Calculation method of expected credit loss and credit management system

Abstract

The invention discloses an expected credit loss calculation method and a credit management system, which can improve the calculation efficiency of expected credit loss. The expected credit loss calculation method can be applied to a credit management system. In a specific implementation, a credit management system obtains a calculation request for an expected credit loss, the calculation request comprising an identifier of a first calculation scenario. The credit management system also screens out a calculation element table from the transaction data of the first calculation scene according to the identifier of the first calculation scene and the feature table. Wherein the feature table comprises calculation elements associated with a plurality of credit risk slow release tools, and the calculation element table comprises a first calculation element associated with a first credit risk slow release tool involved in the transaction data and a value of the first calculation element. And then, the credit management system calculates expected credit loss according to the calculation element table.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method for calculating expected credit loss and a credit management system. Background Technology

[0002] Expected credit loss (EL) refers to the potential losses a company may incur due to borrower default or inability to repay loans on time within a future period. This indicator is crucial for a company's risk management and accurate balance sheet reflection. Currently, expected credit loss can be calculated by finance personnel through collecting loan transaction data, identifying credit risk-related parameters (e.g., contract aging, borrower credit rating), and performing risk assessments based on these parameters. However, it's understandable that as loan transaction scenarios become more complex and the amount of transaction data increases, relying solely on manual searching for credit risk-related parameters would negatively impact the efficiency of expected credit loss calculation. Summary of the Invention

[0003] This application provides a method for calculating expected credit loss and a credit management system, which can improve the efficiency of calculating expected credit loss.

[0004] Firstly, this application provides a method for calculating expected credit loss. This method can be applied to a credit management system. Specifically, the credit management system obtains a calculation request for expected credit loss, which includes an identifier of a first calculation scenario. The credit management system also filters a calculation element table from the transaction data of the first calculation scenario based on the identifier of the first calculation scenario and a feature table. The feature table includes calculation elements associated with multiple credit risk mitigation tools, and the calculation element table includes first calculation elements associated with the first credit risk mitigation tool involved in the aforementioned transaction data and the values ​​of the first calculation elements. Subsequently, the credit management system calculates the expected credit loss based on the calculation element table.

[0005] In the technical solution provided by this application, the credit management system can automatically obtain transaction data of the first calculation scenario according to the expected credit loss calculation request, and filter out the first calculation element and the value of the first calculation element required for calculating the expected credit loss from the transaction data according to the feature table. Then, the expected credit loss is calculated based on the first calculation element and its value. Therefore, the expected credit loss calculation method provided by this application has a high degree of automation and can improve the calculation efficiency of expected credit loss.

[0006] In one possible implementation, the credit management system filters a table of computational elements from the transaction data of the first computational scenario based on the identifier and feature table of the first computational scenario. This includes: the credit management system obtains the transaction data based on the identifier of the first computational scenario; then, based on the matching relationship between the identifier of the first credit risk mitigation tool in the transaction data and the identifier of the credit risk mitigation tool in the feature table, it obtains a first computational element from the feature table. The first computational element is the computational element associated with the credit risk mitigation tool indicated by the matching identifier. Subsequently, the credit management system obtains the value of the first computational element from the transaction data based on the first computational element.

[0007] In the above implementation, the credit management system can identify the first credit risk mitigation tool involved in the transaction data of the first calculation scenario, as well as the first calculation element associated with the first credit risk mitigation tool and the value of the first calculation element, based on the feature table. It should be understood that the feature table is configured by the user. When business requirements are updated—for example, if new business requirements necessitate adding, deleting, or modifying the calculation elements associated with a credit risk mitigation tool—the user can update the feature table according to the new business requirements. The credit management system can then complete the new calculation task based on the new feature table. Throughout this process, no software code modification is required. Therefore, the above implementation method can improve the scalability and maintainability of the software code and reduce the calculation cost of expected credit losses.

[0008] In another possible implementation, the credit management system also generates a calculation factor combination table based on multiple calculation factor tables configured by the user. Each calculation factor table corresponds to a calculation factor in the formula for calculating expected credit loss. The calculation factor table includes the calculation factor and its value. The calculation factor combination table is the combination result determined based on the aforementioned multiple calculation factor tables. The credit management system calculates the expected credit loss based on the calculation factor table, including: the credit management system calculates the expected credit loss based on the calculation factor table and the calculation factor combination table.

[0009] In the above implementation, the credit management system can combine multiple user-configured calculation factor tables into a single calculation factor combination table. This allows the system to directly retrieve the necessary data for calculating expected credit loss from this combination table. Compared to retrieving multiple calculation factor tables, this approach improves data retrieval efficiency.

[0010] In another possible implementation, the credit management system calculates the expected credit loss based on a calculation element table and a calculation factor combination table. This includes: the credit management system obtaining the value of a first calculation factor from the calculation element table based on the matching relationship between the calculation factors in the calculation formula and the calculation elements in the calculation element table; the value of the first calculation factor is the value of the matching calculation element. The credit management system also obtains the value of a second calculation factor from the calculation factor combination table based on the matching relationship between the calculation factors in the calculation formula and the calculation factors in the calculation factor combination table; the value of the second calculation factor is the value of the matching calculation factor. Then, the credit management system obtains the expected credit loss based on the values ​​of the first and second calculation factors.

[0011] Through the above implementation method, the credit management system can obtain the data required to calculate expected credit loss by querying the calculation element table and calculation factor table, thereby completing the calculation of expected credit loss.

[0012] In another possible implementation, the calculation factor combination table also includes the calculation elements associated with the calculation factors and multiple values ​​for the calculation elements. The credit management system retrieves the value of the second calculation factor in the calculation formula from the calculation factor combination table based on the matching relationship between the calculation factors in the calculation formula and the calculation factors in the calculation factor combination table. This includes: the credit management system retrieves the second calculation element associated with the second calculation factor from the calculation factor combination table based on the matching relationship between the calculation factors in the calculation formula and the calculation factors in the calculation factor combination table; then, based on the second calculation element, retrieves the value of the second calculation element from the aforementioned transaction data; and finally, based on the value of the second calculation element, retrieves the value of the second calculation factor from the calculation factor combination table.

[0013] Through the above implementation method, the credit management system can obtain the data required to calculate the expected credit loss from the transaction data of the first calculation scenario based on the calculation factor combination table.

[0014] In another possible implementation, the credit management system also updates at least one of the aforementioned feature table, calculation factor combination table, and calculation formula according to the user's configuration, wherein the updated feature table, calculation factor combination table, and calculation formula meet the new business requirements.

[0015] Secondly, this application provides a credit management system. The system includes an acquisition module and a calculation module. The acquisition module acquires a calculation request for expected credit loss, the calculation request including an identifier of a first calculation scenario. The calculation module filters a calculation element table from transaction data of the first calculation scenario based on the identifier of the first calculation scenario and a feature table. The feature table includes calculation elements associated with multiple credit risk mitigation tools, and the calculation element table includes first calculation elements associated with the first credit risk mitigation tool involved in the aforementioned transaction data and the values ​​of the first calculation elements. The calculation module also calculates the expected credit loss based on the calculation element table.

[0016] In one possible implementation, the calculation module is used to obtain the transaction data based on the identifier of the first calculation scenario; based on the matching relationship between the identifier of the first credit risk mitigation tool in the transaction data and the identifier of the credit risk mitigation tool in the feature table, the first calculation element is the calculation element associated with the credit risk mitigation tool indicated by the matching identifier; and the value of the first calculation element is obtained from the transaction data based on the first calculation element.

[0017] In another possible implementation, the credit management system also includes a configuration module. The configuration module generates a calculation factor combination table based on multiple calculation factor tables configured by the user. Each calculation factor table corresponds to a calculation factor in the formula for calculating expected credit loss. The calculation factor table includes the calculation factor and its value, and the calculation factor combination table is the combination result determined based on the aforementioned multiple calculation factor tables. The calculation module calculates the expected credit loss based on the calculation factor table and the calculation factor combination table.

[0018] In another possible implementation, the calculation module is used to obtain the value of a first calculation factor in the calculation formula from the calculation element table based on the matching relationship between the calculation factor in the calculation formula and the calculation element in the calculation element table, wherein the value of the first calculation factor is the value of the matching calculation element; based on the matching relationship between the calculation factor in the calculation formula and the calculation factor combination table, the module obtains the value of a second calculation factor in the calculation formula from the calculation factor combination table, wherein the value of the second calculation factor is the value of the matching calculation factor; and based on the values ​​of the first calculation factor and the second calculation factor, the expected credit loss is obtained.

[0019] In another possible implementation, the calculation factor combination table also includes calculation elements associated with the calculation factors and multiple values ​​for the calculation elements. The calculation module is used to obtain the second calculation element associated with the second calculation factor from the calculation factor combination table based on the matching relationship between the calculation factors in the calculation formula and the calculation factors in the calculation factor combination table; obtain the value of the second calculation element from the aforementioned transaction data based on the second calculation element; and obtain the value of the second calculation factor from the calculation factor combination table based on the value of the second calculation element.

[0020] In another possible implementation, the configuration module is also used to update at least one of the aforementioned feature table, calculation factor combination table, and calculation formula according to the user's configuration, wherein the updated feature table, calculation factor combination table, and calculation formula meet the new business requirements.

[0021] Thirdly, this application provides a computing device. The computing device includes a processor and a memory, wherein the processor is configured to execute instructions stored in the memory to cause the computing device to perform some or all of the methods described in the first aspect and any implementation thereof.

[0022] Fourthly, this application provides a computing device cluster. The computing device cluster includes at least one computing device, each computing device including a processor and a memory. The processor of the at least one computing device is used to execute instructions stored in the memory of the at least one computing device, causing the computing device cluster to perform some or all of the methods described in the first aspect and any implementation thereof.

[0023] Fifthly, this application provides a computer program product containing instructions. This computer program product may be a software or program product containing instructions that can run on a computing device or be stored on any usable medium. When the computer program product is run on a computing device, it causes the computing device to perform some or all of the methods described in the first aspect and any implementation thereof.

[0024] Sixthly, this application provides a computer-readable storage medium. The computer storage medium includes computer program instructions that, when executed by a computing device, cause the computing device to perform some or all of the methods described in the first aspect and any implementation thereof. Attached Figure Description

[0025] Figure 1 This is a schematic diagram illustrating a calculation scenario for expected credit loss provided in this application;

[0026] Figure 2 This is a schematic diagram of an application scenario provided in this application;

[0027] Figure 3 This is a flowchart illustrating a method for calculating expected credit loss provided in this application;

[0028] Figure 4 This is a schematic diagram of the structure of a credit management system provided in this application;

[0029] Figure 5 This is a schematic diagram of the structure of a computing device provided in this application;

[0030] Figure 6 This is a schematic diagram of the structure of a computing device cluster provided in this application;

[0031] Figure 7 This is a schematic diagram of another computing device cluster provided in this application. Detailed Implementation

[0032] To address the issue of low efficiency in calculating expected credit losses, this application provides a method for calculating expected credit losses. This method can be applied to a credit management system. The credit management system has a pre-set feature table and a calculation factor combination table. The feature table records calculation elements associated with multiple credit risk mitigation tools that need to participate in the calculation of expected credit losses. The calculation factor combination table records multiple calculation factors and their values ​​in the expected credit loss calculation formula. When calculating expected credit losses for transaction data in a first calculation scenario, the credit management system can, based on the aforementioned feature table, filter out a calculation element table from the transaction data of the first calculation scenario. This calculation element table includes calculation elements associated with the credit risk mitigation tools used in that scenario and their values. Then, by retrieving the calculation element table and the calculation factor combination table, the values ​​of the calculation factors in the expected credit loss calculation formula are obtained, thereby calculating the expected credit loss. In the above technical solution, the credit management system can automatically acquire the data required for calculating expected credit losses and complete the calculation based on the acquired data. Therefore, the above method can improve the efficiency of expected credit loss calculation. On the other hand, the aforementioned feature table, calculation factor combination table, and expected credit loss calculation formula can be configured according to the different business needs of different enterprises. Therefore, the above method can be applied to the calculation of expected credit loss in various calculation scenarios, reducing the calculation cost of expected credit loss.

[0033] The technical solution provided in this application will now be described with reference to the accompanying drawings.

[0034] See Figure 1 , Figure 1 A schematic diagram illustrating a calculation scenario for expected credit loss to which this application applies is shown. For example... Figure 1As shown, the scenario includes a client 100, a credit management system 200, and a database system 300. The credit management system 200 is connected to the client 100 and the database system 300 via a network, which can be a wide area network or a local area network.

[0035] Client 100 can be software or applications (such as browsers, applications (APPs)) deployed on terminal devices. The terminal devices can be desktop computers, laptops, tablets, smartphones, wearable devices, etc.

[0036] The credit management system 200 can be deployed on a single computing device or a cluster of computing devices consisting of multiple computing devices. The computing devices can be servers, such as central servers, edge servers, or local servers in local data centers, or terminal devices such as desktop computers, laptops, or smartphones.

[0037] Database system 300 can be an integrated storage and computing system, specifically including a storage cluster. The storage cluster includes one or more servers, where a server is a device with both computing and storage capabilities, such as a desktop computer, ARM server, or x86 server. Database system 300 can also be a storage and computing separated system, specifically including a compute node cluster and a storage node cluster. The compute node cluster includes one or more compute nodes, where a compute node is a computing device, such as a server, desktop computer, or storage array controller. The storage node cluster includes one or more storage nodes, where a storage node is a storage device, which can include hard disks, magnetic disks, or other types of storage media.

[0038] As one possible implementation, the credit management system 200 and database system 300 can be deployed in a cloud data center and provided to users as cloud services based on the basic resources (including computing resources, storage resources, and network resources) provided by the cloud data center. The client 100 can be deployed on the terminal devices used by the user. Figure 2 As shown, the cloud management platform provides the credit management system 200 and the database system 300 as a cloud service to enterprises. Enterprises can obtain access to the credit management system 200 and the database system 300 by purchasing cloud services, and provide enterprise employees with accounts for using the above cloud services. This allows enterprise employees to log in to their assigned accounts through the client 100 and use the credit management system 200 and the database system 300 to calculate expected credit losses.

[0039] It is worth noting that the deployment methods of the credit management system 200 and the database system 300, in addition to referring to... Figure 1Besides the separate deployment shown, it can also be deployed in a unified manner according to business needs. For example, the credit management system 200 and the database system 300 can be deployed on the same hardware resources, and the credit management system 200 and the database system 300 can implement their respective functions through different software modules. For ease of description, the following embodiments use... Figure 1 The deployment methods of the credit management system 200 and database system 300 shown are introduced.

[0040] exist Figure 1 In the scenario shown, the credit management system 200 provides configurability, allowing employees of the first enterprise to configure calculation rules for expected credit losses based on business or enterprise needs. These rules can include a feature table associated with credit risk mitigation tools, a calculation factor table, and a calculation formula for expected credit losses. Each feature table can include multiple credit risk mitigation tools and the calculation elements associated with each tool that need to participate in the expected credit loss calculation. The calculation factor table includes multiple calculation factors and their values. The credit management system 200 can also store the configured calculation rules in the database system 300. When employees of other enterprises belonging to the same enterprise as the employee of the first enterprise need to calculate expected credit losses for transaction data in the first calculation scenario, these employees can initiate a calculation request to the credit management system 200 through the client 100. The credit management system 200 can retrieve the transaction data for the first calculation scenario and the expected credit loss calculation rules from the database system 300 based on the calculation request, and then process the transaction data according to the expected credit loss calculation rules to calculate the expected credit loss. The credit management system 200 processes the transaction data of the first calculation scenario according to the calculation rules of expected credit, including the following steps: The credit management system 200 filters the transaction data of the first calculation scenario from the feature table to obtain a calculation element table, wherein the calculation element table includes the first calculation element associated with the credit risk mitigation tools involved in the transaction data and the value of the first calculation element. Then, based on the calculation element table and the calculation factor combination table, the value of the calculation factor in the calculation formula of expected credit loss is determined, and the expected credit loss is calculated based on the determined value. Afterwards, the credit management system 200 returns the calculation result of expected credit loss to the client 100, so that the user can know the expected credit loss corresponding to the transaction data of the first calculation scenario through the client 100.

[0041] Next, combined Figure 3 The flowchart illustrating the calculation method for expected credit loss provides a more detailed explanation of the process by which the aforementioned credit management system 200 calculates expected credit loss.

[0042] Step 101: The credit management system 200 generates calculation rules for expected credit loss based on the user's configuration.

[0043] Specifically, the credit management system 200 provides an access interface, which can be implemented through an application programming interface (API) or a graphical user interface (GUI). Users can provide configurations to the credit management system 200 through the aforementioned access interface. Correspondingly, the credit management system 200 can also obtain the user's configuration through the aforementioned access interface and generate calculation rules for expected credit losses based on the user's configuration.

[0044] The rules for calculating expected credit losses include a feature table associated with credit risk mitigation instruments (hereinafter referred to as the "Feature Table"), a calculation factor combination table, and a formula for calculating expected credit losses (hereinafter referred to as the "Calculation Formula"). These three parts are described below:

[0045] (1) The feature table records the identifiers of multiple credit risk mitigation instruments, as well as the calculation elements associated with each credit risk mitigation instrument.

[0046] The feature table can be configured by the user according to the credit risk mitigation tools required by the business. Among them, credit risk mitigation tools (also known as "borrower credit enhancement tools") refer to specific tools used in advance to ensure that the borrower can fulfill its payment obligations as agreed. Through credit risk mitigation tools, the risk of loan transactions can be transferred to an acceptable third party, or the probability of occurrence of loan transaction risks or the scale of loss can be reduced to an acceptable range.

[0047] It should be understood that credit risk mitigation tools involve multiple elements, but when calculating expected credit losses, only some of the above elements need to be used. The elements involved in the calculation of expected credit losses are the calculation elements associated with the credit risk mitigation tools.

[0048] To facilitate understanding, the following explanation is provided in conjunction with the feature table shown in Table 1:

[0049] Table 1 Feature Table

[0050] Credit risk mitigation tools Identification of credit risk mitigation instruments Calculation elements associated with credit risk mitigation instruments Credit Risk Mitigation Tool A Identifier A Letter of Credit Return Amount Credit Risk Mitigation Tool B Identifier B Guarantee Amount Credit Risk Mitigation Tool B Identifier B Currency types

[0051] The contents of Table 1 will be introduced next:

[0052] ① Credit Risk Mitigation Tool A refers to tools that mitigate credit risk based on letters of credit (LCs). A letter of credit is a written document issued by the importer's bank, guaranteeing payment to the exporter upon delivery of goods or services that meet contractual requirements. Through letters of credit, exporters obtain payment security, and importers can ensure payment only after the goods meet contractual requirements, thereby reducing transaction risk. A letter of credit may include information about the issuing bank, exporter, importer, and other parties, as well as transaction data such as goods description, payment amount, shipment date, and withdrawal amount.

[0053] Credit risk mitigation instrument A is identified as Identifier A, and the calculation element associated with credit risk mitigation instrument A is the letter of credit exit amount. In other words, when using credit risk mitigation instrument A, the element used in the calculation of expected credit loss is the "letter of credit exit amount," while other elements in the letter of credit (such as issuing bank information, exporter information, shipment date, etc.) do not need to be included in the calculation of expected credit loss.

[0054] ② Credit Risk Mitigation Tool B refers to tools that mitigate credit risk based on guarantees. A guarantee is a written credit guarantee issued by a bank, insurance company, or other guarantee company, which transfers the risk of a loan transaction to the issuer of the guarantee. A guarantee may include transaction data such as the issuer, beneficiary, applicant, issuance date, terms, amount, and currency.

[0055] Credit Risk Mitigation Instrument B is designated as Identifier B. The calculation elements associated with Credit Risk Mitigation Instrument B include the guarantee amount and the currency. In other words, when using Credit Risk Mitigation Instrument B, the calculation of expected credit losses requires the use of the "guarantee amount" and "currency," while other elements of the guarantee (such as the issuer information, beneficiary information, and issuance date) do not need to be included in the calculation of expected credit losses.

[0056] It should be understood that Table 1 is merely an example illustrating one possible feature table. In practical applications, there may be more calculation elements associated with credit risk mitigation instrument A. Furthermore, other credit risk mitigation instruments may be used in actual loan transactions, in which case the feature table may also include the identifiers of other credit risk mitigation instruments and their associated calculation elements; this application does not limit this.

[0057] (2) The calculation factor combination table records multiple calculation factors, the calculation elements associated with each calculation factor and the values ​​of the calculation elements, as well as the value of each calculation factor. The multiple calculation factors in the calculation factor combination table include those involved in the calculation formula. A calculation factor can be associated with one or more calculation elements, where calculation elements refer to the elements in the loan transaction data that participate in the calculation of expected credit loss. The values ​​of the calculation elements associated with a calculation factor can include one or more, and the value of the calculation factor may differ depending on the value of the calculation element.

[0058] The calculation factor combination table can be obtained by the credit management system 200 by combining multiple calculation factor tables configured by the user. In other words, the calculation factor combination table is the combination result determined by multiple calculation factor tables configured by the user. Each calculation factor table configured by the user corresponds to one calculation factor. Specifically, a calculation factor table can include the calculation factor, the calculation elements associated with the calculation factor and the values ​​of the calculation elements, as well as the value of the calculation factor. In the calculation factor table, the calculation factor can be associated with one or more calculation elements, and the calculation elements can have one or more values. Furthermore, different values ​​of the calculation elements may result in different values ​​for the calculation factor.

[0059] For ease of understanding, the following explanation is based on Tables 2 to 4:

[0060] Table 2 Calculation Factors Table - Loss on Default Rate

[0061] Credit rating\Contract aging 4-6 months 7-12 months 13-24 months A 0.85％ 2.36％ 9.78％ B 2.24％ 5.36％ 17.67％

[0062] Table 3 Calculation Factors Table - Probability of Default

[0063] Credit rating / duration 1 year 2 years 3 years A 0.06％ 0.18％ 0.37％ B 0.8％ 2.24％ 3.95％

[0064] Table 4 Calculation Factor Combination Table

[0065]

[0066] The contents of Tables 2 to 4 will be introduced next:

[0067] ① Tables 2 and 3 are the user configuration calculation factor tables.

[0068] The calculation factor in Table 2 is the loss given default (LGD). The loss given default (LGD) is the percentage of outstanding loan amount that a creditor might lose if the borrower defaults. LGD is usually expressed as a percentage. Table 2 shows the calculation factors associated with the LGD, different values ​​for these factors, and the corresponding LGD values. The calculation factors associated with the LGD include the borrower's credit rating and the contract age. Contract age refers to the length of time the contract between the borrower and creditor has existed, usually calculated in months or years. As shown in Table 2, when the borrower's credit rating is A and the contract age is 4-6 months, the LGD is 0.85%; when the borrower's rating is A and the contract age is 7-12 months, the LGD is 2.36%; and when the borrower's rating is A and the contract age is 13-24 months, the LGD is 9.78%. When the borrower's credit rating is B and the contract age is 4-6 months, the loss due to default is 2.24%; when the borrower's credit rating is B and the contract age is 7-12 months, the loss due to default is 5.36%; and when the borrower's credit rating is B and the contract age is 13-24 months, the loss due to default is 17.67%.

[0069] The calculation factor in Table 3 is the probability of default (PD). The probability of default refers to the likelihood that a borrower will default within a certain period. Default probability is usually expressed as a percentage. Table 3 shows the calculation elements associated with the probability of default, different values ​​of these elements, and the corresponding default probabilities for each value. The calculation elements associated with the probability of default include the borrower's credit rating and the duration of the default period. As shown in Table 3, a borrower with a credit rating of A has a default probability of 0.06%, 0.18%, and 0.37% within the next 1, 2, and 3 years, respectively. A borrower with a credit rating of B has a default probability of 0.8%, 2.24%, and 3.95% within the next 1, 2, and 3 years, respectively.

[0070] ② Table 4 is a table of calculation factor combinations generated by the credit management system 200 based on the calculation factor tables shown in Tables 2 and 3.

[0071] Table 4 encompasses the content of Tables 2 and 3. Specifically, the calculation factors in Table 4 include the default loss ratio from Table 2 and the default probability from Table 3. Table 4 shows the calculation elements associated with the default loss ratio, different values ​​of these elements, and the corresponding default loss ratios under different values. Similarly, it shows the calculation elements associated with the default probability, different values ​​of these elements, and the corresponding default probabilities under different values. As shown in Table 4, it includes six columns: "XDimension_name", "XDimension_value", "YDimension_name", "YDimension_value", "target_name", and "target_value".

[0072] The data in the “XDimension_name” column is the row header attribute name from Table 2 or Table 3, namely “Credit Rating”.

[0073] The data in the “XDimension_value” column is the row header attribute value from Table 2 or Table 3, namely “A” or “B”.

[0074] The data in the “YDimension_name” column is the column header attribute name of Table 2 or Table 3. The column header attribute name of Table 2 is “Contract Aging”, and the column header attribute name of Table 3 is “Duration”.

[0075] The data in the “YDimension_value” column is the column header attribute value of Table 2 or Table 3. The column header attribute values ​​of Table 2 include “4-6 months”, “7-12 months”, and “13-24 months”, while the column header attribute values ​​of Table 3 include “1 year”, “2 years”, and “3 years”.

[0076] The data in the “target_name” column is the attribute name (i.e., the calculated factor) of the return value of Table 2 or Table 3. The attribute name of the return value of Table 2 is “Default Loss Rate”, and the attribute name of the return value of Table 3 is “Default Probability”.

[0077] The data in the “target_value” column is the return value (i.e., the value of the calculated factor) of Table 2 or Table 3. Specifically, when the row header attribute value of Table 2 is “A” and the column header attribute value is “4-6 months”, the return value of Table 2 is “0.85%”; when the row header attribute value of Table 2 is “A” and the column header attribute value is “7-12 months”, the return value of Table 2 is “2.36%”; when the row header attribute value of Table 2 is “A” and the column header attribute value is “13-24 months”, the return value of Table 2 is “9.78%”; when the row header attribute value of Table 2 is “B” and the column header attribute value is “4-6 months”, the return value of Table 2 is “2.24%”; when the row header attribute value of Table 2 is “B” and the column header attribute value is “7-12 months”, the return value of Table 2 is “5.36%”; and when the row header attribute value of Table 2 is “B” and the column header attribute value is “13-24 months”, the return value of Table 2 is “17.67%”. When the row header attribute value of Table 3 is "A" and the column header attribute value is "1 year", the return value of Table 3 is "0.06%"; when the row header attribute value of Table 3 is "A" and the column header attribute value is "2 years", the return value of Table 3 is "0.18%"; when the row header attribute value of Table 3 is "A" and the column header attribute value is "3 years", the return value of Table 3 is "0.37%"; when the row header attribute value of Table 3 is "B" and the column header attribute value is "1 year", the return value of Table 3 is "0.8%"; when the row header attribute value of Table 3 is "B" and the column header attribute value is "2 years", the return value of Table 3 is "2.24%"; when the row header attribute value of Table 3 is "B" and the column header attribute value is "3 years", the return value of Table 3 is "3.95%".

[0078] Specifically, taking one row of data in Table 4 as an example, the following explains how the credit management system 200 generates Table 4 based on user-configured Tables 2 and 3: The credit management system 200 extracts the row header attribute name (i.e., "Credit Rating") from Table 2 and records the "Credit Rating" in the "XDimension_name" column of Table 4; it extracts the row header attribute value (i.e., "A") from Table 2 and records "A" in the "XDimension_value" column of Table 4; it extracts the column header attribute name (i.e., "Contract Aging") from Table 2 and records the "Contract Aging" in Table 4. The table retrieves the "YDimension_name" column from Table 2; extracts the column header attribute value (i.e., "4-6 months") from Table 2 and records "4-6 months" into the "YDimension_value" column; extracts the loss due to default (ODB) value (i.e., "0.85%) corresponding to a credit rating of A and a contract age of 4-6 months from Table 2 and records "0.85%" into the "target_value" column in Table 4; and records the corresponding calculation factor (i.e., "ODB") from Table 2 into the "target_name" column in Table 4. This yields the second row of data in Table 4. Similarly, the data for the other rows in Table 4 can be obtained, resulting in the calculation factor combination table shown in Table 4.

[0079] It should be understood that Table 4 is merely an illustrative representation of one possible calculation factor table. In practical applications, the loss ratio can be associated with more calculation elements, and the calculation elements associated with the loss ratio can have more values. Similarly, the probability of default can be associated with more calculation elements, and the calculation elements associated with the probability of default can have more values. Furthermore, the calculation of expected credit loss may use other calculation factors, so the calculation factor table may also include other calculation factors and their associated calculation elements, which is not limited in this application.

[0080] (3) The calculation formula is used to calculate expected credit loss.

[0081] The calculation formula is preset by the user based on the aforementioned feature table and calculation factor combination table. In specific implementation, the user can preset the calculation factors in the calculation formula and the operation rules between the calculation factors based on the credit risk mitigation tools and their associated calculation elements in the feature table, as well as the calculation factors in the calculation factor combination table, so that the calculation formula can meet the actual loan transaction business needs.

[0082] (4) When business requirements change (e.g., adding or deleting credit risk mitigation tools), users can provide new configurations based on the new business requirements. Accordingly, the credit management system 200 can update at least one of the above-mentioned feature table, calculation factor combination table, and calculation formula according to the user's configuration. The updated feature table, calculation factor combination table, and calculation formula meet the new business requirements.

[0083] Step 102: The credit management system 200 sends the calculation rules for expected credit loss to the database system 300. Accordingly, the database system 300 receives and stores the calculation rules for expected credit loss.

[0084] Step 103: Client 100 sends a calculation request to credit management system 200. Accordingly, credit management system 200 receives the expected credit loss calculation request sent by client 100.

[0085] Step 104: The credit management system 200 obtains the transaction data of the first calculation scenario and the calculation rules of expected credit loss from the database system 300 according to the calculation request.

[0086] Specifically, the calculation request includes an identifier for a first calculation scenario. Database system 300 stores identifiers for multiple calculation scenarios and transaction data associated with each scenario. Therefore, credit management system 200 can retrieve transaction data for the first calculation scenario from database system 300 based on the identifier of the first calculation scenario in the calculation request. The first calculation scenario can be a calculation scenario determined by the user based on multiple dimensions such as loan transaction, customer, region, time, and industry. Transaction data for the first calculation scenario can include information related to credit risk mitigation tools used in the loan transaction (hereinafter referred to as "first credit risk mitigation tools," the number of which can be one or more), the borrower's credit rating, contract aging, etc. The calculation request also includes a request type, which indicates whether the calculation request is for calculating expected credit loss. The request type is associated with an identifier for the expected credit loss calculation rule. Database system 300 also stores the expected credit loss calculation rule and its identifier. Therefore, credit management system 200 can also retrieve the expected credit loss calculation rule from database system 300 based on the identifier of the aforementioned calculation rule associated with the request type in the calculation request.

[0087] More specifically, the credit management system 200 can generate a data query request based on the identifier of the first calculation scenario and the request type in the calculation request. The data query request includes the identifier of the first calculation scenario and the identifier of the calculation rule for expected credit loss. Then, the credit management system 200 sends the data query request to the database system 300. Upon receiving the data query request, the database system 300 retrieves the transaction data associated with the identifier of the first calculation scenario and the calculation rule indicated by the identifier of the aforementioned calculation rule, and then sends the transaction data of the first calculation scenario and the calculation rule for expected credit loss to the credit management system 200. Thus, the credit management system 200 can obtain the transaction data of the first calculation scenario and the calculation rule for expected credit loss.

[0088] Step 105: The credit management system 200 selects the calculation element table from the transaction data of the first calculation scenario based on the feature table.

[0089] Specifically, the transaction data of the first calculation scenario includes the identifier of the first credit risk mitigation tool, and the feature table includes the identifiers of multiple credit risk mitigation tools and the calculation elements associated with each credit risk mitigation tool indicated by the identifier. Therefore, the credit management system 200 can obtain the calculation elements associated with the first credit risk mitigation tool (hereinafter referred to as "first calculation elements") from the feature table based on the matching relationship between the identifier of the first credit risk mitigation tool and the identifiers of credit risk mitigation tools in the feature table. The first calculation element is the calculation element associated with the credit risk mitigation tool indicated by the matching identifier. Then, the credit management system 200 obtains the value of the first calculation element from the transaction data of the first calculation scenario based on the first calculation element, and then associates and stores the identifier of the first credit risk mitigation tool, the first calculation element, and the value of the first calculation element to obtain a calculation element table. The calculation element table includes the identifier of the first credit risk mitigation tool, the first calculation element, and the value of the first calculation element.

[0090] More specifically, taking the identifier of a first credit risk mitigation tool (hereinafter referred to as "Identifier S") and the identifier of a credit risk mitigation tool in the feature table (hereinafter referred to as "Identifier R") as an example, the credit management system 200 can determine the matching relationship between the two in the following way: The credit management system 200 can use a similarity algorithm to calculate the similarity between Identifier S and Identifier R. When the similarity between Identifier S and Identifier R is greater than or equal to a threshold, it means that Identifier S and Identifier R match; when the similarity between Identifier S and Identifier R is less than the threshold, it means that Identifier S and Identifier R do not match. It should be understood that the similarity algorithm used in this step can be an existing algorithm in the industry that has a better effect on text similarity calculation, such as cosine similarity algorithm, word embedding algorithm (such as Word2Vec, FastText), etc.

[0091] As one possible implementation, the transaction data in the first computing scenario is stored in the form of data tables. The data table recording relevant information about the first credit risk mitigation instrument is called a "source data table," and there can be one or more source data tables. A source data table includes at least one record row. Each record row records relevant information about a credit risk mitigation instrument, such as the identifier of the credit risk mitigation instrument, the elements associated with the credit risk mitigation instrument and their values, and the record row ID.

[0092] Optionally, the calculation element table may also include the ID of the record line or other elements associated with the first credit mitigation instrument, which can be determined according to the actual calculation needs, and this application does not limit this.

[0093] To facilitate understanding of this step, a specific example is provided below, along with the feature table shown in Table 1:

[0094] First, please refer to Table 5, which shows two record lines (i.e., record line 1 and record line 2). Record line 1 records information related to Credit Risk Mitigation Instrument A, specifically including the identifier of Credit Risk Mitigation Instrument A, the name of the letter of credit, the name of the issuing bank, the name of the exporter, the name of the importer, the description of the goods, the date of issuance, the amount of the letter of credit withdrawn, and the ID of record line 1. Record line 2 records information related to Credit Risk Mitigation Instrument B, specifically including the identifier of Credit Risk Mitigation Instrument B, the name of the guarantee, the name of the issuer, the name of the beneficiary, the name of the applicant, the date of issuance, the amount of the guarantee, the currency, and the ID of record line 2.

[0095] Table 5 Source Data Table

[0096]

[0097] Based on the feature table shown in Table 1, the calculation element table shown in Table 6 can be extracted from the source data table shown in Table 5.

[0098] Table 6 Calculation Element Table

[0099] ID Identification of credit risk mitigation instruments Calculation elements Calculate the value of the element 10001 Identifier A Letter of Credit Withdrawal Amount 20000 10002 Identifier B Guarantee Amount 14000 10003 Identifier B Currency types Dollar

[0100] The contents of Table 6 will be introduced next:

[0101] ① Table 6 includes the values ​​of the “ID” field (i.e., “10001”), the “Identifier of Credit Risk Mitigation Instrument” field (i.e., “Identifier A”), the “Letter of Credit Withdrawal Amount”, and the “Letter of Credit Withdrawal Amount” (i.e., “20000”) in record row 1.

[0102] ② Table 6 also includes the values ​​of the “ID” field (i.e., “10002”), the “Identifier of Credit Risk Mitigation Instrument” field (i.e., “Identifier B”), the “Guarantee Amount” field (i.e., “14000”), the “Currency Type” field, and the “Currency Type” field (i.e., “USD”) in record row 2.

[0103] It should be understood that Table 5 is merely an example showing one possible source data table. In practical applications, the source data table may also include more elements associated with credit risk mitigation instrument A and credit risk mitigation instrument B. Similarly, Table 6 is also merely an example showing one possible calculation element table. In practical applications, the calculation element table may also include more elements associated with credit risk mitigation instrument A and credit risk mitigation instrument B, and this application does not limit this.

[0104] Step 106: The credit management system 200 calculates the expected credit loss based on the calculation element table and the calculation factor combination table.

[0105] Specifically, the calculation formula includes multiple calculation factors. The credit management system 200 can determine the values ​​of the calculation factors in the calculation formula based on the matching relationship between the calculation factors in the calculation formula and the calculation elements in the calculation element table and the calculation factors in the calculation factor table, and then obtain the expected credit loss based on the values ​​of the calculation factors in the calculation formula.

[0106] In one implementation, the credit management system 200 can determine the values ​​of the calculation factors in the calculation formula through the following steps:

[0107] Step 1061: The credit management system 200 retrieves the value of the first calculation factor in the calculation formula from the calculation element table based on the matching relationship between the calculation factor in the calculation formula and the calculation element in the calculation element table. The value of the first calculation factor is the value of the matched calculation element.

[0108] Specifically, taking a calculation factor (hereinafter referred to as "calculation factor M") in the calculation formula as an example: The credit management system 200 matches calculation factor M with calculation element N in the calculation element table. If they match, the value of calculation element N is obtained from the calculation result table, and the value of calculation factor M is the value of calculation element N. If they do not match, the system continues to match calculation factor M with the next calculation element in the calculation element table, and so on, until a calculation element that matches calculation factor M is determined, or it is determined that calculation factor M does not match any calculation element in the calculation element table.

[0109] In the above process, the credit management system 200 can determine the matching relationship between the calculation factor M and the calculation element N in the following way: The credit management system 200 uses a similarity algorithm to calculate the similarity between the calculation factor M and the calculation element N. When the similarity between the calculation factor M and the calculation element N is greater than or equal to a threshold, it indicates that the calculation factor M and the calculation element N match; when the similarity between the calculation factor M and the calculation element N is less than the threshold, it indicates that the calculation factor M and the calculation element N do not match. It should be understood that the similarity algorithm used in this step can be an existing algorithm in the industry that has a better effect on text similarity calculation, such as cosine similarity algorithm, word embedding algorithm (such as Word2Vec, FastText), etc.

[0110] Step 1062: The credit management system 200, based on the matching relationship between the calculation factors in the calculation formula and the calculation factors in the calculation factor combination table, obtains the value of the second calculation factor in the calculation formula from the calculation factor combination table. The value of the second calculation factor is the value of the matched calculation factor.

[0111] Specifically, the credit management system 200 obtains the calculation element associated with the second calculation factor (hereinafter referred to as "second calculation element") from the calculation factor combination table based on the matching relationship between the calculation factor in the calculation formula and the calculation factor in the calculation factor combination table. Then, it obtains the value of the second calculation element from the transaction data of the first calculation scenario based on the second calculation element, and then obtains the value of the second calculation factor from the calculation factor combination table based on the value of the second calculation element.

[0112] Taking a calculation factor (hereinafter referred to as "calculation factor P") in the calculation formula and a calculation factor (hereinafter referred to as "calculation factor Q") in the calculation factor combination table as an example, the credit management system 200 can determine the matching relationship between the two in the following way: The credit management system 200 can use a similarity algorithm to calculate the similarity between the above-mentioned calculation factor P and calculation element Q. When the similarity between calculation factor P and calculation element Q is greater than or equal to a threshold, it indicates that calculation factor P and calculation element Q match; when the similarity between calculation factor P and calculation element Q is less than the threshold, it indicates that calculation factor P and calculation element Q do not match. It should be understood that the similarity algorithm used in this step can be an existing algorithm in the industry that has a better effect on text similarity calculation, such as cosine similarity algorithm, word embedding algorithm (such as Word2Vec, FastText), etc.

[0113] To facilitate understanding, a specific example is provided below, along with Tables 4 and 5, to illustrate step 106:

[0114] In this example, the transaction data for the first calculation scenario includes the following information: the borrower's credit rating is "A", the contract age is "7-12 months", the default probability corresponds to a duration of "1 year", and the information recorded in Table 5.

[0115] The calculation formula includes the following calculation factors: the amount of the letter of credit exit (hereinafter referred to as "calculation factor 1"), the amount of the guarantee (hereinafter referred to as "calculation factor 2"), the currency of the guarantee amount (hereinafter referred to as "calculation factor 3"), the loss rate in default (hereinafter referred to as "calculation factor 4"), and the probability of default (hereinafter referred to as "calculation factor 5").

[0116] The credit management system 200 matches the calculation factors in the calculation formula with the calculation element table shown in Table 5, and determines that the value of calculation factor 1 is "20000", the value of calculation factor 2 is "14000", and the value of calculation factor 3 is "currency type".

[0117] The credit management system 200 also matches the calculation factors in the calculation formula with the calculation factor combination table shown in Table 4, and determines that calculation factor 4 matches "default loss rate" in the calculation factor table, and calculation factor 5 matches "default probability" in the calculation factor table.

[0118] For calculation factor 4, the credit management system 200 determines that the calculation elements associated with "loss of default rate" are "credit rating" and "contract aging" according to the calculation factor combination table. Then, based on the determined calculation elements, it obtains the value of "credit rating" as "A" and the value of "contract aging" as "7-12 months" from the transaction data of the first calculation scenario. Then, based on the values ​​of "loss of default rate", "credit rating" ("A"), "contract aging" ("7-12 months"), it obtains the value of "loss of default rate" as "2.36%" from the calculation factor table, that is, the value of calculation factor 4 is "2.36%".

[0119] For calculation factor 5, the credit management system 200 determines the calculation elements associated with "default probability" as "credit rating" and "duration" according to the calculation factor table. Then, based on the determined calculation elements, it obtains the value of "credit rating" as "A" and the value of "duration" as "1 year" from the transaction data of the first calculation scenario. Then, based on the values ​​of "default probability", "credit rating" ("A"), "duration" ("1 year"), and "duration" ("0.06%), it obtains the value of "default probability" as "0.06%" from the calculation factor table, that is, the value of calculation factor 5 is "0.06%".

[0120] Thus, the values ​​of the five calculation factors in the formula are determined. The credit management system 200 can then perform calculations on the values ​​of these five factors according to the operational rules defined in the formula, thereby obtaining the expected credit loss.

[0121] Step 107: The credit management system 200 returns the calculation result of the expected credit loss to the client 100. Accordingly, the client 100 receives the calculation result of the expected credit loss.

[0122] In the process of calculating expected credit loss described in steps 101 to 107 above, the credit management system 200 can automatically filter the data required for calculating expected credit loss from transaction data according to preset calculation rules, and complete the calculation of expected credit loss. Therefore, the method provided in this application can automate the calculation of expected credit loss and improve the calculation efficiency of expected credit loss. In addition, when business requirements change, users only need to update the feature table, calculation factor table, and calculation formula accordingly. The software code for calculating expected credit loss does not need to be changed, which improves the scalability and maintainability of the software code and reduces the calculation cost of expected credit loss.

[0123] The above text combined Figures 1 to 3 This application details the method for calculating expected credit loss, and the following section provides an example. Figure 4 The structure of the credit management system 200 that implements the above method is described.

[0124] See Figure 4 , Figure 4 A schematic diagram of the structure of a credit management system 200 is shown. It should be understood that... Figure 4 This example merely illustrates one possible way to divide the structure of the credit management system 200. In practical applications, the credit management system 200 can be divided in other ways, and this embodiment does not limit this. Figure 4 As shown, the credit management system 200 includes a configuration module 201, a sending module 202, an acquisition module 203, and a calculation module 204.

[0125] Configuration module 201 is used to perform step 101 above. Sending module 202 is used to perform steps 102 and 107 above. Acquisition module 203 is used to perform the calculation request for receiving expected credit loss in step 103 above, as well as the related steps in step 104 above. Calculation module 204 is used to perform steps 105 and 106 above.

[0126] The configuration module 201, sending module 202, acquiring module 203, and calculation module 204 can all be implemented in software or in hardware. For example, the implementation of the configuration module 201 will be described below. Similarly, the implementation of the sending module 202, acquiring module 203, and calculation module 204 can refer to the implementation of the configuration module 201.

[0127] As an example of a software functional unit, configuration module 201 may include code running on a compute instance. The compute instance may include at least one of a physical host (compute device), a virtual machine, or a container. Further, the compute instance may be one or more. For example, configuration module 201 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed within the same region or in different regions. Further, the multiple hosts / virtual machines / containers used to run the code may be distributed within the same availability zone (AZ) or in different AZs, each AZ including one or more geographically proximate data centers. Typically, a region may include multiple AZs.

[0128] Similarly, multiple hosts / virtual machines / containers used to run this code can be distributed within the same Virtual Private Cloud (VPC) or across multiple VPCs. Typically, a VPC is set up within a region. Communication between two VPCs within the same region, as well as between VPCs in different regions, requires a communication gateway to be set up within each VPC to enable interconnection between VPCs.

[0129] As an example of a hardware functional unit, configuration module 201 may include at least one computing device, such as a server. Alternatively, configuration module 201 may also be a device implemented using a central processing unit (CPU), an application-specific integrated circuit (ASIC), or a programmable logic device (PLD). The aforementioned PLD may be implemented using a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), generic array logic (GAL), a data processing unit (DPU), a neural network processing unit (NPU), a system-on-chip (SoC), an offload card, an accelerator card, or any combination thereof.

[0130] The multiple computing devices included in configuration module 201 can be distributed in the same region or in different regions. Similarly, the multiple computing devices included in configuration module 201 can be distributed in the same Availability Zone (AZ) or in different AZs. Likewise, the multiple computing devices included in configuration module 201 can be distributed in the same VPC or in multiple VPCs. Furthermore, the multiple computing devices included in configuration module 201 can be any combination of computing devices such as servers, ASICs, PLDs, CPLDs, FPGAs, GALs, DPUs, NPUs, SoCs, offloading cards, and accelerator cards.

[0131] It should be noted that, in other embodiments, the configuration module 201 can be used to execute any step in the above-described method for calculating expected credit loss, the sending module 202 can be used to execute any step in the above-described method for calculating expected credit loss, the acquisition module 203 can be used to execute any step in the above-described method for calculating expected credit loss, and the calculation module 204 can be used to execute any step in the above-described method for calculating expected credit loss. The steps implemented by the configuration module 201, sending module 202, acquisition module 203, and calculation module 204 can be specified according to actual needs. By implementing different steps in the above-described method for calculating expected credit loss through the configuration module 201, sending module 202, acquisition module 203, and calculation module 204, all functions of the credit management system 200 can be realized.

[0132] This application also provides a computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. The computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0133] Figure 5 A schematic diagram of the structure of the computing device provided in this application is shown as an example. Figure 5 As shown, the computing device 400 includes a bus 401, a processor 402, a memory 403, and a communication interface 404, and the processor 402, the memory 403, and the communication interface 404 communicate with each other via the bus 401. It should be understood that this application does not limit the number of processors 402 and memory 403 in the computing device 400; for simplicity, Figure 5 The description will be based on an example of a processor 402 and a memory 403.

[0134] Bus 401 can be a Peripheral Component Interconnect Express (PCIe) bus, an Extended Industry Standard Architecture (EISA) bus, a Unified Bus (Ubus or UB), a Compute Express Link (CXL) bus, a Cache Coherent Interconnect for Accelerators (CCIX) bus, etc. The Unified Bus is also known as the Lingqu Bus. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 5 The bus 401 is represented by only one line, but this does not mean that the computing device 400 has only one bus or one type of bus. The bus 401 may include a path for transmitting information between various components of the computing device 400 (e.g., processor 402, memory 403, and communication interface 404).

[0135] Processor 402 may include any one or more computing units with computing capabilities, such as CPU, graphics processing unit (GPU), microprocessor (MP), digital signal processor (DSP), ASIC, FPGA, CPLD, NPU, SoC, offload card, accelerator card, etc.

[0136] Memory 403 may include volatile memory, such as random access memory (RAM). Memory 403 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD). Furthermore, memory 403 may also be implemented using storage class memory (SCM), phase change memory (PCM), or other types of storage media.

[0137] It is worth noting that the same type of storage medium can be configured in the same computing device to realize the function of memory 403, or two or more types of storage media can be configured to realize the function of memory 403. This application does not limit this.

[0138] The memory 403 stores executable program code. The processor 402 executes this executable program code to implement the functions of the configuration module 201, the sending module 202, the acquisition module 203, and the calculation module 204, respectively, thereby realizing the above-mentioned method for calculating expected credit loss. That is to say, the memory 403 stores instructions for executing the above-mentioned method for calculating expected credit loss.

[0139] The communication interface 404 uses transceiver modules such as, but not limited to, network interface cards and transceivers to enable communication between the computing device 400 and other devices or communication networks. For example, the computing device 400 communicates with the client 100 through the communication interface 404.

[0140] It should be understood that the computing device 400 provided according to this application can correspond to the computing device provided in this application. Figure 1 The credit management system 200 shown can be used to execute the system provided in this application. Figure 3 The credit management system 200 and the various modules in the computing device 400 described above and other operations and / or functions are respectively implemented to achieve Figure 3 For the sake of brevity, the corresponding processes of each method in the code will not be elaborated here.

[0141] This application also provides a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. The computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0142] Figure 6 An exemplary schematic diagram of the computing device cluster provided in this application is shown. Figure 6 As shown, the computing device cluster 500 includes at least one computing device 400. The memory 403 of one or more computing devices 400 in the computing device cluster 500 may store identical instructions for performing the aforementioned method for calculating expected credit loss.

[0143] In one implementation, the memory 403 of one or more computing devices 400 in the computing device cluster 500 may also store partial instructions for executing the above-mentioned expected credit loss calculation method, that is, a combination of one or more computing devices 400 can jointly execute the above-mentioned expected credit loss calculation method.

[0144] It should be noted that the memory 403 in different computing devices 400 within the computing device cluster 500 can also store different instructions, which are used to execute certain functions of the credit management system 200. That is, the instructions stored in the memory 403 of different computing devices 400 can implement the functions of one or more modules among the configuration module 201, the sending module 202, the acquisition module 203, and the computing module 204.

[0145] In one implementation, multiple computing devices 400 in a computing device cluster 500 can be connected via a network, which can be a wide area network or a local area network, etc. Figure 7 One possible implementation is shown. For example... Figure 7 As shown, computing devices 400A and 400B are connected via a network. Specifically, they are connected to the network through communication interfaces in computing devices 400A and 400B. In this implementation, the memory 403 in both computing device 400A and computing device 400B stores instructions for executing the functions of configuration module 201, sending module 202, acquisition module 203, and computing module 204.

[0146] Figure 7 The connection method between the computing device cluster 500 shown can be considered in high-concurrency computing scenarios. For example, if computing device 400A receives a large number of computing requests in the same time period, in order to respond to these computing requests as quickly as possible, computing device 400A can offload some computing tasks to computing device 400B. It should be understood that... Figure 7 The functions of computing device 400A can also be performed by multiple computing devices 400, and similarly, the functions of computing device 400B can also be performed by multiple computing devices 400.

[0147] It should be understood that the computing device cluster 500 provided according to this application may correspond to the computing device cluster 500 provided in this application. Figure 1 The credit management system 200 shown can be used to execute the system provided in this application. Figure 3 The credit management system 200 and the various modules in the computing device cluster 500 described above and other operations and / or functions are respectively implemented to achieve Figure 3 For the sake of brevity, the corresponding processes of each method in the code will not be elaborated here.

[0148] This application also provides a computer program product containing instructions. The computer program product may be a software or program product containing instructions that can run on a computing device or be stored on any available medium. When the computer program product is run on a computing device, it causes the computing device to perform the method for calculating the expected credit loss described above.

[0149] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium capable of being stored by a computing device, or a data storage device such as a data center containing one or more available media. The aforementioned available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives), etc. The computer-readable storage medium includes instructions that instruct the computing device to perform the method for calculating the expected credit loss described above.

[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of this application.

Claims

1. A method for calculating expected credit loss, applied to a credit management system, characterized in that, The method includes: Obtain a calculation request for the expected credit loss, the calculation request including an identifier of a first calculation scenario; Based on the identifier and feature table of the first calculation scenario, a calculation element table is selected from the transaction data of the first calculation scenario; wherein, the feature table includes calculation elements associated with multiple credit risk mitigation tools, and the calculation element table includes the first calculation element associated with the first credit risk mitigation tool involved in the transaction data and the value of the first calculation element; The expected credit loss is calculated based on the aforementioned calculation element table.

2. The method according to claim 1, characterized in that, The step of filtering the calculation element table from the transaction data associated with the first calculation scenario based on the identifier and feature table of the first calculation scenario includes: The transaction data is obtained based on the identifier of the first calculation scenario; Based on the matching relationship between the identifier of the first credit risk mitigation tool in the transaction data and the identifier of the credit risk mitigation tool in the feature table, the first calculation element is obtained from the feature table, and the first calculation element is the calculation element associated with the credit risk mitigation tool indicated by the matching identifier. The value of the first calculation element is obtained from the transaction data based on the first calculation element.

3. The method according to claim 1 or 2, characterized in that, The method further includes: A calculation factor combination table is generated based on multiple calculation factor tables configured by the user; wherein, each calculation factor table corresponds to a calculation factor in the calculation formula of the expected credit loss, and the calculation factor table includes the calculation factor and the value of the calculation factor; the calculation factor combination table is the combination result determined based on the multiple calculation factor tables; The step of calculating the expected credit loss based on the calculation element table includes: The expected credit loss is calculated based on the calculation element table and the calculation factor combination table.

4. The method according to claim 3, characterized in that, The step of calculating the expected credit loss based on the calculation element table and the calculation factor configuration table includes: Based on the matching relationship between the calculation factor in the calculation formula and the calculation element in the calculation element table, the value of the first calculation factor in the calculation formula is obtained from the calculation element table, and the value of the first calculation factor is the value of the matching calculation element. Based on the matching relationship between the calculation factors in the calculation formula and the calculation factors in the calculation factor combination table, the value of the second calculation factor in the calculation formula is obtained from the calculation factor combination table, and the value of the second calculation factor is the value of the matching calculation factor. The expected credit loss is obtained based on the values ​​of the first calculation factor and the second calculation factor.

5. The method according to claim 4, characterized in that, The calculation factor combination table also includes the calculation elements associated with the calculation factors and multiple values ​​of the calculation elements; The step of obtaining the value of the second calculation factor in the calculation formula from the calculation factor combination table based on the matching relationship between the calculation factors in the calculation formula and the calculation factors in the calculation factor combination table includes: Based on the matching relationship between the calculation factors in the calculation formula and the calculation factors in the calculation factor combination table, the second calculation element associated with the second calculation factor is obtained from the calculation factor combination table; The value of the second calculation element is obtained from the transaction data based on the second calculation element; The value of the second calculation factor is obtained from the calculation factor combination table based on the value of the second calculation element.

6. The method according to claim 3, characterized in that, The method further includes: Based on the user's configuration, at least one of the feature table, the calculation factor combination table, and the calculation formula is updated, and the updated feature table, calculation factor combination table, and calculation formula meet the new business requirements.

7. A credit management system, characterized in that, The system includes: The acquisition module is used to acquire a calculation request for expected credit loss, wherein the calculation request includes an identifier of a first calculation scenario; The calculation module is used to filter out a calculation element table from the transaction data of the first calculation scenario based on the identifier and feature table of the first calculation scenario; wherein, the feature table includes calculation elements associated with multiple credit risk mitigation tools, and the calculation element table includes a first calculation element associated with a first credit risk mitigation tool involved in the transaction data and the value of the first calculation element; and calculates the expected credit loss based on the calculation element table.

8. The system according to claim 7, characterized in that, The calculation module is configured to obtain the transaction data based on the identifier of the first calculation scenario; obtain the first calculation element from the feature table based on the matching relationship between the identifier of the first credit risk mitigation tool in the transaction data and the identifier of the credit risk mitigation tool in the feature table, wherein the first calculation element is the calculation element associated with the credit risk mitigation tool indicated by the matching identifier; and obtain the value of the first calculation element from the transaction data based on the first calculation element.

9. The system according to claim 7 or 8, characterized in that, The system also includes a configuration module. The configuration module is used to generate a calculation factor combination table based on multiple calculation factor tables configured by the user; wherein, one calculation factor table corresponds to one calculation factor in the calculation formula of the expected credit loss, and the calculation factor table includes the calculation factor and the value of the calculation factor; the calculation factor combination table is the combination result determined based on the multiple calculation factor tables; The calculation module is used to calculate the expected credit loss based on the calculation element table and the calculation factor combination table.

10. The system according to claim 9, characterized in that, The calculation module is used to obtain the value of the first calculation factor in the calculation formula from the calculation element table according to the matching relationship between the calculation factor in the calculation formula and the calculation element in the calculation element table, wherein the value of the first calculation factor is the value of the matching calculation element. Based on the matching relationship between the calculation factors in the calculation formula and the calculation factors in the calculation factor combination table, the value of the second calculation factor in the calculation formula is obtained from the calculation factor combination table, and the value of the second calculation factor is the value of the matching calculation factor. The expected credit loss is obtained based on the values ​​of the first calculation factor and the second calculation factor.

11. The system according to claim 10, characterized in that, The calculation factor combination table also includes the calculation elements associated with the calculation factors and multiple values ​​of the calculation elements; The calculation module is used to obtain the second calculation element associated with the second calculation factor from the calculation factor combination table based on the matching relationship between the calculation factor in the calculation formula and the calculation factor in the calculation factor combination table. The value of the second calculation element is obtained from the transaction data based on the second calculation element; The value of the second calculation factor is obtained from the calculation factor combination table based on the value of the second calculation element.

12. The system according to claim 9, characterized in that, The configuration module is further configured to update at least one of the feature table, the calculation factor combination table, and the calculation formula according to the user's configuration, wherein the updated feature table, calculation factor combination table, and calculation formula meet new business requirements.

13. A computing device, characterized in that, The computing device includes a processor and memory; The processor is configured to execute instructions stored in the memory to cause the computing device to perform the operational steps of the method as described in any one of claims 1 to 6.

14. A computing device cluster, characterized in that, It includes at least one computing device, each computing device including a processor and memory; The processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device to cause the cluster of computing devices to perform the operational steps of the method as described in any one of claims 1 to 6.

15. A computer program product containing instructions, characterized in that, When the instruction is executed by the computing device cluster, the computing device cluster causes the computing device cluster to perform the operation steps of the method as described in any one of claims 1 to 6.

16. A computer-readable storage medium, characterized in that, It includes computer program instructions, which, when executed by a cluster of computing devices, perform the operational steps of the method as described in any one of claims 1 to 6.

Images