Risk assessment method and device based on green credit, electronic equipment, medium and program product

Abstract

The invention provides a risk assessment method based on green credit, relates to application of a large model in the field of financial science and technology, and can be applied to the technical field of artificial intelligence, the technical field of big data and the field of financial science and technology. The method comprises the following steps: acquiring financial dimension data and environment dimension data of a target enterprise; based on the financial dimension data and the environmental dimension data, obtaining a financial risk assessment result and an environmental risk assessment result of the target enterprise, the environmental risk assessment result at least comprising an environmental protection declaration authenticity verification result; integrating the financial risk assessment result and the environmental risk assessment result to obtain a target risk level; and obtaining a risk assessment result based on the target risk level.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of artificial intelligence, the technical field of big data, and the technical field of financial technology, and more particularly to a green credit-based risk assessment method, device, equipment, medium, and program product. BACKGROUND

[0002] Green credit risk assessment aims to ensure that credit funds are invested in environmental protection, energy saving, and low-carbon economic fields. By comprehensively considering factors such as environmental friendliness, policy compliance, technology maturity, market prospects, and financial feasibility, the environmental benefits and credit risks of a project are assessed.

[0003] However, the existing assessment methods face several technical problems. First, financial risk assessment and environmental risk assessment lack effective integration, and the combined impact of both on overall risk cannot be comprehensively considered, resulting in insufficient accuracy of the assessment results. Furthermore, traditional methods fail to adequately verify the authenticity of corporate environmental protection declarations, making it difficult to comprehensively assess environmental risks. SUMMARY

[0004] In view of the above problems, the present disclosure provides a green credit-based risk assessment method, device, equipment, medium, and program product.

[0005] According to a first aspect of the present disclosure, a green credit-based risk assessment method is provided, the method comprising: obtaining financial dimension data and environmental dimension data of a target enterprise; based on the financial dimension data and the environmental dimension data, obtaining a financial risk assessment result and an environmental risk assessment result of the target enterprise, wherein the environmental risk assessment result at least includes a verification result of the authenticity of an environmental protection declaration; integrating the financial risk assessment result and the environmental risk assessment result to obtain a target risk level; and based on the target risk level, obtaining a risk assessment result.

[0006] According to an embodiment of the present disclosure, the financial risk assessment result of the target enterprise is obtained based on the financial dimension data, specifically comprising: extracting time sequence features of the financial dimension data, obtaining the correlation strength between financial subjects based on the time sequence features; performing target scenario simulation based on the time sequence features and the correlation strength to obtain a financial performance simulation result; and obtaining the financial risk assessment result based on the financial performance simulation result.

[0007] According to an embodiment of the present disclosure, the environment risk assessment result is obtained based on the environment dimension data, specifically comprising: performing semantic analysis and authenticity verification on the environmental protection declaration data of the target enterprise to obtain an environmental protection declaration authenticity verification result; comparing the environmental performance of the target enterprise with industry benchmark data to obtain an industry benchmark comparison result; analyzing the environmental behavior trajectory of the target enterprise to obtain a behavior trajectory analysis result; and generating the environment risk assessment result based on the environmental protection declaration authenticity verification result, the industry benchmark comparison result and the behavior trajectory analysis result.

[0008] According to an embodiment of the present disclosure, the environmental protection declaration authenticity verification result is obtained based on the semantic analysis and authenticity verification on the environmental protection declaration data of the target enterprise, specifically comprising: decomposing the environmental protection declaration data into environmental protection commitment data, environmental protection measure data and environmental protection effectiveness data, calculating a matching degree based on the environmental protection commitment data, the environmental protection measure data and the environmental protection effectiveness data to obtain a authenticity score; performing fuzzy identification on the environmental protection declaration data to obtain a fuzzy expression label; and obtaining the environmental protection declaration authenticity verification result based on the authenticity score and the fuzzy expression label.

[0009] According to an embodiment of the present disclosure, the financial dimension data and the environment dimension data of the target enterprise are obtained based on structured data sources, unstructured data sources and dynamic data streams.

[0010] According to an embodiment of the present disclosure, the method further comprises: performing multi-source data fusion and spatio-temporal alignment on the financial dimension data and the environment dimension data to obtain synchronized multi-source data; performing standardization processing and feature extraction on the synchronized multi-source data to obtain standardized feature data; and performing missing value filling on the standardized feature data to obtain filled data.

[0011] According to an embodiment of the present disclosure, the method further comprises: constructing a causal graph between environmental protection behaviors and green credit defaults; performing counterfactual reasoning based on the causal graph and a target environmental protection decision-making scenario to obtain a counterfactual reasoning result; and obtaining a causal chain analysis result based on the causal graph and the counterfactual reasoning result.

[0012] According to an embodiment of the present disclosure, the method further comprises: integrating a causal discovery algorithm, the causal discovery algorithm being used to dynamically update the action direction between variables in the causal graph; and in response to detecting a change in policies and regulations, triggering the causal discovery algorithm to reconstruct the causal graph.

[0013] The second aspect of the present disclosure provides a green credit-based risk assessment device, comprising: a data acquisition module configured to acquire financial dimension data and environmental dimension data of a target enterprise; a risk assessment module configured to acquire a financial risk assessment result and an environmental risk assessment result of the target enterprise based on the financial dimension data and the environmental dimension data, wherein the environmental risk assessment result at least includes an environmental protection declaration authenticity verification result; a result integration module configured to integrate the financial risk assessment result and the environmental risk assessment result to obtain a target risk level; and an assessment result acquisition module configured to acquire a risk assessment result based on the target risk level.

[0014] According to an embodiment of the present disclosure, the risk assessment module can be further configured to extract time sequence features of the financial dimension data, acquire correlation strengths between financial subjects based on the time sequence features, perform target scenario simulation based on the time sequence features and the correlation strengths to obtain a financial performance simulation result, and acquire the financial risk assessment result based on the financial performance simulation result.

[0015] According to an embodiment of the present disclosure, the risk assessment module can be further configured to perform semantic analysis and authenticity verification based on environmental protection declaration data of the target enterprise to acquire an environmental protection declaration authenticity verification result, compare environmental performance of the target enterprise with industry benchmark data to acquire an industry benchmark comparison result, analyze an environmental behavior trajectory of the target enterprise to acquire a behavior trajectory analysis result, and generate the environmental risk assessment result based on the environmental protection declaration authenticity verification result, the industry benchmark comparison result, and the behavior trajectory analysis result.

[0016] According to an embodiment of the present disclosure, the risk assessment module can be further configured to decompose the environmental protection declaration data into environmental protection commitment data, environmental protection measure data, and environmental protection effectiveness data, calculate a matching degree based on the environmental protection commitment data, the environmental protection measure data, and the environmental protection effectiveness data to obtain an authenticity score, perform fuzzy identification on the environmental protection declaration data to obtain a fuzzy expression label, and acquire the environmental protection declaration authenticity verification result based on the authenticity score and the fuzzy expression label.

[0017] According to an embodiment of the present disclosure, the risk assessment device can be further configured to perform multi-source data fusion and space-time alignment on the financial dimension data and the environmental dimension data to acquire synchronized multi-source data, perform standardization processing and feature extraction on the synchronized multi-source data to obtain standardized feature data, and perform missing value filling on the standardized feature data to acquire filled data.

[0018] According to embodiments of this disclosure, the risk assessment device can also be used to construct a causal graph between environmental protection behavior and green credit default; perform counterfactual reasoning based on the causal graph and the target environmental decision-making scenario to obtain counterfactual reasoning results; and obtain causal chain analysis results based on the causal graph and the counterfactual reasoning results.

[0019] According to embodiments of this disclosure, the risk assessment device can also be used to integrate a causal discovery algorithm, which is used to dynamically update the direction of interaction between variables in the causal graph; and to trigger the causal discovery algorithm to reconstruct the causal graph in response to the detection of policy and regulatory changes.

[0020] A third aspect of this disclosure provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0021] A fourth aspect of this disclosure also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0022] The fifth aspect of this disclosure also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method.

[0023] According to embodiments of this disclosure, by integrating financial and environmental data and verifying the authenticity of environmental claims, the problems of data fragmentation and inefficient processing in traditional methods are avoided. This makes the assessment process more efficient, comprehensively improving the accuracy and completeness of the assessment, thereby effectively enhancing computational performance. Simultaneously, the integrated financial and environmental risk assessment results provide users with clearer and more reliable risk level information, helping them to more intuitively understand the company's overall risk situation, thus improving the accuracy of decision-making. Verifying the authenticity of environmental claims can accurately identify greenwashing behavior, further enhancing the user experience. Attached Figure Description

[0024] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0025] Figure 1 The illustration schematically depicts application scenarios of risk assessment methods, apparatuses, devices, media, and program products based on green credit according to embodiments of this disclosure;

[0026] Figure 2A flowchart illustrating a risk assessment method based on green credit according to an embodiment of this disclosure is shown schematically.

[0027] Figure 3 A flowchart illustrating a method for obtaining financial risk assessment results according to an embodiment of this disclosure is shown schematically.

[0028] Figure 4 A flowchart illustrating a method for obtaining environmental risk assessment results according to an embodiment of the present disclosure is shown schematically.

[0029] Figure 5 A schematic diagram illustrating a structural block diagram of a risk assessment apparatus based on green credit according to an embodiment of the present disclosure; and

[0030] Figure 6 A block diagram schematically illustrates an electronic device suitable for implementing a risk assessment method based on green credit, according to an embodiment of the present disclosure. Detailed Implementation

[0031] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0032] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0033] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0034] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0035] First, the technical terms used in this article are explained and clarified as follows.

[0036] An ESG report is a performance report on a company's environmental, social, and governance (ESG) aspects. It reflects a company's actions, policies, and achievements in sustainability, including how it addresses climate change, social responsibility, employee welfare, and corporate governance structure. ESG reports are frequently used to assess a company's long-term risks and opportunities, and are particularly prevalent in green finance, where they are used to measure a company's environmental friendliness and social responsibility.

[0037] Monte Carlo simulation is a numerical computation method based on probability and statistics, primarily used to simulate stochastic processes in complex systems. It generates a large amount of random input data, runs the system model, and observes the output results to analyze the system's behavior and risks. It is commonly used in risk assessment, financial modeling, physics, and engineering. In green credit risk assessment, Monte Carlo simulation can be used to simulate a company's financial performance under different scenarios and assess potential changes in environmental and financial risks.

[0038] Spatiotemporal alignment algorithms are used to process data with temporal and spatial dimensions, primarily aiming to resolve differences or discrepancies between these dimensions. For example, a company's financial data, environmental data, and policy and regulatory data may have different timestamps or collection frequencies. Spatiotemporal alignment algorithms adjust these data to allow for comparison and analysis within the same timeline and spatial framework. In green credit risk assessment, spatiotemporal alignment algorithms ensure accurate alignment between financial and environmental data, enhancing the accuracy and timeliness of the assessment model.

[0039] Causal inference networks are graphical models based on causal relationships, primarily used to model and infer causal relationships between variables. In green credit risk assessment, causal inference networks can help reveal how a company's environmental behavior affects its financial condition or credit default risk. For example, it can be used to analyze the causal relationship between environmental protection behavior and green credit default, identify potential risk factors, and provide a basis for decision-making. By building causal graphs, the system can infer the impact of different environmental decisions on corporate credit, providing a more accurate risk assessment.

[0040] Counterfactual reasoning is a reasoning method used to simulate how the results may change under different assumptions.

[0041] Green credit risk assessment, as a method for evaluating the environmental and financial risks of green credit projects, aims to ensure that credit funds are invested in environmentally friendly, energy-saving, and low-carbon economic sectors that align with sustainable development principles. With the deepening of the global sustainable development concept, financial institutions are playing an increasingly important role in promoting green credit. Green credit risk assessment systems typically consider multiple factors, including a project's environmental friendliness, policy compliance, technological maturity, market prospects, and financial feasibility. Through quantitative indicators and qualitative analysis, they comprehensively evaluate the potential environmental benefits and credit risks of a project. This type of assessment not only helps financial institutions rationally allocate credit resources and promote the development of green industries but also reduces the environmental and financial risks of credit operations, making it an important tool for achieving green finance and sustainable finance development.

[0042] In the implementation of green credit, the accuracy, comprehensiveness, and dynamism of assessment methods are crucial for financial institutions' decision-making. However, existing green credit risk assessment methods face several challenges in application. First, traditional assessment methods often struggle to effectively integrate unstructured environmental data with financial time-series characteristics. Environmental data typically comes from unstructured forms such as reports, public opinion information, or satellite remote sensing data, while financial data is mostly presented as time-series data, making effective integration of these two data types difficult. In particular, existing methods cannot fully verify the authenticity of corporate environmental claims regarding "greenwashing," making it difficult to comprehensively assess environmental risks, thus posing both environmental and financial risks to credit operations.

[0043] Secondly, regarding industry comparisons, many traditional green credit assessment methods use fixed industry benchmarks and thresholds for comparison, failing to consider the dynamic changes within industries. The performance of green credit projects is influenced by various factors, including policy changes, market changes, and technological advancements in the industry. Fixed thresholds are difficult to adapt to changes in industry and markets, easily leading to biased assessment results. For example, when assessing a company's carbon emission intensity or energy consumption, differences in standards across different regions and industries may not be effectively reflected under fixed thresholds, resulting in distorted risk assessments of projects.

[0044] Furthermore, existing assessment methods lack the dynamic perception capability of multi-source data fusion, making it impossible to capture abnormal fluctuations in a company's environmental performance in real time. A company's environmental performance is often influenced by numerous factors, including policy changes, public opinion, and market demand. However, traditional assessment methods mostly rely on static data analysis, which cannot respond to changes in a company's environmental performance in a short period. For example, when a company's environmental investment suddenly increases or decreases, or when its environmental compliance becomes abnormal, traditional methods struggle to capture these changes in a timely manner, thus affecting the risk assessment and funding allocation of green loans.

[0045] Finally, traditional methods often suffer from highly coupled assessment modules, leading to insufficient system scalability and interpretability. Green credit risk assessment involves numerous variables and assessment factors, encompassing multiple dimensions such as financial, environmental, policy, and market factors. Effectively integrating and dynamically analyzing this complex information is a major challenge for existing assessment systems. While modular design helps improve system scalability and flexibility, existing methods often lack sufficient module decoupling, making it difficult to introduce new data sources or assessment dimensions and adapt to constantly changing market demands and policy environments.

[0046] Based on this, embodiments of this disclosure provide a risk assessment method based on green credit. The method includes: acquiring financial and environmental data of a target enterprise; acquiring financial and environmental risk assessment results based on the financial and environmental data, wherein the environmental risk assessment results include at least the verification results of environmental declarations; weightedly fusing the financial and environmental risk assessment results to obtain a target risk level; and obtaining a risk assessment result based on the target risk level. The risk assessment method based on green credit provided by this disclosure, by integrating financial and environmental data and verifying the authenticity of environmental declarations, avoids the problems of data fragmentation and inefficient processing in traditional methods, making the assessment process more efficient and comprehensively improving the accuracy and comprehensiveness of the assessment, thereby effectively improving computational performance. Simultaneously, the integrated financial and environmental risk assessment results provide users with clearer and more reliable risk level information, helping users to more intuitively understand the comprehensive risk status of enterprises, thereby improving the accuracy of decision-making. Verification of the authenticity of environmental declarations can accurately identify greenwashing behavior, further improving the user experience.

[0047] It should be noted that the risk assessment methods, devices, equipment, media, and program products based on green credit as defined in this disclosure relate to the application of large-scale models in the fintech field. They can be used in the fields of artificial intelligence, big data, and fintech, and also in various other fields besides these. The application areas of the risk assessment methods, devices, equipment, media, and program products based on green credit provided in the embodiments of this disclosure are not limited.

[0048] In the technical solution disclosed herein, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation entry points are provided for users to choose to authorize or refuse.

[0049] In scenarios involving automated decision-making using personal information, the methods, devices, and systems provided in this disclosure all offer users corresponding entry points for choosing to agree to or reject the automated decision-making results. If the user chooses to reject, the process proceeds to the expert decision-making stage. Here, "automated decision-making" refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests, or economic, health, and credit status through computer programs, and then making a decision. Here, "expert decision-making" refers to the activity of making decisions by personnel who specialize in a particular field, possess specialized experience, knowledge, and skills, and have reached a certain level of professional expertise.

[0050] Figure 1 The illustration shows an application scenario of a risk assessment method, apparatus, device, medium, and program product based on green credit according to embodiments of the present disclosure.

[0051] like Figure 1 As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0052] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0053] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0054] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0055] It should be noted that the risk assessment method based on green credit provided in this disclosure embodiment can generally be executed by server 105. Correspondingly, the risk assessment device based on green credit provided in this disclosure embodiment can generally be located in server 105. The risk assessment method based on green credit provided in this disclosure embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the risk assessment device based on green credit provided in this disclosure embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105.

[0056] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0057] The following will be based on Figure 1 The described scene, through Figures 2 to 4 The risk assessment method based on green credit, as disclosed in the embodiments, is described in detail.

[0058] Figure 2 A flowchart illustrating a risk assessment method based on green credit according to an embodiment of this disclosure is shown schematically.

[0059] like Figure 2 As shown, the risk assessment method 200 based on green credit in this embodiment includes operations S210 to S240, and the risk assessment method based on green credit can be executed by server 105.

[0060] In operation S210, obtain the target company's financial and environmental data.

[0061] In embodiments of this disclosure, financial dimension data may include the target company's financial statement data, covering multiple financial indicators. To ensure the accuracy and timeliness of the data, it can be obtained through various methods.

[0062] For example, API interfaces that connect to a company's financial system or a third-party financial data provider can be used to automatically obtain the latest financial statement data of the target company. This can include: basic financial indicators such as the company's total assets, total liabilities, net assets, operating revenue, profit, and cash flow; as well as historical time series of various financial indicators, especially the company's financial performance over the past few years, providing time-series data for subsequent financial risk analysis.

[0063] For example, in the absence of an API interface, financial statement data provided by the target company can be manually entered. This data can come from the company's annual reports, quarterly reports, or other financial statement documents, and is verified and entered by financial analysts.

[0064] For example, to enhance the credibility of data, financial data can be extracted from independent audit reports to ensure the reliability and transparency of financial information. This is especially true for larger companies, where reports from third-party audit firms can provide more credible financial information.

[0065] For example, for companies that do not have direct connections, their financial information can be obtained from external databases such as public market information, industry associations, and government statistics, serving as an important supplement to financial evaluation.

[0066] In embodiments of this disclosure, environmental dimension data may include various types of data related to a company’s environmental protection behavior, compliance status, and environmental impact.

[0067] For example, this could include the target company's annual environmental reports, sustainability reports, and ESG (Environmental, Social, and Governance) reports. These reports typically detail the company's environmental information, such as carbon emissions, waste management, resource consumption, and the application of green technologies.

[0068] For example, this could include the target company's past environmental compliance history, especially environmental penalties, warnings, and violations, which reflect the company's historical performance in environmental protection. Environmental compliance history can be obtained from publicly available records of government environmental regulatory departments or industry regulatory bodies.

[0069] For example, in industries with high environmental requirements, target companies may install environmental sensors (such as air quality monitoring equipment and water quality sensors) to monitor their emissions and resource usage. Therefore, real-time environmental data, such as greenhouse gas emissions and wastewater discharge data from factories, can be obtained through IoT devices, providing real-time monitoring information on the company's environmental behavior.

[0070] For example, large enterprises or resource extraction companies can use satellite remote sensing technology for environmental monitoring. By analyzing remote sensing images and geographic data, they can assess the company's environmental impact in certain areas, such as deforestation and mining pollution.

[0071] For example, a company's environmental performance is also influenced by public opinion. By scraping data from online platforms, we can analyze public perception of a company's environmental behavior and assess its environmental reputation.

[0072] In operation S220, based on the financial dimension data and the environmental dimension data, the financial risk assessment results and environmental risk assessment results of the target enterprise are obtained, wherein the environmental risk assessment results include at least the verification results of the authenticity of the environmental protection declaration.

[0073] In the embodiments of this disclosure, financial dimension data and environmental dimension data can be analyzed in depth to obtain corresponding evaluation results.

[0074] For example, financial risk assessment and environmental statement authenticity verification can be based on statistical methods and deep learning. For instance, a statistical regression model can be used to analyze the historical financial data of a target company. By extracting the time-series characteristics of key financial indicators such as cash flow, profitability, and debt-to-equity ratio, the regression model can identify key factors related to the company's financial stability. Furthermore, scenario simulation analysis can be used to analyze the company's financial performance under different market environments, such as simulating the impact of economic recessions and industry contraction on cash flow. For example, semantic parsing can be performed on environmental statements, and a pre-trained BERT model can be used to extract semantic features of the environmental statements, thereby assessing their authenticity.

[0075] For example, environmental risk assessment can be based on traditional financial analysis and industry comparison. For instance, financial ratio analysis can be used to assess a company's profitability, solvency, and other financial risk indicators, and further risk simulation analysis can be conducted, such as the impact of rising interest rates on a company's debt repayment ability. For example, environmental performance data of other companies in the industry can be obtained, and the target company's environmental indicators can be compared with the industry average to generate an industry performance distribution map, thereby identifying whether the target company's environmental performance meets industry standards.

[0076] For example, environmental risk assessment and financial performance simulation can be based on machine learning and behavioral trajectory analysis. For instance, machine learning algorithms can be used to analyze the historical financial data of a target company. By training on financial time-series features, the model can identify complex relationships between different financial indicators and simulate financial performance under different economic environments. For example, behavioral trajectory analysis can be performed on time-series data of environmental inputs, using temporal convolutional networks to detect abnormal fluctuations in environmental protection investments.

[0077] In operation S230, the financial risk assessment results and the environmental risk assessment results are integrated to obtain the target risk level. An appropriate integration method can be flexibly selected based on the specific circumstances of the green credit project.

[0078] For example, the results of financial risk assessment and environmental risk assessment can be integrated using a weighted average method, and the weighted average formula can be used to calculate the target risk level.

[0079] For example, multiple dimensions of scoring can be integrated to consider the interactions between different risk types, combining financial and environmental risks using a multi-dimensional risk scoring model. Specifically, a multi-dimensional risk scoring model can be constructed, employing methods such as machine learning or logistic regression to dynamically adjust the weights of each dimension based on the company's performance in the financial and environmental dimensions. For instance, if a company has a higher environmental risk, the weight of its environmental dimension may increase, thus having a greater impact on the risk assessment.

[0080] For example, the financial risk assessment results and environmental risk assessment results of a company can be converted into a classification label. A threshold rule can then be set based on the specific classification of financial and environmental risks.

[0081] In operation S240, a risk assessment result is obtained based on the target risk level.

[0082] In embodiments of this disclosure, the risk assessment results may take the form of intelligent reports, credit decision support, and / or real-time alerts.

[0083] In embodiments of this disclosure, an interactive 3D assessment report can be automatically generated using a data visualization engine to display the risk assessment results. The interactive 3D assessment report can integrate various graphical data representation formats, such as risk heatmaps, time-series trend charts, and industry benchmarking radar charts. For example, a risk heatmap can show the risk distribution of a target company in different regions and dimensions, a time-series trend chart can show the changing trends of a company's financial and environmental performance over time, and an industry benchmarking radar chart helps compare the environmental performance of the target company with that of companies in the same industry.

[0084] Furthermore, the interactive 3D assessment report allows users to flexibly view assessment results from different dimensions by adjusting parameters. For example, users can choose to view a comprehensive assessment result that combines financial risk, environmental risk, or both.

[0085] In the embodiments of this disclosure, credit decision support can dynamically generate differentiated interest rate fluctuation ranges based on the target risk level. Specifically, based on the financial and environmental risk assessment results of the target enterprise, credit decision support can provide an appropriate credit interest rate range for the target enterprise. For example, when an enterprise has a good financial condition but a high environmental risk, the credit interest rate may be higher to offset the potential risk. In addition to generating interest rate ranges based on assessment results, credit decision support can also combine the specific characteristics of the target enterprise to match applicable financial instruments in the green credit product library. This ensures that green credit products are customized to suit the financial condition and environmental protection needs of different enterprises, helping financial institutions better allocate credit resources and promote the development of green industries.

[0086] In the embodiments disclosed herein, a red alert triggering mechanism for greenwashing behavior can be set to monitor the environmental performance of target enterprises in real time and promptly push high-risk alerts to the risk control department. Specifically, when the target enterprise's target risk level reaches a preset value, the system will automatically generate an alert to remind the risk control department to pay attention to potential credit risks.

[0087] Figure 3 A flowchart illustrating a method for obtaining financial risk assessment results according to an embodiment of this disclosure is shown schematically.

[0088] like Figure 3 As shown, the method for obtaining financial risk assessment results in this embodiment may include operations S310 to S330.

[0089] In operation S310, the time-series features of the financial dimension data are extracted, and the correlation strength between financial accounts is obtained based on the time-series features. Financial accounts are standardized accounting units used by the system to classify, record, and report economic transactions. For example, they may include economic attribute dimensions, such as asset accounts and liability accounts; they may include accounting level accounts, such as general ledger accounts and subsidiary ledger accounts.

[0090] In the embodiments of this disclosure, financial dimension data is typically presented in time series format. To better understand a company's financial situation and its changing trends, it is necessary to extract time series features from the time series data. Time series features may include factors such as trends (e.g., long-term growth or decline in revenue), seasonality (e.g., sales peaks at specific times each year), and cyclicality (e.g., the impact of economic cycles on company finances). Specifically, time series feature extraction can be performed using an LSTM (Long Short-Term Memory) network model.

[0091] In the embodiments of this disclosure, a bidirectional LSTM network can be used to calculate the correlation strength between different financial items. The bidirectional LSTM network can not only capture past dependencies but also improve the understanding of the relationships between financial items by considering the impact of future data. For example, there is often a dependency between a company's cash flow and debt ratio; a bidirectional LSTM can simultaneously consider the bidirectional impact of both, thereby providing more accurate information for subsequent financial assessments.

[0092] In operation S320, a target scenario simulation is performed based on the aforementioned time-series characteristics and the correlation strength to obtain the financial performance simulation results. The target scenario can be extreme scenarios such as rising interest rates or declining revenue, thereby obtaining the financial resilience performance of the target company under extreme scenarios.

[0093] In embodiments of this disclosure, extreme scenarios can be simulated using a combination of Monte Carlo simulation and scenario analysis.

[0094] In operation S330, the financial risk assessment result is obtained based on the financial performance simulation results. Specifically, the financial performance simulation results can be compared with preset financial health standards to output a comprehensive financial risk assessment result.

[0095] Figure 4 A flowchart illustrating a method for obtaining environmental risk assessment results according to an embodiment of this disclosure is shown schematically.

[0096] like Figure 4 As shown, the method for obtaining environmental risk assessment results in this embodiment may include operations S410 to S440.

[0097] In operation S410, semantic parsing and authenticity verification are performed based on the environmental declaration data of the target company to obtain the authenticity verification results of the environmental declaration.

[0098] Preferably, the target company's environmental declaration data can be semantically decomposed into three core parts: environmental commitment data, environmental measures data, and environmental effectiveness data. Environmental commitment data includes the company's commitments to environmental protection, such as reducing carbon emissions and improving energy efficiency; environmental measures data includes the specific environmental measures implemented by the company, such as equipment upgrades and the application of pollution control technologies; and environmental effectiveness data includes the actual effects of the company's implemented measures, such as the reduction in carbon emissions and energy consumption.

[0099] Furthermore, the correlation and consistency between environmental commitments, environmental measures, and environmental outcomes can be assessed through matching degree calculations. Specifically, techniques such as logistic regression models can be used to calculate the matching degree between environmental commitment data and environmental measure data (ScoreCM), and the matching degree between environmental measure data and environmental outcome data (ScoreME). For example, if a company commits to implementing a certain environmental technology, and that technology has been successfully implemented and has brought about substantial results, then the matching degree between the environmental commitment and the measures is high, and the matching degree between the environmental measures and the outcomes is also high. Finally, the environmental statement's truthfulness score is obtained by multiplying the two matching degrees:

[0100] Scoretruth=ScoreCM×ScoreME (1)

[0101] This rating comprehensively measures the authenticity of a company's environmental statements and reflects the company's actual performance in the field of environmental protection.

[0102] Building upon the accuracy calculation, embodiments of this disclosure further employ fuzzy expression recognition technology to further evaluate the accuracy of environmental claims. Environmental claims often contain vague statements such as "striving to promote" or "continuously improving," which often lack clear quantitative targets and measurement standards. By using text classification algorithms and fuzzy recognition devices, vague expressions in environmental claims can be identified, and corresponding fuzzy expression tags can be generated. For example, if a claim uses "we plan to reduce carbon emissions" without mentioning specific reduction targets or timeframes, this part will be marked as a fuzzy expression. Fuzzy expression tagging helps further evaluate the executability and specificity of the claim, ensuring the accuracy of the accuracy verification results. Specifically:

[0103] Let F be the set of fuzzy representations and C be the set of non-fuzzy representations. Using a text classification algorithm, calculate the probability P(T∈F) that a text T belongs to F. If P(T∈F) exceeds a preset threshold θ, then the statement is considered to contain a fuzzy representation.

[0104] In the embodiments disclosed herein, a large language model can also be used to perform in-depth semantic analysis on the environmental statements submitted by the target company to verify the authenticity, compliance, and potential greenwashing risks of the content.

[0105] For example, the large language model can be fine-tuned for the environmental field to automatically identify core elements in statements, including corporate environmental commitments, actual environmental measures, and achieved environmental results. Through in-depth text analysis, the large language model can extract key information such as quantitative indicators, target implementation status, and timelines, and calculate the authenticity score for each statement. The authenticity score can comprehensively consider vague expressions, missing quantitative targets, and actual evidence of environmental effects in the statement. Specifically, a greenwashing rhetoric feature detector can be developed to capture the semantic disconnect between embellished expressions and substantive actions in the text through an attention mechanism. Combined with contextual coherence analysis, it can expose selective disclosure behavior and compare semantic consistency with the company's actual behavior trajectory to identify potential subjective risks such as semantic greenwashing, false disclosure, and logical contradictions, further enhancing the credibility of environmental dimension scores and the ability to control subjective risks.

[0106] When operating S420, the environmental performance of the target company is compared with industry benchmark data to obtain the industry benchmark comparison results.

[0107] In the embodiments of this disclosure, a distribution map of industry environmental performance can be dynamically generated using the kernel density estimation method for enterprise performance comparison. Let the enterprise environmental performance index be I, and the set of environmental performance indices for all enterprises in the industry be I0. i The formula for calculating the KDE distribution function f(I) is:

[0108] (2)

[0109] Where K is the kernel function, h is the bandwidth, and N is the number of firms.

[0110] Furthermore, we can analyze the changes in the target company's environmental performance percentile ranking within the industry to identify whether the company's environmental performance has improved or declined within the industry. The calculation formula is: Let the company's current environmental performance percentile be Qt, and the previous period's percentile be Qt-1.

[0111] Quantile migration ΔQ = Qt - Qt-1 (3)

[0112] By operating S430, we analyze the environmental behavior trajectory of the target enterprise and obtain the behavior trajectory analysis results.

[0113] In the embodiments of this disclosure, a TCN (Temporal Convolutional Network) model can be used to analyze the time series data of corporate environmental protection investments to detect any abnormal fluctuations and obtain behavioral trajectory analysis results. Specifically, let the time series of corporate environmental protection investments be: X=[x1,x2,...,x...] TThe TCN model output is Y=[y1,y2,...,y]. T [] represents the predicted value at each time point. Fluctuation detection can be achieved by calculating the difference between X and Y: D=XY.

[0114] In embodiments of this disclosure, a matrix can be constructed to assess whether a company's environmental investment growth matches its capacity expansion. Specifically, the environmental investment growth rate can be denoted as renov, and the capacity expansion rate as rcap. A reasonableness score S is calculated by comparing the relationship between the two.

[0115] S=renv / rcap (4)

[0116] In operation S440, an environmental risk assessment result is generated based on the verification results of the environmental declaration's authenticity, the industry benchmark comparison results, and the behavioral trajectory analysis results. Specifically, the verification results of the environmental declaration's authenticity reflect the actual fulfillment of the company's environmental commitments, the industry benchmark comparison results demonstrate the company's relative environmental performance within the industry, and the behavioral trajectory analysis results reveal the stability and potential problems of the company's environmental performance. By comprehensively considering these factors, the level of environmental risk that the company may face in the future can be assessed.

[0117] In the embodiments disclosed herein, the financial and environmental data of the target enterprise are comprehensively acquired through a multi-data acquisition method involving structured data sources, unstructured data sources, and dynamic data streams. This ensures that the financial status and environmental performance of the enterprise are fully covered, thereby guaranteeing the integrity and accuracy of the data during the assessment process.

[0118] Specifically, structured data sources refer to easily accessible quantitative data stored in standardized formats. For financial data, structured data sources can include a company's financial statements, balance sheets, income statements, and cash flow statements. For environmental data, structured data sources can include a company's environmental compliance records, environmental law compliance status, carbon emission data, and environmental penalty records. This data can also be obtained through standardized data interfaces and integrated with other data sources.

[0119] Unstructured data sources refer to data types that are difficult to store and query using traditional databases, such as text data, images, audio, or video data. In green credit risk assessment, unstructured data sources mainly include textual information such as environmental statements, environmental reports, news reports, and public opinion released by companies, as well as environmental monitoring data such as satellite remote sensing imagery. For financial data, unstructured data sources can include news reports, analytical articles, and investor relations information related to a company's financial situation. For environmental data, unstructured data sources mainly involve documents such as environmental reports, social responsibility reports, and environmental statements released by companies.

[0120] Dynamic data streams refer to real-time, constantly changing data streams that provide immediate insights into corporate performance and environmental changes. In green credit risk assessment, dynamic data streams primarily include real-time environmental monitoring data (such as pollutant emission monitoring data), real-time market price fluctuation data (such as carbon price changes in the carbon trading market), and real-time public opinion feedback. For financial data, dynamic data streams can include real-time data such as stock prices, bond interest rates, and market trends obtained through API interfaces, reflecting the market's real-time assessment of the company's financial health. For environmental data, dynamic data streams can include real-time data on the company's environmental protection investments, pollution emissions, and resource consumption, as well as real-time updates on environmental policies from governments and environmental organizations.

[0121] By combining structured data sources, unstructured data sources, and dynamic data streams, we can ensure comprehensive data coverage, timely information delivery, and provide rich and reliable data support for subsequent risk assessments.

[0122] In the embodiments of this disclosure, to ensure that financial and environmental data can be comprehensively analyzed on the same platform, multi-source data fusion and spatiotemporal alignment can be performed on these data. Preferably, a spatiotemporal alignment algorithm can be used to align data from different sources according to a unified timestamp based on the target enterprise's unified social credit code and other identifying information. For example, the timestamps for financial data may be provided monthly or quarterly, while environmental data may be recorded daily or hourly. The spatiotemporal alignment algorithm can unify these data to the same time granularity, enabling time-series data from different sources to be compared and analyzed within the same analytical framework. Through spatiotemporal alignment, synchronized multi-source data can be obtained, ensuring that financial and environmental data are compared within the same time frame and avoiding deviations in analysis results due to inconsistent data timing.

[0123] After acquiring data from multiple sources simultaneously, further standardization and feature extraction can be performed. For example, for financial data such as revenue and profit, Z-score standardization can be used to convert them into a standard normal distribution with a mean of 0 and a standard deviation of 1, thereby eliminating dimensional differences between different data dimensions. For environmental data such as carbon emissions and resource consumption, normalization can be performed according to industry benchmarks to ensure a consistent range of data values, facilitating subsequent analysis.

[0124] Building upon standardized processing, the embodiments of this disclosure can also extract features from synchronous multi-source data to extract key features that reflect a company's financial and environmental performance. For example, from financial data, indicators of financial health such as revenue growth rate and debt-to-equity ratio can be extracted; from environmental data, indicators of environmental performance such as carbon emission intensity and environmental protection investment growth rate can be extracted.

[0125] Furthermore, data gaps are a common problem in actual green credit risk assessment, especially in environmental monitoring and financial reporting, where data may be missing or incomplete for various reasons. Therefore, missing values ​​can be filled into standardized feature data to ensure data integrity.

[0126] In the embodiments disclosed herein, in order to comprehensively assess the green credit risk of the target enterprise, a causal graph between environmental protection behavior and green credit default can also be constructed. The causal graph can be constructed with the target enterprise's environmental protection behavior (such as environmental investment, environmental compliance, changes in carbon emissions, etc.) and green credit default risk (such as the probability of corporate default, debt repayment ability, etc.) as the main nodes.

[0127] Furthermore, counterfactual reasoning can be performed based on the constructed causal graph and target environmental decision-making scenarios to simulate the potential impact of different environmental decisions on green credit default risk. Counterfactual reasoning can assess a company's possible credit performance under different environmental behaviors by constructing different hypothetical scenarios, thereby helping financial institutions understand how a company's financial condition and credit default probability change under specific environmental policy changes.

[0128] Based on the constructed causal graph and counterfactual reasoning results, further causal chain analysis can be conducted to reveal the transmission mechanism between environmental protection behavior and green credit default. Causal chain analysis can identify key drivers influencing green credit default risk by tracing the impact path between corporate environmental behavior and credit default risk. Therefore, the results of causal chain analysis can not only help identify the direct impact of corporate environmental behavior on financial condition and debt repayment ability, but also reveal its indirect impact, such as the role of external factors like policy changes and industry trends.

[0129] By analyzing causal chain results, financial institutions can clearly understand which environmental protection measures are most effective in reducing the risk of green loan defaults under different circumstances. For example, if certain environmental practices can significantly improve a company's financial health and thus reduce its default risk, financial institutions can prioritize these companies and offer them more favorable green loan products. Conversely, if a company's environmental statements do not match its actual practices, or if its environmental performance fails to meet standards, it may increase the risk of default, thereby affecting credit decisions.

[0130] In the embodiments of this disclosure, in order to further improve the accuracy and adaptability of green credit risk assessment, a causal discovery algorithm is also adopted to dynamically update the direction of interaction between variables in the causal graph. The causal graph is automatically reconstructed according to changes in external factors, thereby ensuring that the causal graph can reflect new risk patterns in a timely manner when policies, market environment or other external factors change, thereby improving the flexibility and accuracy of green credit decision-making.

[0131] Specifically, causal discovery algorithms can automatically detect causal relationships between variables based on various inputs, including a target company's financial data, environmental data, and policy changes. For example, they can determine how financial performance is affected by environmental protection practices, or how changes in green credit policies affect a company's debt repayment ability. The algorithm uses a data-driven approach, leveraging statistical and machine learning methods (such as Granger causality tests and Bayesian networks) to continuously optimize and update the causal graph. This algorithm can not only process existing data but also identify potential new causal relationships, allowing the causal graph to continuously evolve with the input of new data.

[0132] In the field of green credit, changes in policies and regulations have a significant impact on enterprises' environmental behavior and their credit default risks. Therefore, timely response to policy and regulatory changes is crucial for green credit risk assessment. The embodiments of this disclosure utilize a causal graph reconstruction mechanism to automatically trigger updates to the causal graph upon detecting policy and regulatory changes, thereby reflecting the impact of new policy changes on green credit risk.

[0133] For example, new environmental policies or carbon emission regulations introduced by the government may lead to changes in companies' environmental behavior, thereby affecting their credit risk levels. In this case, the causal discovery algorithm will reassess the causal relationships between variables based on the new policy information and adjust the direction of action in the causal graph. Through causal graph reconstruction, the system can automatically identify and adapt to new policy changes, ensuring that risk assessment results are always consistent with the latest market and policy environment.

[0134] The embodiments disclosed herein comprehensively assess a company's financial and environmental risks by integrating financial and environmental data. Specifically, through time-series feature analysis and scenario simulation of financial data, the correlation strength between financial items is analyzed based on these features, and target scenario simulations are conducted to obtain simulated financial performance results, further assessing the company's financial risks and improving the accuracy and scientific rigor of the financial assessment. Simultaneously, through verification of the authenticity of environmental statements, comparison with industry benchmarks, and behavioral trajectory analysis, it helps to identify potential inaccuracies in environmental information, enhancing the credibility of the environmental risk assessment. Data collection encompasses structured, unstructured, and dynamic data streams, and multi-source data fusion and spatiotemporal alignment ensure data integrity and consistency, further improving the accuracy of risk assessment. Furthermore, a causal graph is constructed for counterfactual reasoning, combined with a dynamic causal discovery algorithm to address policy changes, making the assessment results more accurate, timely, and adaptable, thus providing a scientific basis for green credit decisions.

[0135] Figure 5 A schematic block diagram of a risk assessment apparatus based on green credit according to an embodiment of the present disclosure is shown.

[0136] like Figure 5 As shown, the risk assessment device 500 based on green credit in this embodiment includes a data acquisition module 510, a risk assessment module 520, a result integration module 530, and an assessment result acquisition module 540.

[0137] The data acquisition module 510 can be used to acquire financial and environmental data of the target enterprise. In one embodiment, the data acquisition module 510 can be used to perform the operation S210 described above, which will not be repeated here.

[0138] The risk assessment module 520 can be used to obtain the financial risk assessment results and environmental risk assessment results of the target enterprise based on the financial dimension data and the environmental dimension data, wherein the environmental risk assessment results include at least the verification results of the authenticity of the environmental protection declaration. In one embodiment, the risk assessment module 520 can be used to perform the operation S220 described above, which will not be repeated here.

[0139] The result integration module 530 can be used to integrate the financial risk assessment results and the environmental risk assessment results to obtain the target risk level. In one embodiment, the result integration module 530 can be used to perform the operation S230 described above, which will not be repeated here.

[0140] The assessment result acquisition module 540 can be used to obtain a risk assessment result based on the target risk level. In one embodiment, the assessment result acquisition module 540 can be used to perform the operation S240 described above, which will not be repeated here.

[0141] According to embodiments of this disclosure, the risk assessment module 520 can also be used to extract the time-series features of the financial dimension data, obtain the correlation strength between financial items based on the time-series features; perform target scenario simulation based on the time-series features and the correlation strength to obtain financial performance simulation results; and obtain the financial risk assessment results based on the financial performance simulation results.

[0142] According to embodiments of this disclosure, the risk assessment module 520 can also be used to obtain multiple multimodal feature vectors output by the second sub-network, align the multimodal feature vectors to obtain a modality alignment vector set; perform correlation modeling based on the modality alignment vector set to obtain attention weight information; and perform weighted fusion of the multimodal feature vectors based on the attention weight information to obtain the transaction behavior feature vector.

[0143] According to embodiments of this disclosure, the risk assessment module 520 can also be used to perform semantic parsing and authenticity verification based on the environmental declaration data of the target enterprise, and obtain the authenticity verification result of the environmental declaration; compare the environmental performance of the target enterprise with industry benchmark data, and obtain the industry benchmark comparison result; analyze the environmental behavior trajectory of the target enterprise, and obtain the behavior trajectory analysis result; and generate an environmental risk assessment result based on the authenticity verification result of the environmental declaration, the industry benchmark comparison result and the behavior trajectory analysis result.

[0144] According to embodiments of this disclosure, the risk assessment module 520 can also be used to semantically decompose the environmental declaration data into environmental commitment data, environmental measures data, and environmental effectiveness data; calculate the matching degree based on the environmental commitment data, environmental measures data, and environmental effectiveness data to obtain an authenticity score; perform fuzzy recognition on the environmental declaration data to obtain fuzzy expression tags; and obtain the authenticity verification result of the environmental declaration based on the authenticity score and the fuzzy expression tags.

[0145] According to embodiments of this disclosure, the risk assessment device 500 can also be used to perform multi-source data fusion and spatiotemporal alignment on the financial dimension data and the environmental dimension data to obtain synchronous multi-source data; to perform standardization processing and feature extraction on the synchronous multi-source data to obtain standardized feature data; and to fill missing values ​​in the standardized feature data to obtain filled data.

[0146] According to embodiments of this disclosure, the risk assessment device 500 can also be used to construct a causal graph between environmental protection behavior and green credit default; perform counterfactual reasoning based on the causal graph and the target environmental decision-making scenario to obtain counterfactual reasoning results; and obtain causal chain analysis results based on the causal graph and the counterfactual reasoning results.

[0147] According to embodiments of this disclosure, the risk assessment device 500 can also be used to integrate a causal discovery algorithm, which is used to dynamically update the direction of interaction between variables in the causal graph; and to trigger the causal discovery algorithm to reconstruct the causal graph in response to the detection of policy and regulatory changes.

[0148] According to embodiments of this disclosure, any multiple modules among the data acquisition module 510, risk assessment module 520, result integration module 530, and evaluation result acquisition module 540 can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this disclosure, at least one of the data acquisition module 510, risk assessment module 520, result integration module 530, and evaluation result acquisition module 540 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the data acquisition module 510, risk assessment module 520, result integration module 530, and assessment result acquisition module 540 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0149] Figure 6 A block diagram schematically illustrates an electronic device suitable for implementing a risk assessment method based on green credit, according to an embodiment of the present disclosure.

[0150] like Figure 6 As shown, an electronic device 600 according to an embodiment of the present disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage portion 606 into a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0151] RAM 603 stores various programs and data required for the operation of electronic device 600. Processor 601, ROM 602, and RAM 603 are interconnected via bus 604. Processor 601 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 602 and / or RAM 603. It should be noted that the programs may also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0152] According to embodiments of this disclosure, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to a bus 604. The electronic device 600 may also include one or more of the following components connected to the input / output (I / O) interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 610 as needed so that computer programs read from it can be installed into the storage section 608 as needed.

[0153] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0154] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include ROM 602 and / or RAM 603 and / or one or more memories other than ROM 602 and RAM 603 described above.

[0155] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement the risk assessment method based on green credit provided in embodiments of this disclosure.

[0156] When the computer program is executed by the processor 601, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0157] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 609, and / or installed from the removable medium 611. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0158] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0159] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can execute entirely on a user's computing device, partially on a user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0160] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0161] Those skilled in the art will understand that the features described in the various embodiments of this disclosure can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments of this disclosure can be combined and / or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0162] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A risk assessment method based on green credit, characterized in that, The method includes: Obtain financial and environmental data from the target company; Based on the financial dimension data and the environmental dimension data, obtain the financial risk assessment results and environmental risk assessment results of the target enterprise, wherein the environmental risk assessment results include at least the verification results of the authenticity of the environmental protection declaration; The financial risk assessment results and the environmental risk assessment results are integrated to obtain the target risk level; and Based on the target risk level, a risk assessment result is obtained.

2. The method according to claim 1, characterized in that, The financial risk assessment results of the target company are obtained based on the aforementioned financial dimension data, specifically including: Extract the temporal features of the financial dimension data, and obtain the correlation strength between financial items based on the temporal features; Based on the aforementioned time-series features and the aforementioned correlation strength, a target scenario simulation is performed to obtain financial performance simulation results; and The financial risk assessment results are obtained based on the simulated financial performance results.

3. The method according to claim 1 or 2, characterized in that, The environmental risk assessment results are obtained based on the aforementioned environmental dimension data, specifically including: Semantic analysis and authenticity verification are performed on the environmental declaration data of the target company to obtain the authenticity verification results of the environmental declaration; The environmental performance of the target company is compared with industry benchmark data to obtain industry benchmark comparison results; Analyze the environmental behavioral trajectories of the target company and obtain the behavioral trajectory analysis results; and Based on the verification results of the environmental protection statement's authenticity, the comparison results with industry benchmarks, and the analysis results of behavioral trajectories, an environmental risk assessment result is generated.

4. The method according to claim 3, characterized in that, The process of semantic parsing and authenticity verification based on the environmental declaration data of the target enterprise to obtain the authenticity verification results of the environmental declaration specifically includes: The environmental declaration data is semantically decomposed into environmental commitment data, environmental measures data, and environmental effectiveness data. Based on the environmental commitment data, environmental measures data, and environmental effectiveness data, the matching degree is calculated to obtain an authenticity score. The environmental statement data is subjected to fuzzy recognition to obtain fuzzy representation tags; and Based on the authenticity score and the fuzzy expression marker, the authenticity verification result of the environmental protection statement is obtained.

5. The method according to any one of claims 1, 2, and 4, characterized in that, The financial dimension data and environmental dimension data of the target enterprise are obtained based on structured data sources, unstructured data sources, and dynamic data streams.

6. The method according to claim 5, characterized in that, The method further includes: Multi-source data fusion and spatiotemporal alignment are performed on the financial dimension data and the environmental dimension data to obtain synchronous multi-source data; The synchronized multi-source data is standardized and its features are extracted to obtain standardized feature data; and Missing values ​​are filled into the standardized feature data to obtain the filled data.

7. The method according to any one of claims 1, 2, 4, and 6, characterized in that, The method further includes: Constructing a causal graph between environmental behavior and green credit defaults; Based on the causal diagram and the target environmental decision-making scenario, counterfactual reasoning is performed to obtain counterfactual reasoning results; and The causal chain analysis results are obtained based on the causal graph and the counterfactual reasoning results.

8. The method according to claim 7, characterized in that, The method further includes: An integrated causal discovery algorithm is used to dynamically update the interaction directions between variables in the causal graph; and In response to the detection of policy and regulatory changes, the causal discovery algorithm is triggered to reconstruct the causal graph.

9. A risk assessment device based on green credit, characterized in that, The device includes: The data acquisition module is used to acquire financial and environmental data of the target company. The risk assessment module is used to: obtain the financial risk assessment results and environmental risk assessment results of the target enterprise based on the financial dimension data and the environmental dimension data, wherein the environmental risk assessment results include at least the verification results of the authenticity of the environmental protection declaration; The results integration module is used to: integrate the financial risk assessment results and the environmental risk assessment results to obtain the target risk level; and The assessment result acquisition module is used to obtain risk assessment results based on the target risk level.

10. An electronic device, comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 8.

12. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 8.

Images