claim event may have a greater impact on the current credit status than a claim event a few years ago. Therefore, the contextual information of the previous and next time points, so as to more comprehensively reflect the user's overall behavior pattern, the present application performs causal association-driven record semantic dynamic walking on the time series distribution of the semantic embedding coding vector of the user credit record and the time series distribution of the semantic embedding coding vector of the user insurance record to obtain the user credit record time-series semantic walking coding vector and the user insurance record time-series semantic walking coding vector. In this way, the intrinsic correlation and key information of the complex characteristics of users' credit and insurance records can be captured more efficiently, noise interference can be reduced, the limitation of focusing only on surface local patterns can be broken through, and the hidden global dependency relationships can be deeply explored, making the evaluation results more accurate and credible.

[0038] In one embodiment, Figure 5As shown, the record time series semantic walking unit 33 includes: a user credit record semantic implicit feature mining subunit 331, which is used to perform implicit feature mining on each user credit record semantic embedding coding vector in the time series distribution of the user credit record semantic embedding coding vector to obtain the time series distribution of the user credit record semantic embedding depth implicit feature coding vector; a user credit record semantic causal feature construction subunit 332, which is used to construct a user credit record semantic causal association topological feature matrix based on the time series distribution of the user credit record semantic embedding depth implicit feature coding vector; a user credit record semantic surface context dynamic walking subunit 333, which is used to construct a user credit record semantic causal association topological feature matrix and the user credit record semantic embedding coding vector. The time series distribution of the quantity is subjected to the dynamic graph convolution-driven feature dynamic walk to obtain the user credit record semantic surface context dynamic walk semantic coding vector; the user credit record semantic hidden layer feature extraction subunit 334 is used to perform the dynamic graph convolution-driven feature dynamic walk on the user credit record semantic causal association topological feature matrix and the time series distribution of the user credit record semantic embedding depth implicit feature coding vector to obtain the user credit record semantic hidden layer context dynamic walk semantic coding vector; the user credit record feature fusion subunit 335 is used to fuse the user credit record semantic surface context dynamic walk semantic coding vector and the user credit record semantic hidden layer context dynamic walk semantic coding vector to obtain the user credit record time series semantic walk semantic coding vector.

[0039] Exemplarily, in the user credit record semantic implicit feature mining subunit 331, it should be understood that the core of implicit feature mining is to alleviate the noise problem and limitations of explicit features. For example, some short-term behaviors of users may fluctuate due to external factors, but if risk assessment is based solely on these short-term behavior data, it may lead to misjudgment. Through implicit feature mining, the system can capture the user's long-term behavior patterns and potential trends, thereby more comprehensively understanding their credit status. For example, a user has had several minor overdue repayment records in the past few years, but the overall proportion of on-time repayment is high. Implicit feature mining can help identify this potential good credit habit, rather than simply focusing on individual overdue events. Further, implicit feature mining can efficiently compress important information while eliminating redundant information. This means that the system can still maintain efficiency and accuracy when processing large amounts of complex data. For example, when processing a user data with many years of credit history, implicit feature mining can help the system focus on the key factors that really affect the credit score, such as repayment frequency, default amount, and default time interval, rather than paying too much attention to some insignificant details or outliers. In addition, implicit feature mining can also reveal the intrinsic correlation and causal relationship between data. In credit records, multiple behaviors of users are often interrelated. For example, frequent small overdue payments may indicate greater financial pressure, thereby increasing the risk of default in the future. By mining these implicit features, the system can better understand these intrinsic connections and convert them into time series distributions of deep implicit feature encoding vectors. This representation method not only retains the time order of the original data, but also enhances the ability to capture dynamic changes in user behavior. Specifically, the processing process of the user credit record semantic implicit feature mining subunit 331 can be expressed as follows: in, is the time series distribution of the semantic embedding coding vector of the user's credit record, , , and are the first, second, and third in the time series distribution of the semantic embedding encoding vector of the user's credit record. and The semantic embedding encoding vector of user credit records, is the point convolutional coding, is the convolutional coding activation function, , , , and They are the first, second, and third in the time series distribution of the semantic embedding deep implicit feature encoding vector of the user’s credit record. , and The semantic embedding of user credit records into deep implicit feature encoding vectors, It is the time series distribution of the semantic embedded deep implicit feature encoding vector of the user’s credit record.

[0040] In one embodiment, Figure 6 As shown, the user credit record semantic causal feature construction subunit 332 includes: a user credit record semantic causal factor calculation secondary subunit 3321, which is used to calculate the semantic causal association factor between any two user credit record semantic embedding depth implicit feature coding vectors in the time series distribution of the user credit record semantic embedding depth implicit feature coding vector to obtain a user credit record semantic causal association topological matrix composed of multiple user credit record semantic causal association factors; a user credit record semantic causal triggering secondary subunit 3322, which is used to perform causal triggering of the gated activation function on the user credit record semantic causal association topological matrix to obtain the user credit record semantic causal association topological feature matrix.

[0041] In one embodiment, the user credit record semantic causal factor calculation secondary subunit 3321 includes: a semantic association matrix calculation tertiary subunit, which is used to calculate the association matrix between any two user credit record semantic embedded deep implicit feature coding vectors to obtain the user credit record semantic association matrix; a statistical feature-based causal factor calculation tertiary subunit, which is used to perform statistical feature-based causal factor calculation on the user credit record semantic association matrix to obtain the user credit record semantic causal association factor corresponding to the user credit record semantic association matrix, wherein the user credit record semantic causal association factor is related to the mean, variance, maximum value and causal bias value of the user credit record semantic association matrix.

[0042] in, is matrix multiplication, yes The transposed vector of yes and The semantic association matrix of user credit records between yes The variance of To obtain The maximum value in yes The mean of is the causal bias value, yes The corresponding semantic causal correlation factor of the user's credit record, , , and are the semantic causal association factors of user credit records at each position in the semantic causal association topological matrix of user credit records, It is the semantic causal association topology matrix of user credit records.

[0043] In one embodiment, the third-level sub-unit for calculating causal factors based on statistical features is used to: in response to the variance of the user credit record semantic association matrix being greater than or equal to a predetermined threshold, use the weighted average of the distances between any two user credit record semantic embedding deep implicit feature coding vectors in the time series distribution of the user credit record semantic embedding deep implicit feature coding vectors as the causal bias value; in response to the variance of the user credit record semantic association matrix being less than the predetermined threshold, use the weighted value of the mean of the user credit record semantic association matrix as the causal bias value.

[0044] Specifically, when calculating the semantic causal correlation factor between any two semantically embedded deep implicit feature encoding vectors of user credit records, the system is actually looking for the intrinsic connection and causal relationship between these features. For example, a user's overdue repayment behavior may be related to a previous credit limit adjustment or income change. By constructing a global fine-grained association matrix and using the causal association energy measurement function for quantitative encoding expression, the system can capture these complex causal relationships. Specifically, the system analyzes the degree of correlation between each pair of deep implicit feature encoding vectors and identifies which factors have significant causal relationships. This step helps the system extract the key factors that really affect the user's credit status from massive data.

[0045] Preferably, in another specific example of the present application, the statistical feature-based causal factor calculation three-level subunit is used to: in response to the variance of the user credit record semantic association matrix being greater than or equal to a predetermined threshold, use the weighted average of the distances between any two user credit record semantic embedding depth implicit feature coding vectors in the time series distribution of the user credit record semantic embedding depth implicit feature coding vector as the causal bias value; in response to the variance of the user credit record semantic association matrix being less than the predetermined threshold, use the weighted value of the mean of the user credit record semantic association matrix as the causal bias value. The process is expressed by the formula: in, yes and The semantic association matrix of user credit records between yes The variance of yes The mean of for and The distance between yes The number of vectors in is a predetermined threshold, and is a weighted hyperparameter, is the causal bias value.

[0046] In the above preferred example, by considering the low-level causal associations in the complex system as molecular-level relationships inferred based on statistical correlation, the intervention prediction of the causal energy can be further performed on the basis of the global fine-grained statistical association representation, so as to study the fine-grained structure of causal associations and its dynamic regulation based on the high-dimensional and heterogeneous representation of causal omics. Among them, when the aggregation distribution representation of the causal graph is greater than a predetermined threshold, the source data is integrated based on the matrix graph node effect representation of the semantic causal association factor of the user's credit record. When the aggregation distribution representation of the causal graph is less than a predetermined threshold, the condensed structure modeling can be directly performed through the integration and compression of the feature pattern. In this way, not only can the causal energy in the description system be encoded, but also the implicit causal intervention prediction results can be condensed, so as to obtain a more efficient disclosure of key causal associations.

[0047] Exemplarily, in the user credit record semantic causal triggering secondary subunit 3322, considering that a user credit record semantic causal association topological matrix composed of multiple user credit record semantic causal association factors is obtained, the system enters the next step - causal triggering of the gated activation function on this matrix. The core of this process is to dynamically extract and model causal topological relationships to enhance the expressiveness and accuracy of the model. The gating mechanism allows the model to identify key causal paths in a dynamic context, thereby strengthening important associations and reducing noise interference. For example, some short-term behavioral fluctuations (such as a small overdue payment) should not excessively affect the overall risk assessment results. Through the gating mechanism, the system can ignore these unimportant details and focus on those long-term and significant causal relationships, such as the impact of multiple large defaults on credit scores. On this basis, the nonlinear activation function enhances the expressiveness of the model by introducing nonlinear elements, capturing the high-order regularities hidden in the complex causal structure. Specifically, the processing process of the user credit record semantic causal triggering secondary subunit 3322 can be expressed by the formula: in, is a nonlinear activation function, is the normalized threshold, For Perform gated activation processing, It is the semantic causal association topological feature matrix of user credit records.

[0048] Exemplarily, in the user credit record semantic surface context dynamic walk subunit 333, the user credit record semantic causal association topological feature matrix and the user credit record semantic embedding coding vector time series distribution are subjected to dynamic graph convolution driven feature dynamic walk, aiming to generate the user credit record semantic surface context dynamic walk semantic coding vector. This process simulates the propagation of features in complex topological structures, recursively aggregates the explicit semantics between nodes from local to global, and thus captures the user's deep credit behavior pattern. Specifically, first, the system has obtained the time series distribution of the user credit record semantic embedding deep implicit feature coding vector through implicit feature mining, and constructed the user credit record semantic causal association topological feature matrix. Next, the system combines these two key inputs - the causal association topological feature matrix and the time series distribution of the semantic embedding coding vector - and uses a dynamic graph convolutional neural network (GCN) to perform feature dynamic walk. In this process, the dynamic graph convolution driven feature dynamic walk simulates the propagation process of features in complex topological structures. Specifically, the system starts from each node (i.e., the semantic embedding encoding vector of the user's credit record at each time point) and dynamically walks along the edges in the graph (i.e., the causal association path). This walking method not only considers the characteristics of the node itself, but also integrates the information of its neighboring nodes, thereby forming a more comprehensive and rich contextual representation. For example, suppose that a user's overdue repayment record (node ​​A) is causally related to his previous normal repayment records (nodes B, C, and D). Through the dynamic walking of features driven by dynamic graph convolution, the system can capture the mutual influence between these records. Starting from node A, the system gradually aggregates the feature information of these nodes along the causal paths connected to it (such as nodes B, C, and D) to form a comprehensive contextual representation. This method enables the system to better understand the user's overall credit behavior pattern, rather than just looking at a single event in isolation. After the above dynamic walking process, the system finally generates a semantic encoding vector of the user's credit record semantic surface context dynamic walking. This vector not only contains the feature information of each node itself, but also integrates the influence of its neighboring nodes, thereby providing a more comprehensive and detailed contextual representation. Specifically, this surface context dynamic walking semantic encoding vector can help the system better understand the user's long-term credit behavior pattern and its changing trend. Specifically, the processing process of the user credit record semantic surface context dynamic walking subunit 333 can be expressed as follows: in, It is the graph convolution processing, It is the dynamic walking semantic encoding vector of the semantic surface context of the user’s credit record.

[0049] Exemplarily, in the user credit record semantic latent feature extraction subunit 334, for the user credit record semantic embedding deep implicit feature encoding vector, the dynamic walking mechanism conducts a deeper semantic exploration of the implicit features. Hidden semantics enables the potential embedding of features, so it is necessary to focus on the multi-hop propagation of high-order information and the distributed decoupling of deep features during the walking process to avoid the over-smoothing phenomenon caused by the propagation of deep topological features. This latent representation provides greater generalization capability for feature expression by capturing profound temporal dependencies and complex semantic patterns. Specifically, the processing of the user credit record semantic latent feature extraction subunit 334 can be expressed by the formula: in, It is the dynamic walking semantic encoding vector of the semantic hidden context of the user’s credit record.

[0050] Exemplarily, in the user credit record feature fusion subunit 335, the system has generated two types of semantic coding vectors through dynamic graph convolution-driven feature dynamic walks: user credit record semantic surface context dynamic walk semantic coding vector and user credit record semantic hidden context dynamic walk semantic coding vector. The surface context dynamic walk semantic coding vector mainly focuses on the characteristics of the node itself and the influence of its direct neighbors, providing a more intuitive and explicit credit behavior pattern; while the hidden context dynamic walk semantic coding vector deeply explores higher-level relationships and deep-level causal associations, revealing those complex patterns and potential risk factors hidden under the surface. By fusing the surface and hidden semantic coding vectors, the system can capture multi-dimensional information about the user's credit behavior. For example, when evaluating a user's credit risk, the surface context dynamic walk semantic coding vector can help the system identify the user's recent repayment behavior and default situation, and provide short-term behavior pattern analysis. At the same time, the hidden context dynamic walk semantic coding vector can reveal the user's long-term behavior trends and potential risk factors, such as income fluctuations, the impact of emergencies on credit behavior, etc. Specifically, the processing process of the user credit record feature fusion subunit 335 can be expressed as follows: in, is the fusion weight parameter, It is the temporal semantic walk encoding vector of the user's credit record.

[0051] Exemplarily, in the record multimodal joint unit 34, it is used to fuse the user credit record time series semantic walk coding vector and the user insurance record time series semantic walk coding vector to obtain the user credit risk multimodal joint coding vector. It should be understood that considering that the user credit record time series semantic walk coding vector mainly reflects its credit performance in the credit field, such as repayment ability and willingness; the user insurance record time series semantic walk coding vector focuses on showing the user's behavior and potential risk status in risk protection. Both portray the user from different angles. Based on this, in the technical solution of the present application, the user credit record time series semantic walk coding vector and the user insurance record time series semantic walk coding vector are fused to comprehensively consider the credit and insurance record information, and the user credit risk multimodal joint coding vector is obtained, thereby more comprehensively reflecting the user's overall behavior pattern. In a specific example, the user credit record time series semantic walk coding vector and the user insurance record time series semantic walk coding vector are fused by weighted summation or attention mechanism to comprehensively consider the credit and insurance record information.

[0052] Exemplarily, the credit risk judgment unit 35 is used to obtain a recognition result indicating whether the insurance request is within the credit risk security level according to the multimodal joint coding vector of the user's credit risk. In one embodiment, the credit risk judgment unit is used to: input the multimodal joint coding vector of the user's credit risk into a credit risk identifier based on a classifier to obtain the recognition result, and the recognition result is used to indicate whether the insurance request is within the credit risk security level. That is, the multimodal joint coding vector of the user's credit risk obtained by multimodal combination of the user's credit record time series semantic wandering coding vector and the user's insurance record time series semantic wandering coding vector is classified and processed, so as to realize automatic judgment of whether the insurance request is within the credit risk security level. It should be understood that the classifier has a strong pattern recognition ability, and after training, it can learn the mapping relationship between different credit risk features and security levels. The credit risk identifier constructed based on the classifier can give full play to its advantages, analyze and judge the credit risk features contained in the multimodal joint coding vector, and thus give an accurate recognition result. By clarifying whether the insurance request is within the credit risk safety level, financial institutions and insurance companies can effectively screen out high-risk insurance requests and avoid providing insurance services to users with excessively high credit risks, thereby reducing potential default risks and economic losses and ensuring the sound operation of their own businesses. In a specific example, the classifier-based credit risk identifier uses a softmax classifier, which consists of the following parts: Input layer: Receives the multimodal joint encoding vector of the user's credit risk. Fully Connected Layer: Maps the input vector to a hidden layer or directly to the Softmax layer. Usually contains a weight matrix. Softmax layer: Converts the output of the fully connected layer into a probability distribution.

[0053] In summary, the blockchain-based full-process management and risk control system of the electronic letter of guarantee according to the embodiment of the present application is explained, which retrieves the user credit record and the user insurance record according to the user identity information, and uses the natural language analysis and processing algorithm based on deep learning to perform data splitting and semantic encoding on the user credit record and the user insurance record. Then, the semantic embedding of each encoded user credit record and the semantic embedding features of each user insurance record are respectively subjected to causal association semantic dynamic walking, so as to realize the automatic judgment of whether the insurance request is within the credit risk security level according to the multimodal joint representation of credit risk between the temporal semantic walking features of the user credit record and the temporal semantic walking features of the user insurance record obtained after the association. The present application can effectively parse and utilize the complex semantic information in the user credit record and the user insurance record, so as to make the basis for credit risk judgment more comprehensive and accurate. At the same time, the model can be continuously updated and adjusted according to new data, no longer constrained by fixed formulas and preset coefficient factors, and improve the accuracy and effectiveness of risk judgment in different scenarios.

[0054] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0055] It should be understood that the specific examples in this article are only intended to help those skilled in the art better understand the embodiments of the present application, rather than to limit the scope of the embodiments of the present application.

[0056] It should also be understood that in the various embodiments of the present application, the size of the serial number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0057] It should also be understood that the various implementation methods described in this specification can be implemented individually or in combination, and the embodiments of the present application are not limited to this.

[0058] Unless otherwise stated, all technical and scientific terms used in the embodiments of the present application are the same as the meanings generally understood by those skilled in the art of the technical field of the present application. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the scope of the present application. The term "and / or" used in the present application includes any and all combinations of one or more related listed items. The singular forms of "a kind of", "above" and "the" used in the embodiments of the present application and the appended claims are also intended to include majority forms, unless the context clearly indicates other meanings. In addition, the terms "first", "second" etc. are only used for description purposes and cannot be understood as indicating or suggesting relative importance.

[0059] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0060] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0061] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0062] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0063] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A blockchain-based electronic guarantee full-process management and risk control system, including: Platform building module, used to build an electronic letter of guarantee management platform using blockchain technology; A request module, used to initiate an insurance request to the electronic letter of guarantee management platform; A verification module, for verifying the identity of the user in response to the electronic letter of guarantee management platform receiving the insurance request to determine whether the insurance request is within the credit risk security level; a letter of guarantee issuance module, for approving the insurance request in response to the insurance request being within the credit risk security level, and issuing an electronic letter of guarantee for the approved insurance request and encrypting and storing it; It is characterized in that the verification module includes: a retrieval unit, which is used to retrieve the user credit record and the user insurance record according to the user identity information; a record splitting and encoding unit, which is used to perform data splitting and semantic encoding on the user credit record and the user insurance record to obtain the time series distribution of the user credit record semantic embedding coding vector and the time series distribution of the user insurance record semantic embedding coding vector; a record temporal semantic walking unit, which is used to perform causal association-driven record semantic dynamic walking on the time series distribution of the user credit record semantic embedding coding vector and the time series distribution of the user insurance record semantic embedding coding vector to obtain the user credit record temporal semantic walking coding vector and the user insurance record temporal semantic walking coding vector; a record multimodal joint unit, which is used to fuse the user credit record temporal semantic walking coding vector and the user insurance record temporal semantic walking coding vector to obtain the user credit risk multimodal joint coding vector; a credit risk judgment unit, which is used to obtain an identification result indicating whether the insurance request is within the credit risk security level according to the user credit risk multimodal joint coding vector.

2. The blockchain-based electronic letter of guarantee full-process management and risk control system according to claim 1 is characterized in that: The record splitting encoding unit comprises: A data splitting subunit, used to split the user credit record and the user insurance record according to the time dimension to obtain a list of user credit record items and a list of user insurance items; A user credit record semantic encoding subunit, configured to semantically encode each user credit record item in the list of user credit record items to obtain a time series distribution of the user credit record semantic embedding encoding vector; The user insurance record semantic coding subunit is used to semantically code each user insurance item in the list of user insurance items to obtain a time series distribution of the semantic embedded coding vector of the user insurance record.

3. The blockchain-based electronic letter of guarantee full-process management and risk control system according to claim 2 is characterized in that: The user credit record semantic encoding subunit is used to: use a semantic encoder including an RNN to semantically encode each user credit record item in the list of user credit record items to obtain a time series distribution of the user credit record semantic embedding encoding vector.

4. The blockchain-based electronic guarantee full-process management and risk control system according to claim 3 is characterized in that: The user insurance record semantic encoding subunit is used to: use the semantic encoder containing RNN to semantically encode each user insurance item in the list of user insurance items to obtain the time series distribution of the user insurance record semantic embedding encoding vector.

5. The blockchain-based electronic guarantee full-process management and risk control system according to claim 4 is characterized in that: The recording time sequence semantic walking unit includes: A user credit record semantic implicit feature mining subunit is used to perform implicit feature mining on each user credit record semantic embedding coding vector in the time series distribution of the user credit record semantic embedding coding vector to obtain a time series distribution of the user credit record semantic embedding depth implicit feature coding vector; A user credit record semantic causal feature construction subunit, for constructing a user credit record semantic causal association topological feature matrix based on the time series distribution of the user credit record semantic embedding deep implicit feature coding vector; A user credit record semantic surface context dynamic walking subunit, used to perform dynamic graph convolution-driven feature dynamic walking on the user credit record semantic causal association topological feature matrix and the time series distribution of the user credit record semantic embedding coding vector to obtain a user credit record semantic surface context dynamic walking semantic coding vector; A user credit record semantic hidden layer feature extraction subunit is used to perform the dynamic graph convolution-driven feature dynamic walk on the user credit record semantic causal association topological feature matrix and the time series distribution of the user credit record semantic embedding depth implicit feature coding vector to obtain the user credit record semantic hidden layer context dynamic walk semantic coding vector; The user credit record feature fusion subunit is used to fuse the user credit record semantic surface context dynamic wandering semantic coding vector and the user credit record semantic hidden context dynamic wandering semantic coding vector to obtain the user credit record time series semantic wandering coding vector.

6. The blockchain-based electronic guarantee full-process management and risk control system according to claim 5 is characterized in that: The user credit record semantic causal feature construction subunit includes: A user credit record semantic causal factor calculation secondary subunit is used to calculate the semantic causal association factor between any two user credit record semantic embedding depth implicit feature coding vectors in the time series distribution of the user credit record semantic embedding depth implicit feature coding vector to obtain a user credit record semantic causal association topological matrix composed of multiple user credit record semantic causal association factors; The user credit record semantic causal triggering secondary sub-unit is used to perform causal triggering of the gated activation function on the user credit record semantic causal association topology matrix to obtain the user credit record semantic causal association topology feature matrix.

7. The blockchain-based electronic guarantee full-process management and risk control system according to claim 6 is characterized in that: The user credit record semantic causal factor calculation secondary subunit includes: The semantic association matrix calculation tertiary subunit is used to calculate the association matrix between any two user credit record semantic embedding deep implicit feature encoding vectors to obtain the user credit record semantic association matrix; The third-level sub-unit for causal factor calculation based on statistical features is used to perform causal factor calculation based on statistical features on the user credit record semantic association matrix to obtain the user credit record semantic causal association factor corresponding to the user credit record semantic association matrix, wherein the user credit record semantic causal association factor is related to the mean, variance, maximum value and causal bias value of the user credit record semantic association matrix.

8. The blockchain-based electronic letter of guarantee full-process management and risk control system according to claim 7 is characterized in that: The statistical feature-based causal factor calculation three-level sub-unit is used to: In response to the variance of the user credit record semantic association matrix being greater than or equal to a predetermined threshold, taking the weighted average of the distances between any two user credit record semantic embedding deep implicit feature encoding vectors in the time series distribution of the user credit record semantic embedding deep implicit feature encoding vectors as the causal bias value; In response to the variance of the user credit record semantic association matrix being less than the predetermined threshold, a weighted value of the mean of the user credit record semantic association matrix is ​​used as the causal bias value.

9. The blockchain-based electronic guarantee full-process management and risk control system according to claim 8 is characterized in that: The credit risk judgment unit is used to: input the user credit risk multimodal joint encoding vector into a classifier-based credit risk identifier to obtain the identification result, and the identification result is used to indicate whether the insurance request is within the credit risk safety level.

Images