Carbon asset pledge financing post-loan risk early warning method and device, electronic equipment and storage medium

Abstract

The invention discloses a carbon asset pledge financing post-loan risk early warning method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining carbon asset related data; and inputting the carbon asset related data into the prediction model, extracting to obtain carbon asset related risk feature data, and obtaining a risk prediction result after enterprise carbon asset pledge financing loan, the prediction model being obtained by using multiple groups of data through machine learning training. Each group of data in the multiple groups of data comprises normal operation enterprise data and default enterprise data; and according to the risk prediction result after the enterprise carbon asset pledge financing loan, sending a risk early warning to a downstream system under the condition that the risk prediction result is higher than a threshold value. According to the invention, the effectiveness of carbon asset pledge financing post-loan risk early warning is improved.

Description

Technical Field

[0001] The present application relates to the technical field of post-credit risk assessment and risk warning, and in particular to a post-credit risk warning method, device, electronic equipment, and storage medium for carbon asset mortgage financing. Background Art

[0002] Carbon assets refer to new types of assets generated by the carbon emission trading mechanism.

[0003] Against the backdrop of the continuous development of the carbon trading market, carbon asset mortgage financing continues to innovate. However, financial institutions' post-loan management of mortgage financing still relies on the evaluation of historical information on corporate industry and commerce, taxation, finance, etc.

[0004] In relevant technologies, insufficient attention is paid to the characteristics of carbon assets, a new type of collateral. There are deficiencies such as the post-loan management method not taking into account the particularity of carbon assets, the quality of carbon asset-related data is poor, the risk warning model is not interpretable, and it cannot respond to market changes and risk dynamics in a timely manner. Summary of the invention

[0005] The embodiments of the present application provide a carbon asset mortgage financing post-loan risk warning method, device, electronic equipment, and storage medium to improve the effectiveness of carbon asset mortgage financing post-loan risk warning.

[0006] The present application embodiment adopts the following technical solutions:

[0007] In a first aspect, an embodiment of the present application provides a risk warning method, wherein the method comprises:

[0008] Obtain data related to carbon assets;

[0009] The carbon asset-related data is input into the prediction model, and the carbon asset-related risk characteristic data is extracted to obtain the post-loan risk prediction result of the enterprise carbon asset mortgage financing, wherein the prediction model is obtained by machine learning training using multiple sets of data, and each set of data in the multiple sets of data includes data of normally operating enterprises and data of defaulting enterprises;

[0010] Based on the post-loan risk prediction results of the enterprise's carbon asset mortgage financing, when the risk prediction result is higher than the threshold, a risk warning will be issued to the downstream system.

[0011] In some embodiments, the carbon asset related data includes:

[0012] Connecting with the first type of node through an application programming interface to obtain external business information data of the enterprise, wherein the first type of node includes at least one of the following: an enterprise node, a mortgage registration node, a registration node, and a market transaction node;

[0013] Connecting to the second type of nodes through an application programming interface to obtain internal business information data of the enterprise, wherein the second type of nodes at least includes a bank node;

[0014] Based on the automated process, updated business information data is obtained regularly, and the enterprise data is updated in real time.

[0015] In some embodiments, the carbon asset related data further includes:

[0016] When an enterprise uses carbon assets as collateral for a loan, basic enterprise information, carbon asset ownership information, carbon asset mortgage registration information, and carbon market transaction information are obtained according to the enterprise node, the carbon asset mortgage registration node, the carbon asset registration node, and the market transaction node;

[0017] as well as

[0018] The enterprise loan information is obtained according to the bank node.

[0019] In some embodiments, the inputting of the carbon asset related data into the prediction model, extracting the carbon asset related risk characteristic data, and obtaining the post-loan risk prediction result of the enterprise carbon asset mortgage financing includes:

[0020] After the carbon asset related data is input into the prediction model, the carbon asset market risk characteristics, carbon asset behavior risk characteristics, and carbon asset target risk characteristics are extracted according to the prediction model;

[0021] Outputting the quantitative results of the enterprise post-loan risk indicators through the prediction model according to the carbon asset market risk characteristics, the carbon asset behavior risk characteristics and the carbon asset target risk characteristics;

[0022] Perform risk grading based on the quantitative results of the enterprise's post-loan risk indicators to obtain the enterprise's carbon asset mortgage financing post-loan risk prediction results;

[0023] in,

[0024] The carbon asset market risk is used to characterize the carbon asset market risk of changes in the market price of carbon asset collateral to the collateral guarantee balance;

[0025] The carbon asset behavior risk characteristics are used to characterize the carbon asset behavior risks generated by the lending enterprise when the key changes occur in the carbon market;

[0026] The carbon asset target risk characteristics are used to characterize the carbon asset target risks of the enterprise's carbon assets as collateral during the period of existence.

[0027] In some embodiments, according to the post-loan risk prediction result of the carbon asset mortgage financing of the enterprise, when the risk prediction result is higher than a threshold, a risk warning is issued to the downstream system, including:

[0028] According to the post-loan risk prediction results of the carbon asset mortgage financing of the enterprise, the corresponding risk level is obtained to provide a post-loan risk warning for the enterprise;

[0029] Display the enterprise's carbon asset market risk data, carbon asset behavior risk data, carbon asset target risk data, and final risk score and risk level information;

[0030] When the enterprise's risk level reaches the preset high risk, risk warning information is delivered in real time through any one or more methods such as SMS, email, and application push.

[0031] In some embodiments, the method further comprises:

[0032] The post-loan default probability and risk level output from the post-loan risk prediction results of the enterprise's carbon asset mortgage financing are transmitted to the downstream credit system to provide post-loan risk warning to the enterprise.

[0033] In some embodiments, the method further comprises:

[0034] Adjusting the acquired enterprise data into a common format through a standardization process;

[0035] storing the processed data in a common format in a network database;

[0036] The frequency of data consumption requests is set, and the enterprise data is re-queried and acquired at regular intervals, and the storage of the network database is updated.

[0037] In a second aspect, the embodiment of the present application further provides a carbon asset mortgage financing post-loan risk warning device, wherein the device comprises:

[0038] Data collection module, for carbon asset related data;

[0039] A feature extraction module is used to input the carbon asset related data into the prediction model, extract the carbon asset related risk feature data, and obtain the post-loan risk prediction result of the enterprise carbon asset mortgage financing, wherein the prediction model is obtained by machine learning training using multiple sets of data, and each set of data in the multiple sets of data includes data of normal operating enterprises and data of defaulting enterprises;

[0040] The early warning module is used to issue a risk warning to the downstream system based on the post-loan risk prediction result of the carbon asset mortgage financing of the enterprise, when the risk prediction result is higher than the threshold.

[0041] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to perform the above method.

[0042] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple application programs, the electronic device executes the above method.

[0043] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects: by obtaining carbon asset related data, and then inputting the carbon asset related data into the prediction model, extracting the carbon asset related risk characteristic data, and obtaining the post-loan risk prediction result of the enterprise's carbon asset mortgage financing. Therefore, according to the post-loan risk prediction result of the enterprise's carbon asset mortgage financing, when the risk prediction result is higher than the threshold, a risk warning is issued to the downstream system. Through the above method, by inputting the carbon asset related data into the prediction model, extracting the carbon asset related risk data, and obtaining the post-loan risk prediction result of the enterprise's carbon asset mortgage financing, a real-time post-loan risk warning is realized based on data-driven and machine learning algorithms, thereby improving the effectiveness of the post-loan risk warning of carbon asset pledge financing. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0045] Figure 1 This is a flow chart of the post-loan risk warning method for carbon asset mortgage financing in the embodiment of this application;

[0046] Figure 2 This is a flow chart of a post-loan risk warning method for carbon asset pledge financing in a post-loan risk warning method for carbon asset pledge financing in an embodiment of the present application;

[0047] Figure 3 This is a schematic diagram of the multi-dimensional data collection process for carbon asset pledge financing post-loan risk warning in the carbon asset pledge financing post-loan risk warning method in the embodiment of the present application;

[0048] Figure 4 This is a functional structure diagram of a data processing program in a carbon asset mortgage financing post-loan risk early warning method in an embodiment of the present application;

[0049] Figure 5 This is a schematic diagram of the structure of the carbon asset mortgage financing post-loan risk warning device in the embodiment of the present application;

[0050] Figure 6 This is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0052] The technical terms involved in this application are as follows:

[0053] Carbon assets: new types of assets generated by the carbon emission trading mechanism.

[0054] Carbon trading market: It is a market artificially created by the government to control emissions from energy-consuming enterprises. It is an institutional design that trades carbon dioxide emission rights as a commodity. Usually, the government determines a total carbon emission amount and allocates carbon emission quotas to enterprises according to certain rules. When fulfilling the contract, if the enterprise has a surplus of initial quotas (quotas > actual emissions), it can sell quotas in the carbon market; if the quota is insufficient (actual emissions > quotas), it needs to purchase carbon emission quotas or CCERs for fulfillment and settlement. At the same time, some enterprises adopt energy-saving and emission reduction technologies, and their final carbon emissions are lower than the quotas they have obtained. They can sell excess quotas through the carbon trading market to obtain economic benefits.

[0055] Carbon market financing tools: Carbon financial products that use carbon assets as the basis for various types of financing. They mainly include carbon bonds, carbon asset mortgage financing, carbon asset repurchase, and carbon asset custody.

[0056] CCER: Nationally certified voluntary emission reductions refer to the greenhouse gas emission reduction effects of renewable energy, forestry carbon sinks, methane utilization and other projects in my country that are quantitatively verified and registered in the national greenhouse gas voluntary emission reduction trading registration system.

[0057] CEA: National carbon emission quota refers to the total amount of greenhouse gas emissions (measured in carbon dioxide equivalent) into the atmosphere obtained by enterprises within a certain period of time, as approved by the government department in charge, in order to encourage enterprises to save energy and reduce emissions, and actively promote the green and low-carbon transformation of industrial structure. In addition to being initially issued by the government department in charge, carbon emission quotas can also be obtained from carbon emission trading markets.

[0058] In order to support the low-carbon transformation of the economy, the carbon trading market has become one of the important institutional arrangements for achieving emission reduction targets. There are two types of basic products in the carbon trading market, namely carbon quotas (China Emission Allowance, CEA) and Chinese Certified Emission Reduction (CCER). Because they have economic attributes and can bring asset returns to enterprises, financial loan products with carbon assets as collateral have emerged.

[0059] Carbon asset mortgage financing refers to a financing contract in which the holder of a carbon asset pledges its carbon asset as collateral to the fund provider to obtain a loan, and then releases the collateral by paying the principal and interest when the loan is due. If the borrower is unable to repay the loan when it is due, the fund provider has the right to deal with the collateral to make up for the loan loss. Carbon assets have the characteristics of scarcity, substitutability, free allocation of carbon quotas, and due performance and liquidation. They are significantly different from traditional collateral in terms of management and operation mechanisms, market trading mechanisms, and price formation mechanisms.

[0060] At present, banks generally use qualitative and quantitative methods for post-loan management methods, integrating enterprise operation, finance, industry and commerce, taxation, finance and other information to monitor and evaluate the borrower, loan purpose, repayment account, guarantee measures and other contents. Among them, the monitoring of collaterals focuses on value changes, ownership of rights, and status of collaterals. When carbon assets are used as new collateral, the post-loan management methods of banks have the shortcomings of not considering the particularity of carbon assets, poor quality of carbon asset-related data, weak interpretability of risk warning models, and failure to respond to market changes and risk dynamics in a timely manner.

[0061] To solve the above problems, the carbon asset mortgage financing post-loan risk warning method in the embodiment of the present application, taking into account the particularity of carbon assets, obtains enterprise carbon asset-related information in real time through multi-dimensional, and performs post-loan real-time risk warning based on data-driven and machine learning algorithms to improve the effectiveness of carbon asset mortgage financing post-loan risk warning. Considering that banking business has the characteristics of complex processes, multiple related business lines, and strict risk control requirements, in this context, with the help of data-driven integration of multi-dimensional data, combined with machine learning algorithms to improve the effectiveness of model prediction, it can effectively solve the post-loan risk management problem of carbon asset mortgage financing.

[0062] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0063] The present application embodiment provides a carbon asset mortgage financing post-loan risk warning method, such as Figure 1 As shown, a schematic diagram of the process flow of the carbon asset mortgage financing post-loan risk warning method in an embodiment of the present application is provided, and the method at least includes the following steps S110 to S130:

[0064] Step S110, obtaining carbon asset related data.

[0065] like Figure 2 As shown, the target enterprise's business data is obtained, the data format and standard are unified, the enterprise's initial data is formed and stored separately.

[0066] After authorization by the enterprise, the original data such as enterprise basic information, carbon asset ownership information, carbon asset mortgage registration information, carbon market transaction information, etc. are obtained from external enterprise nodes, carbon asset registration nodes, carbon asset mortgage registration nodes, and market transaction nodes through API (application programming interface) technology, and enterprise loan information is obtained from internal bank nodes.

[0067] It should be noted that when the acquired data is stored and transmitted, an encrypted channel is used for data transmission to reduce the risk of data leakage.

[0068] It should be noted that enterprise data of the same dimensions are collected and obtained during both the training and verification stages of the prediction model.

[0069] Step S120, inputting the carbon asset related data into the prediction model, extracting the carbon asset related risk characteristic data, and obtaining the post-loan risk prediction result of the enterprise carbon asset mortgage financing, wherein the prediction model is obtained by machine learning training using multiple sets of data, and each set of data in the multiple sets of data includes normal operating enterprise data and default enterprise data.

[0070] The prediction model uses multiple sets of data for machine learning training. Each set of data includes data of normal operating companies and data of defaulting companies. For the defaulting companies and normal operating companies, industry experts are organized to label their risk levels and score labels for normal operating and non-defaulting companies, further increasing the amount of training data, and finally storing the integrated data in the intranet database.

[0071] Carbon assets with carbon quotas and national certified emission reductions as the target have become new assets for enterprises. In the business scenario where banks accept enterprises to use such assets as collateral for financing, in the specific implementation, the prediction model is trained by characterizing the data during the training phase of the model. In the verification phase of the model, it is necessary to extract the risk data related to carbon assets to evaluate the post-loan risk prediction results of carbon asset mortgage financing.

[0072] It can be understood that when processing data characterization, fine feature extraction is performed on three types of indicators covering the main post-loan risks of carbon asset pledge financing. The three types of indicators include: carbon asset market risk, carbon asset behavioral risk, and carbon asset target risk.

[0073] like Figure 2 As shown in the figure, the initial data of the enterprise is processed to form the characteristic data of the enterprise's carbon assets, and the risk warning indicators are calculated to prepare for the next step of the risk quantification calculation model. According to the characteristic data of carbon assets, the enterprise's post-loan risk data is labeled and collected, and a quantitative prediction model of the enterprise's post-loan credit indicators based on the logistic regression algorithm is constructed, and a loss function is constructed for optimization training.

[0074] The indicator data is characterized to obtain the characteristic functions of the enterprise's carbon assets. Compared with the original data, these characteristics can more intuitively and accurately reflect the market risk, behavioral risk and target risk of the enterprise using carbon assets as collateral, and also provide accurate data for the next step of model calculation.

[0075] Step S130, based on the post-loan risk prediction result of the carbon asset mortgage financing of the enterprise, when the risk prediction result is higher than a threshold, a risk warning is issued to the downstream system.

[0076] like Figure 2 As shown, based on the obtained quantitative prediction model of enterprise post-loan risk indicators, the risk warning data is sent to the business system for the dynamic forecast period risks of the concerned enterprises, business warnings are made and post-loan management measures are responded.

[0077] According to the risk scoring results, set thresholds and output levels: According to the set scoring intervals or standards, set multiple scoring thresholds. For example, the risk level can be finally divided into five intervals, corresponding to the five levels of "very low risk", "low risk", "medium risk", "high risk" and "very high risk".

[0078] Through the above method, by inputting the carbon asset related data into the prediction model, the carbon asset related risk characteristic data is extracted, and the post-loan risk prediction result of the enterprise carbon asset mortgage financing is obtained. Based on data-driven and machine learning algorithms, real-time post-loan risk warning is realized, thereby improving the effectiveness of post-loan risk warning of carbon asset pledge financing.

[0079] Through the above method, the post-loan risk level is assessed in real time through the prediction model, and an early warning mechanism is designed. Finally, the risk warning information is transmitted in real time through multiple message channels.

[0080] Different from related technologies, in terms of risk control models, the existing post-loan risk warning models mainly focus on historical information of enterprises in terms of industry and commerce, taxation, finance, etc., and do not fully consider the characteristics of carbon assets including market fluctuations, corporate behavior, and changes in the underlying assets. The model is not extensible and robust enough, and cannot provide real-time warning and control for carbon asset risks, and cannot effectively support this business scenario. Through the above method, a multi-dimensional carbon asset pledge financing post-loan risk indicator system construction method is provided: for directly obtained multi-source data, a multi-dimensional indicator system is constructed in combination with the characteristics of the pledge financing post-loan management business scenario, including: characterizing the carbon asset market risk of changes in the market price of carbon asset pledges to the pledged balance, characterizing the carbon asset behavior risk generated by the key changes in the behavior of the loan enterprise in the carbon market, and characterizing the carbon asset target risk of the enterprise's carbon assets as collateral during the period of existence.

[0081] Different from related technologies, in terms of data sources, the existing data sources used for carbon asset risk assessment are not representative enough, the risk timeliness is delayed, and the feature data is not complete enough, resulting in poor overall data quality and large deviations in post-loan risk assessment, which ultimately leads to biased model training and poor interpretability. Through the above method, a real-time acquisition method for carbon asset business data based on multi-source data fusion is provided: through API interface and process automation technology, carbon asset business data is acquired and updated in real time through multiple channels, and market changes and risk dynamics are obtained in a timely manner. Technical features include: the multiple channels connected by API include mortgage registration nodes, registration nodes, market transaction nodes, enterprise nodes, and bank nodes; process automation can obtain the latest business information data through multiple channels on a regular basis, including carbon asset-related information and bank business data.

[0082] In one embodiment of the present application, the acquisition of carbon asset related data includes: connecting with a first type of node through an application programming interface to acquire external business information data of the enterprise, wherein the first type of node includes at least one of the following: enterprise node, mortgage registration node, registration node, market transaction node; connecting with a second type of node through an application programming interface to acquire internal business information data of the enterprise, wherein the second type of node includes at least a bank node; regularly acquiring updated business information data based on an automated process, and updating the enterprise data in real time.

[0083] like Figure 3 As shown, by connecting with the first type of nodes through the application programming interface, the external business information data of the enterprise is obtained, that is, through the mortgage registration node, registration node, market transaction node, enterprise node, and bank node of the carbon asset, multi-channel real-time acquisition of business data reflecting market changes and risk dynamics and construction of post-loan risk indicators, enriching the data dimension of carbon asset risk assessment and improving the reliability of model training data.

[0084] Based on the automated process, the updated business information data is obtained regularly, and the enterprise data is updated in real time, that is, the real-time acquisition of carbon asset business data based on multi-source data fusion. Through API interface and process automation technology, carbon asset business data is obtained and updated in real time through multiple channels, and market changes and risk dynamics are obtained in a timely manner. Through the mortgage registration node, registration node, market transaction node, enterprise node, and bank node of carbon assets, business data reflecting market changes and risk dynamics are obtained in real time through multiple channels and post-loan risk indicators are constructed to enrich the data dimension of carbon asset risk assessment and improve the reliability of model training data.

[0085] In one embodiment of the present application, the acquisition of carbon asset related data also includes: when an enterprise uses carbon assets as collateral for a loan, based on the enterprise node, the mortgage registration node of the carbon assets, the registration node of the carbon assets and the market transaction node, the enterprise basic information, carbon asset ownership information, carbon asset mortgage registration information, and carbon market transaction information are respectively obtained; and the enterprise loan information is obtained based on the bank node.

[0086] like Figure 3 As shown, before carrying out the post-loan inspection task, follow the principle of "user authorization, minimum sufficient", clearly indicate the purpose, method and scope of user data collection and use through privacy policy documents, customer information authorization, etc., and obtain information collection and information query authorization from corporate customers. The authorization content includes the collection and query of external and internal original business data.

[0087] Collect external and internal original business data. After authorization by the enterprise, obtain the original data such as basic enterprise information, carbon asset ownership information, carbon asset mortgage registration information, carbon market transaction information from the external enterprise node, carbon asset registration node, carbon asset mortgage registration node, and market transaction node through API technology, and obtain enterprise loan information from the internal bank node. When data is stored and transmitted, encrypted channels are used for data transmission to reduce the risk of data leakage.

[0088] in,

[0089] (1) External business information includes but is not limited to:

[0090] Obtain basic information about the company's carbon assets from the company node: Obtain the company's historical annual emissions E from the company information system history 、The company's estimated emissions this year E expect , Enterprise unified social credit code U company 、National Voluntary Emission Reduction Registration System General Holding Account N ccer , voluntary greenhouse gas emission reduction trading institutions open trading system accounts N cea 、Performance and payment amount A contract wait.

[0091] Obtain carbon asset ownership information from the carbon asset registration node: Obtain the total amount of the enterprise's national certified resource emission reduction from the national greenhouse gas voluntary emission reduction registration system A sset1 Validity period T ccer 、Enterprise Account StatusS ccersign , failure transition period T passby , obtain the total amount of corporate carbon emission quotas A from the national carbon emission rights registration system sset2 Validity period T cea 、The start and end time of the current fulfillment cycle T begin With T end 、Enterprise account status T ceasign wait.

[0092] Obtain carbon asset mortgage registration information from the carbon asset mortgage registration node: Obtain enterprise pledge owner B from the People's Bank of China Credit Information Center movable property financing unified registration and publicity system people , Unified social credit code of the pledgee's owner C people 、Pledgee W people , Pledge registration status S people 、Pledge registration quantity A peoplepledge . Obtain the collateral owner B from the national carbon emission trading system carbon , Unified social credit code of the pledgee's owner C carbon 、Pledgee W carbon , Pledge registration status S carbon 、Pledge registration quantity A carbonpledge wait.

[0093] Obtain carbon asset market transaction information from market transaction nodes: Obtain carbon emission rights market price P1 and enterprise account status S from the national carbon emission rights trading system ceatrade， The market price of voluntary emission reductions certified by the National Voluntary Greenhouse Gas Emission Reduction Trading System P2 and the enterprise account status S ccertrade wait.

[0094] in,

[0095] (2) Internal business information includes but is not limited to:

[0096] Get enterprise loan information from the bank node: get loan amount L, loan interest rate RI, contract term T loan , Guarantee amount I, Guarantee duration T pledge , pledge rate R pledge wait.

[0097] Preferably, considering that the above raw data usually exists in multiple formats, such as comma separated value (CSV) files, Excel files, or files written in scripting languages ​​(such as XML, HTML, etc.), in order to facilitate processing and storage, JavaScript Object Notation (JSON), a lightweight data exchange format, is selected as a unified format to store and represent data in a text form that is easy to read and write, and is also easy for machines to parse and generate.

[0098] Through API interface and process automation technology, carbon asset business data can be obtained and updated in real time through multiple channels, and market changes and risk dynamics can be obtained in a timely manner. Technical features include: multiple channels connected by API include mortgage registration nodes, registration nodes, market transaction nodes, enterprise nodes, and bank nodes; process automation can obtain the latest business information data through multiple channels at regular intervals, including carbon asset-related information and bank business data.

[0099] In one embodiment of the present application, the carbon asset-related data is input into the prediction model, and the carbon asset-related risk characteristic data is extracted to obtain the post-loan risk prediction result of the enterprise carbon asset mortgage financing, including: after the carbon asset-related data is input into the prediction model, the carbon asset market risk characteristics, carbon asset behavior risk characteristics, and carbon asset target risk characteristics are extracted according to the prediction model; according to the carbon asset market risk characteristics, the carbon asset behavior risk characteristics, and the carbon asset target risk characteristics, the quantitative results of the enterprise post-loan risk indicators are output through the prediction model; according to the quantitative results of the enterprise post-loan risk indicators, risk classification is performed to obtain the post-loan risk prediction result of the enterprise carbon asset mortgage financing; wherein, the carbon asset market risk is used to characterize the carbon asset market risk of the change in the market price of the carbon asset collateral to the collateral guarantee balance; the carbon asset behavior risk characteristics are used to characterize the carbon asset behavior risk generated by the loan enterprise when the key change behavior occurs in the carbon market; the carbon asset target risk characteristics are used to characterize the carbon asset target risk of the enterprise carbon asset as collateral during the existence period.

[0100] Data characterization processing. Apply data characterization processing methods to perform fine feature extraction on at least three types of indicators covering the main post-loan risks of carbon asset pledge financing, including: carbon asset market risk, carbon asset behavior risk, and carbon asset target risk.

[0101] (1) Feature extraction of carbon asset market risk indicators.

[0102] Carbon asset market risk indicators include credit warning α1 and credit limit α2. These characteristics characterize the impact of market price changes of carbon asset collateral on the collateral guarantee balance. Carbon asset market risk is a function of its characteristics.

[0103] M = [α1, α2]

[0104] Feature α1: Is the market value of the carbon asset collateral lower than the credit warning line?

[0105]

[0106] Where L alert It is a configurable parameter and can be set to different levels based on actual business requirements.

[0107] Feature α2: Is the market value of the carbon asset collateral lower than the credit limit?

[0108]

[0109] (2) Feature extraction of carbon asset behavior risk indicators. Carbon asset behavior risk indicators include mortgage registration risk β1, performance risk β2, and transaction restriction risk β3. These characteristics characterize the risks generated when loan companies register, trade, settle, pledge, and change in the carbon market. Carbon asset behavior risk is a function of its characteristics.

[0110] B = [β1, β2, β3]

[0111] Feature β1: The company's pledged property has completed double pledge registration

[0112]

[0113] Feature β2: The compliance period is approaching and the stock of carbon assets is sufficient to fulfill the compliance

[0114]

[0115] Feature β3: There are no abnormal transaction restrictions in the registration system and the trading system

[0116]

[0117] (3) Feature extraction of carbon asset target risk indicators. The carbon asset target risk indicators include the enterprise carbon asset ownership γ1 and the collateral failure period γ2. These characteristics characterize the risk of the enterprise carbon asset as collateral during its existence. The carbon asset target risk is a function of its characteristics.

[0118] T = [γ1, γ2]

[0119] Feature γ1: Confirming the company’s ownership of carbon assets

[0120]

[0121] Feature γ2: CCER certification methodology becomes invalid during the pledge period, and the transition period cannot cover the loan term

[0122]

[0123] The indicator data is characterized to obtain the characteristic functions of the enterprise's carbon assets. Compared with the original data, these characteristics can more intuitively and accurately reflect the market risk, behavioral risk and target risk of the enterprise using carbon assets as collateral, thereby providing accurate data for the next step of model calculation.

[0124] Based on the directly acquired multi-source data, a multi-dimensional indicator system is constructed in combination with the characteristics of the pledge financing post-loan management business scenarios, including: characterizing the carbon asset market risk of the pledge guarantee balance due to changes in the market price of carbon asset pledges, characterizing the carbon asset behavior risk generated by the loan company's key changes in the carbon market, and characterizing the carbon asset target risk of the company's carbon assets as collateral during its existence.

[0125] Quantitative calculation of enterprise post-loan risk indicators based on logistic regression algorithm, collection and labeling of enterprise post-loan risk data. In order to provide labeled data for subsequent model training, consider expert data labeling for enterprise post-loan risk. First, collect relevant data including defaulted enterprises and normal operating enterprises. For normal operating and non-defaulting enterprises, organize industry experts to label their risk levels and score labels, further increase the amount of training data, and finally store the integrated data in the intranet database.

[0126] Model construction and training of logistic regression. Use the carbon asset market risk data (M), carbon asset behavior risk data (B) and carbon asset target risk data (T) obtained in the previous step to build a logistic regression model. These features are all expressed in the form of matrices, that is, they are composed of multiple related sub-features, so a risk feature matrix X containing three matrices [M, B, T] is constructed:

[0127] X = [M, B, T]

[0128] And initialize the parameter matrix:

[0129] W = [a, b, c]

[0130] Then, the optimal weight parameters are obtained by training based on historical data through machine learning methods. Define linear combination:

[0131] L=W*X

[0132] Where * represents the product of corresponding elements of the matrix, and the result L is also a matrix.

[0133] Then L is brought into the logistic function. The logistic function is used to map the value of L to between 0 and 1, indicating the probability of default. The final model can be expressed as:

[0134]

[0135] Define the loss function as an indicator for model optimization. Generally, the cross entropy loss function commonly used in logistic regression models is selected:

[0136]

[0137] Where n represents the number of samples, y i is the true label of the i-th sample (0 means no default, 1 means default, and 0-1 means the probability of default or the probability corresponding to the risk level marked by the expert), f(S i ) is the default probability of the i-th sample predicted by the model.

[0138] During the model training process, until the model prediction loss function error can no longer be significantly reduced or reaches the preset number of iterations, it indicates that the model training is completed and the value of the parameter matrix W is finally determined. A quantitative calculation method for post-loan risk based on logistic regression is designed, and an enterprise post-loan risk database is constructed with the help of expert annotations. Further optimization training is performed by constructing a loss function, and machine learning methods are used to improve the accuracy of the model; finally, the post-loan default probability and risk level are output and transmitted to the downstream credit system for early warning, which improves the effectiveness of risk warning and solves the technical problem of insufficient post-loan risk management in the carbon asset pledge financing business scenario.

[0139] Finally, according to the risk scoring results, set thresholds and output levels: According to the set scoring intervals or standards, multiple scoring thresholds are defined. These thresholds are used as demarcation points to identify different risk levels, so as to achieve a more detailed division of the enterprise's risk status. Finally, its risk level is divided into five intervals, corresponding to the five levels of "very low risk", "low risk", "medium risk", "high risk" and "very high risk".

[0140] Through expert annotation, we obtain training data, use machine learning to train optimal weight parameters based on historical data, select cross entropy loss function for model optimization, determine the final parameter matrix and output the risk level, design an early warning mechanism, and finally transmit risk warning information in real time through multiple message channels. Using feature engineering, we focus on the characteristics of carbon assets, integrate the market value risk, behavior risk, and target risk data of carbon assets, and build a highly explanatory carbon asset post-loan risk indicator system.

[0141] In one embodiment of the present application, based on the post-loan risk prediction results of the enterprise's carbon asset mortgage financing, when the risk prediction results are higher than a threshold, a risk warning is issued to the downstream system, including: based on the post-loan risk prediction results of the enterprise's carbon asset mortgage financing, obtaining the corresponding risk level to issue a post-loan risk warning to the enterprise; displaying the enterprise's carbon asset market risk data, the carbon asset behavior risk data, the carbon asset target risk data, and the final risk score and risk level information; when the enterprise's risk level reaches a preset high risk, real-time delivery of risk warning information through any one or more methods such as SMS, email, and application push.

[0142] Collection and labeling of enterprise post-loan risk data. In order to provide labeled data for subsequent model training, consider expert data labeling for enterprise post-loan risks. First, collect relevant data including defaulted enterprises and normal operating enterprises. For normal operating and non-defaulting enterprises, organize industry experts to label their risk levels and score labels, further increase the amount of training data, and finally store the integrated data in the intranet database.

[0143] According to the changes in the information of various indicators of the enterprise, the risk indicator calculation algorithm is run in real time, the enterprise risk score and risk level in the database are updated, and finally the risk calculation results are output to the credit business system.

[0144] Furthermore, the dashboard interface is used in the credit system for integrated information display, specifically displaying the enterprise carbon asset market risk data (M), carbon asset behavior risk data (B), carbon asset target risk data (T) and the final risk score and risk level, so that decision makers can understand the risk status of the enterprise. At the same time, a real-time notification and alarm mechanism is also designed: when the enterprise risk level reaches high risk and extremely high risk, the alarm mechanism is triggered, and finally the risk warning information is delivered in real time through SMS, email, and application push.

[0145] In one embodiment of the present application, the method further includes: transmitting the post-loan default probability and risk level output from the post-loan risk prediction results of the enterprise's carbon asset mortgage financing to the downstream credit system for post-loan risk warning for the enterprise.

[0146] Dynamic risk indicator calculation and early warning information display: According to the changes in the enterprise's various indicator information, the risk indicator calculation algorithm obtained in the third step is run in real time to update the enterprise risk score and risk level in the database, and finally the risk calculation results are output to the credit business system. In the credit system, we designed a dashboard interface for integrated information display, specifically displaying the enterprise's carbon asset market risk data (M), carbon asset behavior risk data (B), carbon asset target risk data (T) and the final risk score and risk level, so that decision makers can understand the enterprise's risk status. At the same time, a real-time notification and alarm mechanism is designed: when the enterprise risk level reaches high risk and extremely high risk, the alarm mechanism is triggered, and finally the risk warning information is delivered in real time through SMS, email, and application push.

[0147] In one embodiment of the present application, the method also includes: adjusting the acquired enterprise data to a common format through a standardized processing procedure; storing the processed data in a common format in a network database; setting a data consumption request frequency, periodically re-querying to acquire the enterprise data, and updating the storage of the network database.

[0148] Clean, preprocess and store the original data set. Adjust the data to a common format through standardization, which includes processing missing values, detecting outliers and cleaning data. Store the processed data in the internal network database, organize and retrieve data in a structured way, and ensure that the data is used in a secure network environment to improve data security and reduce the risk of data leakage and tampering. Set the frequency of data consumption requests, re-query the above enterprise data regularly, and update the storage after digital processing and cleaning to ensure the real-time nature of the data.

[0149] like Figure 4 As shown, it is a functional structure diagram of the data processing program in the risk warning method in the embodiment of the present application, which includes a data collection module, a data preprocessing module, a labeling and training module, and a warning module. The following is the classification and function of each module.

[0150] During the model training phase, the data collection module is carried out. The main task of this module is to collect the required original business data from internal and external data sources. Including relevant characteristic data of enterprises: obtaining basic information of enterprises, carbon asset ownership information, carbon asset mortgage registration information, carbon market transaction information, enterprise financing information, etc. from enterprise nodes, carbon asset registration nodes, carbon asset mortgage registration nodes, market transaction nodes, and bank nodes, and also including historical data of enterprises that have defaulted. For enterprises that have not defaulted, experts are organized to mark their risk levels and scores. The data collection module provides necessary data input for the work of subsequent modules.

[0151] During the model training phase, the data preprocessing module is used to process the collected raw data, including cleaning the data, handling missing values ​​and outliers, data standardization, and storing the processed data. This module is also responsible for regularly updating and storing the latest data to ensure the real-time nature of the data.

[0152] During the model training phase, a model construction and training module is carried out. This module uses the collected enterprise-related data to build and train a risk warning model based on a logistic regression algorithm. At the same time, a unique loss function is defined during the model training process to help optimize the model training results.

[0153] Early warning module: The task of this module is to issue early warnings for the post-loan risks of enterprises based on the trained model and the set thresholds. This module will classify the risk level into five levels, namely "very low risk", "low risk", "medium risk", "high risk" and "very high risk". Then, the module will transmit this risk level information to the downstream credit system for the next step of risk disposal.

[0154] The present application embodiment also provides a carbon asset mortgage financing post-loan risk warning device 500, such as Figure 5 As shown, a schematic diagram of the structure of the carbon asset mortgage financing post-loan risk warning device in the embodiment of the present application is provided. The carbon asset mortgage financing post-loan risk warning device 500 at least includes: a data acquisition module 510, a feature extraction module 520 and an early warning module 530, wherein:

[0155] In one embodiment of the present application, the data acquisition module 510 is specifically used to obtain carbon asset related data.

[0156] like Figure 2 As shown, the target enterprise's business data is obtained, the data format and standard are unified, the enterprise's initial data is formed and stored separately.

[0157] After authorization by the enterprise, the original data such as enterprise basic information, carbon asset ownership information, carbon asset mortgage registration information, carbon market transaction information, etc. are obtained from external enterprise nodes, carbon asset registration nodes, carbon asset mortgage registration nodes, and market transaction nodes through API (application programming interface) technology, and enterprise loan information is obtained from internal bank nodes.

[0158] It should be noted that when the acquired data is stored and transmitted, an encrypted channel is used for data transmission to reduce the risk of data leakage.

[0159] It should be noted that enterprise data of the same dimensions are collected and obtained during both the training and verification stages of the prediction model.

[0160] In one embodiment of the present application, the feature extraction module 520 is specifically used to: input the enterprise data into the prediction model, extract the carbon asset-related risk feature data, and obtain the post-loan risk prediction results of the enterprise's carbon asset mortgage financing. The prediction model is obtained through machine learning training using multiple groups of data, and each group of data in the multiple groups includes normal operating enterprise data and default enterprise data.

[0161] Carbon assets with carbon quotas and national certified emission reductions as the target have become new assets for enterprises. In the business scenario where banks accept enterprises to use such assets as collateral for financing, in the specific implementation, the prediction model is trained by characterizing the data during the training phase of the model. In the verification phase of the model, it is necessary to extract the risk data related to carbon assets to evaluate the post-loan risk prediction results of carbon asset mortgage financing.

[0162] It can be understood that when processing data characterization, fine feature extraction is performed on three types of indicators covering the main post-loan risks of carbon asset pledge financing. The three types of indicators include: carbon asset market risk, carbon asset behavioral risk, and carbon asset target risk.

[0163] like Figure 2 As shown in the figure, the initial data of the enterprise is processed to form the characteristic data of the enterprise's carbon assets, and the risk warning indicators are calculated to prepare for the next step of the risk quantification calculation model. According to the characteristic data of carbon assets, the enterprise's post-loan risk data is labeled and collected, and a quantitative prediction model of the enterprise's post-loan credit indicators based on the logistic regression algorithm is constructed, and a loss function is constructed for optimization training.

[0164] The indicator data is characterized to obtain the characteristic functions of the enterprise's carbon assets. Compared with the original data, these characteristics can more intuitively and accurately reflect the market risk, behavioral risk and target risk of the enterprise using carbon assets as collateral, and also provide accurate data for the next step of model calculation.

[0165] In one embodiment of the present application, the early warning module 530 is specifically used to: based on the post-loan risk prediction result of the enterprise's carbon asset mortgage financing, when the risk prediction result is higher than a threshold, issue a risk warning to the downstream system.

[0166] like Figure 2 As shown, based on the obtained quantitative prediction model of enterprise post-loan risk indicators, the risk warning data is sent to the business system for the dynamic forecast period risks of the concerned enterprises, business warnings are made and post-loan management measures are responded.

[0167] According to the risk scoring results, set thresholds and output levels: According to the set scoring intervals or standards, set multiple scoring thresholds. For example, the risk level can be finally divided into five intervals, corresponding to the five levels of "very low risk", "low risk", "medium risk", "high risk" and "very high risk".

[0168] It can be understood that the above-mentioned risk warning device can implement each step of the risk warning method provided in the above-mentioned embodiment, and the relevant explanations about the risk warning method are applicable to the risk warning device and will not be repeated here.

[0169] In one embodiment of the present application, the data acquisition module 510 is also used to

[0170] Connecting with the first type of node through an application programming interface to obtain external business information data of the enterprise, wherein the first type of node includes at least one of the following: an enterprise node, a mortgage registration node, a registration node, and a market transaction node;

[0171] Connecting to the second type of nodes through an application programming interface to obtain internal business information data of the enterprise, wherein the second type of nodes at least includes a bank node;

[0172] Based on the automated process, updated business information data is obtained regularly, and the enterprise data is updated in real time.

[0173] In one embodiment of the present application, the data acquisition module 510 is also used to

[0174] When an enterprise uses carbon assets as collateral for a loan, basic enterprise information, carbon asset ownership information, carbon asset mortgage registration information, and carbon market transaction information are obtained according to the enterprise node, the carbon asset mortgage registration node, the carbon asset registration node, and the market transaction node;

[0175] as well as

[0176] The enterprise loan information is obtained according to the bank node.

[0177] In one embodiment of the present application, the feature extraction module 520 is also used to

[0178] After the carbon asset related data is input into the prediction model, the carbon asset market risk characteristics, carbon asset behavior risk characteristics, and carbon asset target risk characteristics are extracted according to the prediction model;

[0179] Outputting the quantitative results of the enterprise post-loan risk indicators through the prediction model according to the carbon asset market risk characteristics, the carbon asset behavior risk characteristics and the carbon asset target risk characteristics;

[0180] Perform risk grading based on the quantitative results of the enterprise's post-loan risk indicators to obtain the enterprise's carbon asset mortgage financing post-loan risk prediction results;

[0181] in,

[0182] The carbon asset market risk is used to characterize the carbon asset market risk of changes in the market price of carbon asset collateral to the collateral guarantee balance;

[0183] The carbon asset behavior risk characteristics are used to characterize the carbon asset behavior risks generated by the lending enterprise when the key changes occur in the carbon market;

[0184] The carbon asset target risk characteristics are used to characterize the carbon asset target risks of the enterprise's carbon assets as collateral during the period of existence.

[0185] In one embodiment of the present application, the early warning module 530 is also used to

[0186] According to the post-loan risk prediction results of the carbon asset mortgage financing of the enterprise, the corresponding risk level is obtained to provide a post-loan risk warning for the enterprise;

[0187] Display the enterprise's carbon asset market risk data, carbon asset behavior risk data, carbon asset target risk data, and final risk score and risk level information;

[0188] When the enterprise's risk level reaches the preset high risk, risk warning information is delivered in real time through any one or more methods such as SMS, email, and application push.

[0189] In one embodiment of the present application, it further includes: a processing module for

[0190] The post-loan default probability and risk level output from the post-loan risk prediction results of the enterprise's carbon asset mortgage financing are transmitted to the downstream credit system to provide post-loan risk warning to the enterprise.

[0191] In one embodiment of the present application, it also includes: a standardization module for

[0192] Adjusting the acquired enterprise data into a common format through a standardization process;

[0193] storing the processed data in a common format in a network database;

[0194] The frequency of data consumption requests is set, and the enterprise data is re-queried and acquired at regular intervals, and the storage of the network database is updated.

[0195] Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 6At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0196] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0197] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0198] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a risk warning device at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:

[0199] Obtain data related to carbon assets;

[0200] The carbon asset-related data is input into the prediction model, and the carbon asset-related risk characteristic data is extracted to obtain the post-loan risk prediction result of the enterprise carbon asset mortgage financing, wherein the prediction model is obtained by machine learning training using multiple sets of data, and each set of data in the multiple sets of data includes data of normally operating enterprises and data of defaulting enterprises;

[0201] Based on the post-loan risk prediction results of the enterprise's carbon asset mortgage financing, when the risk prediction result is higher than the threshold, a risk warning will be issued to the downstream system.

[0202] The above application Figure 1The method performed by the risk warning device disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0203] The electronic device may also perform Figure 1 The method for executing the risk warning device in the Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0204] The present application also provides a computer-readable storage medium, which stores one or more programs, wherein the one or more programs include instructions, which, when executed by an electronic device including multiple application programs, enable the electronic device to execute Figure 1 The method performed by the risk warning device in the illustrated embodiment is specifically used to perform:

[0205] Obtain data related to carbon assets;

[0206] The carbon asset-related data is input into the prediction model, and the carbon asset-related risk characteristic data is extracted to obtain the post-loan risk prediction result of the enterprise carbon asset mortgage financing, wherein the prediction model is obtained by machine learning training using multiple sets of data, and each set of data in the multiple sets of data includes data of normally operating enterprises and data of defaulting enterprises;

[0207] Based on the post-loan risk prediction results of the enterprise's carbon asset mortgage financing, when the risk prediction result is higher than the threshold, a risk warning will be issued to the downstream system.

[0208] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0209] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0210] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0211] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1The steps for the functions specified in one or more boxes.

[0212] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0213] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0214] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0215] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0216] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0217] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A post-loan risk warning method for carbon asset mortgage financing, wherein: The method comprises: Obtain data related to carbon assets; The carbon asset-related data is input into the prediction model, and the carbon asset-related risk characteristic data is extracted to obtain the post-loan risk prediction result of the enterprise carbon asset mortgage financing, wherein the prediction model is obtained by machine learning training using multiple sets of data, and each set of data in the multiple sets of data includes data of normally operating enterprises and data of defaulting enterprises; Based on the post-loan risk prediction results of the enterprise's carbon asset mortgage financing, when the risk prediction result is higher than the threshold, a risk warning will be issued to the downstream system.

2. The method of claim 1, wherein: The obtaining of carbon asset related data includes: Connecting with the first type of node through an application programming interface to obtain external business information data of the enterprise, wherein the first type of node includes at least one of the following: an enterprise node, a mortgage registration node, a registration node, and a market transaction node; Connecting to the second type of nodes through an application programming interface to obtain internal business information data of the enterprise, wherein the second type of nodes at least includes a bank node; Based on the automated process, updated business information data is obtained regularly, and the enterprise data is updated in real time.

3. The method according to claim 2, wherein the step of obtaining carbon asset related data further comprises: When an enterprise uses carbon assets as collateral for a loan, basic enterprise information, carbon asset ownership information, carbon asset mortgage registration information, and carbon market transaction information are obtained according to the enterprise node, the carbon asset mortgage registration node, the carbon asset registration node, and the market transaction node; as well as The enterprise loan information is obtained according to the bank node.

4. The method of claim 2, wherein: The step of inputting the carbon asset related data into the prediction model, extracting the carbon asset related risk characteristic data, and obtaining the post-loan risk prediction results of the enterprise carbon asset mortgage financing includes: After the carbon asset related data is input into the prediction model, the carbon asset market risk characteristics, carbon asset behavior risk characteristics, and carbon asset target risk characteristics are extracted according to the prediction model; Outputting the quantitative results of the enterprise post-loan risk indicators through the prediction model according to the carbon asset market risk characteristics, the carbon asset behavior risk characteristics and the carbon asset target risk characteristics; Perform risk grading based on the quantitative results of the enterprise's post-loan risk indicators to obtain the enterprise's carbon asset mortgage financing post-loan risk prediction results; in, The carbon asset market risk is used to characterize the carbon asset market risk of changes in the market price of carbon asset collateral to the collateral guarantee balance; The carbon asset behavior risk characteristics are used to characterize the carbon asset behavior risks generated by the lending enterprise when the key changes occur in the carbon market; The carbon asset target risk characteristics are used to characterize the carbon asset target risks of the enterprise's carbon assets as collateral during the period of existence.

5. The method of claim 4, wherein: According to the post-loan risk prediction results of the carbon asset mortgage financing of the enterprise, when the risk prediction results are higher than the threshold, a risk warning is issued to the downstream system, including: According to the post-loan risk prediction results of the carbon asset mortgage financing of the enterprise, the corresponding risk level is obtained to provide a post-loan risk warning for the enterprise; Display the enterprise's carbon asset market risk data, carbon asset behavior risk data, carbon asset target risk data, and final risk score and risk level information; When the enterprise's risk level reaches the preset high risk, risk warning information is delivered in real time through any one or more methods such as SMS, email, and application push.

6. The method according to any one of claims 1 to 5, further comprising: The post-loan default probability and risk level output from the post-loan risk prediction results of the enterprise's carbon asset mortgage financing are transmitted to the downstream credit system to provide post-loan risk warning to the enterprise.

7. The method according to any one of claims 1 to 5, further comprising: Adjusting the acquired enterprise data into a common format through a standardization process; storing the processed data in a common format in a network database; The frequency of data consumption requests is set, and the enterprise data is re-queried and acquired at regular intervals, and the storage of the network database is updated.

8. A post-loan risk warning device for carbon asset mortgage financing, wherein: The device comprises: Data collection module, for carbon asset related data; A feature extraction module is used to input the carbon asset related data into the prediction model, extract the carbon asset related risk feature data, and obtain the post-loan risk prediction result of the enterprise carbon asset mortgage financing, wherein the prediction model is obtained by machine learning training using multiple sets of data, and each set of data in the multiple sets of data includes data of normal operating enterprises and data of defaulting enterprises; The early warning module is used to issue a risk warning to the downstream system based on the post-loan risk prediction result of the carbon asset mortgage financing of the enterprise, when the risk prediction result is higher than the threshold.

9. An electronic device, comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute any one of the methods of claims 1 to 7.

Images