Risk data management system and method based on data analysis

Abstract

The invention discloses a risk data management system and method based on data analysis, and relates to the technical field of risk evaluation.The risk data management method comprises the steps that all risk data under different business scenes are collected in enterprise historical information, and a risk conduction rule base is defined through the risk data; extracting a business main body from the risk conduction rule base as a node, extracting a risk influence path as a relation edge, and constructing a scenarized sub-graph; when risk detection is carried out in a real-time business scene and risk source data is acquired, calculating a scene coupling degree by using a risk triggering probability, scene influence intensity and risk absorption capability; constructing a risk decision decoupling engine, judging the calculated scene coupling degree, and screening risk source data collected in the real-time service scene; and according to the screened risk source data, a risk source data conduction path is extracted from the scenarized sub-atlas, a risk influence range is constructed, and a risk report is generated and output.

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment, and in particular to a risk data management system and method based on data analysis. Background Art

[0002] In today's digital age, businesses face an increasingly complex risk landscape, encompassing market risk, credit risk, and operational risk. Traditional risk management relies primarily on experience, intuition, and expert judgment, with limited information technology capabilities. This makes it difficult to cope with massive amounts of data and complex and changing risk factors. Data analytics are urgently needed to improve the efficiency and accuracy of risk management. Big data can be used to collect and analyze large amounts of risk data, providing more comprehensive and accurate information for risk management. Artificial intelligence can build risk assessment models and conduct risk forecasting, enabling automated risk identification and assessment. Cloud computing provides flexible and scalable computing resources for risk management information systems, reducing system construction and operating costs.

[0003] However, when conducting systematic management of risk data today, it is difficult to effectively adapt to different scenarios. Due to the diversity of risk data, different risk data are considered for different business scenarios. In this business scenario, although some risk data is high-risk, it will not have an impact on the business. However, when managing the system, it is often impossible to make targeted judgments, and warnings will be issued for all risk data in different business scenarios. This may lead to the loss of potential customers in actual business scenarios and cause unnecessary losses. Summary of the Invention

[0004] The purpose of the present invention is to provide a risk data management system and method based on data analysis to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solutions: A risk data management method based on data analysis, the method comprising the following steps: S100. Collect all risk data under different business scenarios from the enterprise's historical information, the risk data including risk type, risk trigger conditions, and risk impact path; and define a risk transmission rule base using the risk data; Furthermore, the specific steps for defining the risk transmission rule base using risk data are as follows: S101. Collect all risk data under different business scenarios in the enterprise's historical information. The risk data includes risk type, risk trigger conditions and risk impact path, and define risk transmission rules as Rule s ={RiskType,TriggerCondition,ImpactPath}, where Rule sIndicates the risk transmission rule for the s-th business scenario, RiskType indicates the risk type, TriggerCondition indicates the risk trigger condition, and ImpactPath indicates the risk impact path; S102. Define risk transmission rules for each business scenario, and integrate the risk transmission rules of all business scenarios to build a risk transmission rule library.

[0006] Business scenarios include supply chain cooperation and credit approval. Risk types include financial anomalies, personal disputes, and corporate disputes. For example, in a corporate dispute, a risk is triggered when the dispute amount exceeds 10% of the corporate cooperation. For example, the risk impact path is as follows: in a corporate dispute, corporate credit anomaly → account freeze → loan payment failure. An example of a risk transmission rule is: {Legal litigation, amount involved > total order amount × 0.1, funds frozen → supply cut-off}; By collecting risk types, triggering conditions, and impact paths, we clearly define the key components of risk, providing a precise analysis target and foundational framework for subsequent risk analysis. The establishment of a rule base provides data support for the construction of scenario-based sub-graphs, ensuring that subsequent risk analysis is well-defined and based on reliable historical data.

[0007] S200: Extract business entities as nodes from the risk transmission rule base. Business entities include risk source entities, business terminal entities, and risk buffer entities. Extract risk impact paths as relationship edges to construct a scenario-based subgraph. Furthermore, the specific steps to construct the scenario-based sub-graph are as follows: S201. For each business scenario, extract the business subject as a node in the risk transmission rule library. The specific business entities include Entities={E risk , E biz , E buffer}, where E risk Represents the risk source entity, E biz Indicates the business terminal entity, E buffer represents a risk buffer entity; Examples of risk source entities include: low-credit enterprises, dishonest individuals, and corporate disputes; business terminal entities include: purchase orders, loan contracts, etc.; risk buffer entities include: alternative suppliers, collateral, etc. S202, extract the risk impact path in the risk transmission rule base as the relationship edge between the corresponding business entities Edges={(e i , e j , r ij )|e i →r ij →e j}, where ei Indicates the i-th business entity node, e j Represents the jth business entity node, r ij Represents the conduction relationship between the i-th business entity node and the j-th business entity node, e i →r ij →e j Indicates the direction of conduction; S203, connect all business entity nodes and relationship edges to build a scenario sub-graph G s = (V, E), where G s Represents the scenario-based subgraph of the s-th business scenario, V represents the relationship edge set, and E represents the business entity node set; set the hard constraint as follows: all business entity nodes in the constructed scenario-based subgraph are connected to the business terminal entity nodes.

[0008] Clearly distinguishing risk sources, endpoints, and buffer entities allows for rapid identification of risk starting points, endpoints, and intermediate buffers, focusing on key targets for subsequent risk probability calculations and impact analysis. Hard constraints on connectivity between all nodes and business endpoint entities ensure the integrity and relevance of the graph, preventing the omission of important risk transmission paths and ensuring comprehensive risk analysis.

[0009] S300: When risk detection is performed in a real-time business scenario and risk source data is collected, the risk trigger probability is calculated based on the path length from the risk source entity to the business terminal entity in the scenario-based sub-graph. The loss value of the affected node in the scenario-based sub-graph is extracted to calculate the scenario impact intensity. The risk buffer node is extracted to calculate the risk absorption capacity. The scenario coupling degree is calculated using the risk trigger probability, scenario impact intensity, and risk absorption capacity. Furthermore, the specific steps for calculating scenario coupling using risk trigger probability, scenario impact intensity, and risk absorption capacity are as follows: S301. When risk source data is collected in a real-time business scenario, the shortest transmission path length d from the risk source node to the business terminal node corresponding to the collected risk source data is extracted in the scenario-based sub-graph. The number of times the risk source data in the business scenario causes a risk to occur within a historical time T is collected. The basic risk probability of the risk source data is obtained by dividing the number of risk occurrences by the time T. The risk trigger probability is calculated using the transmission path. The formula is: ; In the formula, P represents the risk trigger probability, α represents the basic risk probability, β represents the attenuation coefficient, and d represents the length of the shortest transmission path. e represents a natural constant. d is the number of edges between the risk source node and the business terminal node in the scenario-based subgraph, which is dimensionless. By combining the basic risk probability, attenuation coefficient and the shortest transmission path length, the attenuation effect of risk during the transmission process is taken into account, making the calculation results more in line with the actual possibility of risk occurrence and more accurate than simple basic probability.

[0010] S302: Extract the number of business entity nodes affected by the risk source node and the loss value of the business entity corresponding to the node from the scenario-based sub-graph. Calculate the scenario impact intensity using the business entity nodes affected by the risk source node and the loss value. The formula is: ; In the formula, I represents the impact of the scenario, H represents the loss value of the business entity corresponding to the node, and the loss values ​​of the business entities corresponding to different nodes are standardized before participating in the calculation; N represents the number of business entity nodes affected by the risk source node, and w a Indicates the weight of the a-th business entity node affected by the risk source node. The weight is set by the business personnel; By weighted summing up the loss values ​​of the affected nodes, the importance (weight) and degree of loss of different nodes are comprehensively considered, objectively reflecting the impact of risks on business scenarios and providing quantitative indicators for risk assessment.

[0011] S303. Extract all risk buffer nodes in the scenario-based sub-graph. Business personnel conduct professional assessments on the entities and risk source data corresponding to the risk buffer nodes to obtain the buffering degree of the risk buffer nodes and the risk intensity of the risk source data. The risk absorption capacity of the business scenario is calculated using the standardized buffering degree and risk intensity. The formula is: ; In the formula, A represents the risk absorption capacity of the business scenario, Bu m represents the buffering degree of the mth risk buffer node, Ri represents the risk intensity of the risk source data, and M represents the number of risk buffer nodes in the scenario sub-graph; S304. Calculate the scenario coupling degree using the risk trigger probability, scenario impact intensity, and risk absorption capacity. Standardize the three data to eliminate the dimension. The formula is: ; In the formula, SAC represents the scenario coupling degree of risk source data collected in real time in the business scenario.

[0012] Integrating the risk trigger probability, impact intensity and absorption capacity to calculate the scenario coupling degree realizes a multi-dimensional comprehensive assessment of risks. A single indicator cannot fully reflect the risk situation, while the coupling degree can comprehensively reflect the overall risk situation.

[0013] S400: Build a risk decision decoupling engine to determine the calculated scenario coupling degree and filter the risk source data collected in real-time business scenarios. Furthermore, the specific steps for screening risk source data collected in real-time business scenarios are as follows: S401. Build a risk decision decoupling engine, specifically: ; In the formula, Decision represents risk decision, Accept represents acceptance of risk source data, Reject represents rejection of risk source data, and θ represents the scenario coupling threshold, which is manually configured by professionals. S402. Set constraints, specifically: collect the average loss value caused by the impact of historical risk source data on business entities, standardize it and use it as the basic risk score Iy. When I>Iy, directly reject the risk source data and do not execute the risk decision decoupling engine.

[0014] By setting scenario coupling thresholds and basic risk constraints, we can effectively filter out low-risk data, focus on high-risk data, reduce ineffective analysis, and improve risk decision-making efficiency. The construction of a risk decision decoupling engine standardizes and automates risk decision-making while allowing professionals to configure thresholds. This balances scientific and flexible decision-making and adapts to risk preferences in different business scenarios.

[0015] S500. Based on the screened risk source data, extract the risk source data transmission path in the scenario-based sub-graph, construct the risk impact range, and generate a risk report output.

[0016] Furthermore, the specific steps to generate the risk report output are: S501. Based on the filtered risk source data, extract the risk source data transmission path in the scenario-based sub-graph. Nodes that have a transmission path to the risk source node are identified as affected nodes. All affected nodes are marked with different colors, and the different-colored areas are identified as the risk impact range. S502: Divide the risk report into a three-tier structure, specifically raw risk, scenario analysis, and decision recommendations. Integrate all risk source data detected in the business scenario to generate raw risk. Display the calculated scenario coupling degrees corresponding to different risk source data. After filtering the risk source data using the risk decision decoupling engine, if all risk source data is accepted and no risk source data is rejected, generate a recommendation: no risk is required in the business scenario. If there is rejected risk source data, generate a recommendation: the risk needs to be considered in the business scenario, and manual speculation is recommended. The generated risk report will be output and displayed to the staff in the business scenario.

[0017] A risk data management system based on data analysis, which includes a data acquisition module, a rule base construction module, a scenario sub-graph module, a scenario coupling degree calculation module, a risk decision module and a visualization display module; The data collection module is used to collect all risk data in different business scenarios from the enterprise's historical information; standardize the collected data; The rule base construction module is used to define risk transmission rules for each business scenario, and integrate the risk transmission rules of all business scenarios to build a risk transmission rule base; The scenario-based sub-graph module is used to extract business entities as nodes in the risk transmission rule library to construct a scenario-based sub-graph; The scenario coupling degree calculation module is used to calculate the risk trigger probability based on the path length from the risk source entity to the business terminal entity in the scenario sub-graph, extract the loss value of the affected node in the scenario sub-graph to calculate the scenario impact intensity, extract the risk buffer node to calculate the risk absorption capacity, and calculate the scenario coupling degree using the risk trigger probability, scenario impact intensity and risk absorption capacity; The risk decision module is used to build a risk decision decoupling engine, judge the calculated scenario coupling degree, and screen the risk source data collected in real-time business scenarios; The visualization display module is used to extract the risk source data transmission path in the scenario sub-graph based on the screened risk source data, construct the risk impact range, and generate a risk report output.

[0018] The scenario-based sub-graph module includes entity node units and relationship edge units; The entity node unit is used to extract business entities as nodes in the risk transmission rule library for each business scenario. Business entities include risk source entities, business terminal entities and risk buffer entities. The relationship edge unit is used to extract risk impact paths in the risk transmission rule library as relationship edges between corresponding business entities.

[0019] The scenario coupling degree calculation module includes risk trigger probability unit, scenario impact intensity unit and risk absorption capacity unit; The risk trigger probability unit is used to calculate the risk trigger probability according to the path length from the risk source entity to the business terminal entity in the scenario sub-graph; The scene impact strength unit is used to extract the loss value of the affected node in the scene sub-graph to calculate the scene impact strength; The risk absorption capacity unit is used to extract risk buffer nodes and calculate risk absorption capacity.

[0020] The risk decision module includes a decoupling engine building unit and a constraint unit; The decoupling engine construction unit is used to construct a risk decision decoupling engine to determine whether the risk source data is acceptable; The constraint unit is used to set constraint conditions. When the constraint conditions are met, the risk source data is directly rejected and the risk decision decoupling engine is not executed.

[0021] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention can timely discover potential high-risk hazards by collecting risk source data in real time, calculating scenario coupling degree and screening high-risk data. Combined with the impact scope and decision-making recommendations in the risk report, it enables enterprises to take measures in advance to avoid or reduce risk losses, thereby enhancing the timeliness and effectiveness of risk warning and response.

[0022] 2. The present invention calculates the scenario coupling degree of risk source data in business scenarios to obtain the impact of risk source data in corresponding business scenarios. This not only ensures in-depth analysis of risk data by users and early avoidance, but also avoids risk source data with low impact on real-time business scenarios, and avoids misjudgment due to the lack of correlation between risk source data and business. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a module diagram of a risk data management system based on data analysis in the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0025] Example: Figure 1 As shown, the present invention provides a technical solution. A risk data management method based on data analysis, the method comprising the following steps: S100. Collect all risk data under different business scenarios from the enterprise's historical information, the risk data including risk type, risk trigger conditions, and risk impact path; and define a risk transmission rule base using the risk data; The specific steps for defining the risk transmission rule base using risk data are as follows: S101. Collect all risk data under different business scenarios in the enterprise's historical information. The risk data includes risk type, risk trigger conditions and risk impact path, and define risk transmission rules as Rule s={RiskType,TriggerCondition,ImpactPath}, where Rule s Indicates the risk transmission rule for the s-th business scenario, RiskType indicates the risk type, TriggerCondition indicates the risk trigger condition, and ImpactPath indicates the risk impact path; S102. Define risk transmission rules for each business scenario, and integrate the risk transmission rules of all business scenarios to build a risk transmission rule library.

[0026] Business scenarios include supply chain cooperation and credit approval. Risk types include financial anomalies, personal disputes, and corporate disputes. For example, in a corporate dispute, a risk is triggered when the dispute amount exceeds 10% of the corporate cooperation. For example, the risk impact path is as follows: in a corporate dispute, corporate credit anomaly → account freeze → loan payment failure. An example of a risk transmission rule is: {Legal litigation, amount involved > total order amount × 0.1, funds frozen → supply cut-off}; By collecting risk types, triggering conditions, and impact paths, we clearly define the key components of risk, providing a precise analysis target and foundational framework for subsequent risk analysis. The establishment of a rule base provides data support for the construction of scenario-based sub-graphs, ensuring that subsequent risk analysis is well-defined and based on reliable historical data.

[0027] S200: Extract business entities as nodes from the risk transmission rule base. Business entities include risk source entities, business terminal entities, and risk buffer entities. Extract risk impact paths as relationship edges to construct a scenario-based subgraph. The specific steps to construct a scenario-based sub-graph are: S201. For each business scenario, extract the business subject as a node in the risk transmission rule library. The specific business entities include Entities={E risk , E biz , E buffer}, where E risk Represents the risk source entity, E biz Indicates the business terminal entity, E buffer represents a risk buffer entity; Examples of risk source entities include: low-credit enterprises, dishonest individuals, and corporate disputes; business terminal entities include: purchase orders, loan contracts, etc.; risk buffer entities include: alternative suppliers, collateral, etc. S202, extract the risk impact path in the risk transmission rule base as the relationship edge between the corresponding business entities Edges={(e i , e j , r ij)|e i →r ij →e j}, where e i Indicates the i-th business entity node, e j Represents the jth business entity node, r ij Represents the conduction relationship between the i-th business entity node and the j-th business entity node, e i →r ij →e j Indicates the direction of conduction; S203, connect all business entity nodes and relationship edges to build a scenario sub-graph G s = (V, E), where G s Represents the scenario-based subgraph of the s-th business scenario, V represents the relationship edge set, and E represents the business entity node set; set the hard constraint as follows: all business entity nodes in the constructed scenario-based subgraph are connected to the business terminal entity nodes.

[0028] Clearly distinguishing risk sources, endpoints, and buffer entities allows for rapid identification of risk starting points, endpoints, and intermediate buffers, focusing on key targets for subsequent risk probability calculations and impact analysis. Hard constraints on connectivity between all nodes and business endpoint entities ensure the integrity and relevance of the graph, preventing the omission of important risk transmission paths and ensuring comprehensive risk analysis.

[0029] S300: When risk detection is performed in a real-time business scenario and risk source data is collected, the risk trigger probability is calculated based on the path length from the risk source entity to the business terminal entity in the scenario-based sub-graph. The loss value of the affected node in the scenario-based sub-graph is extracted to calculate the scenario impact intensity. The risk buffer node is extracted to calculate the risk absorption capacity. The scenario coupling degree is calculated using the risk trigger probability, scenario impact intensity, and risk absorption capacity. The specific steps for calculating scenario coupling using risk trigger probability, scenario impact intensity, and risk absorption capacity are as follows: S301. When risk source data is collected in a real-time business scenario, the shortest transmission path length d from the risk source node to the business terminal node corresponding to the collected risk source data is extracted in the scenario-based sub-graph. The number of times the risk source data in the business scenario causes a risk to occur within a historical time T is collected. The basic risk probability of the risk source data is obtained by dividing the number of risk occurrences by the time T. The risk trigger probability is calculated using the transmission path. The formula is: ; In the formula, P represents the risk trigger probability, α represents the basic risk probability, β represents the attenuation coefficient, and d represents the length of the shortest transmission path. e represents a natural constant. d is the number of edges between the risk source node and the business terminal node in the scenario-based subgraph, which is dimensionless. By combining the basic risk probability, attenuation coefficient and the shortest transmission path length, the attenuation effect of risk during the transmission process is taken into account, making the calculation results more in line with the actual possibility of risk occurrence and more accurate than simple basic probability.

[0030] S302: Extract the number of business entity nodes affected by the risk source node and the loss value of the business entity corresponding to the node from the scenario-based sub-graph. Calculate the scenario impact intensity using the business entity nodes affected by the risk source node and the loss value. The formula is: ; In the formula, I represents the impact of the scenario, H represents the loss value of the business entity corresponding to the node, and the loss values ​​of the business entities corresponding to different nodes are standardized before participating in the calculation; N represents the number of business entity nodes affected by the risk source node, and w a Indicates the weight of the a-th business entity node affected by the risk source node. The weight is set by the business personnel; By weighted summing up the loss values ​​of the affected nodes, the importance (weight) and degree of loss of different nodes are comprehensively considered, objectively reflecting the impact of risks on business scenarios and providing quantitative indicators for risk assessment.

[0031] S303. Extract all risk buffer nodes in the scenario-based sub-graph. Business personnel conduct professional assessments on the entities and risk source data corresponding to the risk buffer nodes to obtain the buffering degree of the risk buffer nodes and the risk intensity of the risk source data. The risk absorption capacity of the business scenario is calculated using the standardized buffering degree and risk intensity. The formula is: ; In the formula, A represents the risk absorption capacity of the business scenario, Bu m represents the buffering degree of the mth risk buffer node, Ri represents the risk intensity of the risk source data, and M represents the number of risk buffer nodes in the scenario sub-graph; S304. Calculate the scenario coupling degree using the risk trigger probability, scenario impact intensity, and risk absorption capacity. Standardize the three data to eliminate the dimension. The formula is: ; In the formula, SAC represents the scenario coupling degree of risk source data collected in real time in the business scenario.

[0032] Integrating the risk trigger probability, impact intensity and absorption capacity to calculate the scenario coupling degree realizes a multi-dimensional comprehensive assessment of risks. A single indicator cannot fully reflect the risk situation, while the coupling degree can comprehensively reflect the overall risk situation.

[0033] S400: Build a risk decision decoupling engine to determine the calculated scenario coupling degree and filter the risk source data collected in real-time business scenarios. The specific steps for screening risk source data collected in real-time business scenarios are as follows: S401. Build a risk decision decoupling engine, specifically: ; In the formula, Decision represents risk decision, Accept represents acceptance of risk source data, Reject represents rejection of risk source data, and θ represents the scenario coupling threshold, which is manually configured by professionals. S402. Set constraints, specifically: collect the average loss value caused by the impact of historical risk source data on business entities, standardize it and use it as the basic risk score Iy. When I>Iy, directly reject the risk source data and do not execute the risk decision decoupling engine.

[0034] By setting scenario coupling thresholds and basic risk constraints, we can effectively filter out low-risk data, focus on high-risk data, reduce ineffective analysis, and improve risk decision-making efficiency. The construction of a risk decision decoupling engine standardizes and automates risk decision-making while allowing professionals to configure thresholds. This balances scientific and flexible decision-making and adapts to risk preferences in different business scenarios.

[0035] S500. Based on the screened risk source data, extract the risk source data transmission path in the scenario-based sub-graph, construct the risk impact range, and generate a risk report output.

[0036] The specific steps to generate the risk report output are: S501. Based on the filtered risk source data, extract the risk source data transmission path in the scenario-based sub-graph. Nodes that have a transmission path to the risk source node are identified as affected nodes. All affected nodes are marked with different colors, and the different-colored areas are identified as the risk impact range. S502: Divide the risk report into a three-tier structure, specifically raw risk, scenario analysis, and decision recommendations. Integrate all risk source data detected in the business scenario to generate raw risk. Display the calculated scenario coupling degrees corresponding to different risk source data. After filtering the risk source data using the risk decision decoupling engine, if all risk source data is accepted and no risk source data is rejected, generate a recommendation: no risk is required in the business scenario. If there is rejected risk source data, generate a recommendation: the risk needs to be considered in the business scenario, and manual speculation is recommended. The generated risk report will be output and displayed to the staff in the business scenario.

[0037] A risk data management system based on data analysis, which includes a data acquisition module, a rule base construction module, a scenario sub-graph module, a scenario coupling degree calculation module, a risk decision module and a visualization display module; The data collection module is used to collect all risk data in different business scenarios from the enterprise's historical information; standardize the collected data; The rule base construction module is used to define risk transmission rules for each business scenario, and integrate the risk transmission rules of all business scenarios to build a risk transmission rule base; The scenario-based sub-graph module is used to extract business entities as nodes in the risk transmission rule library to construct a scenario-based sub-graph; The scenario coupling degree calculation module is used to calculate the risk trigger probability based on the path length from the risk source entity to the business terminal entity in the scenario sub-graph, extract the loss value of the affected node in the scenario sub-graph to calculate the scenario impact intensity, extract the risk buffer node to calculate the risk absorption capacity, and calculate the scenario coupling degree using the risk trigger probability, scenario impact intensity and risk absorption capacity; The risk decision module is used to build a risk decision decoupling engine, judge the calculated scenario coupling degree, and screen the risk source data collected in real-time business scenarios; The visualization display module is used to extract the risk source data transmission path in the scenario sub-graph based on the screened risk source data, construct the risk impact range, and generate a risk report output.

[0038] The scenario-based sub-graph module includes entity node units and relationship edge units; The entity node unit is used to extract business entities as nodes in the risk transmission rule library for each business scenario. Business entities include risk source entities, business terminal entities and risk buffer entities. The relationship edge unit is used to extract risk impact paths in the risk transmission rule library as relationship edges between corresponding business entities.

[0039] The scenario coupling degree calculation module includes risk trigger probability unit, scenario impact intensity unit and risk absorption capacity unit; The risk trigger probability unit is used to calculate the risk trigger probability according to the path length from the risk source entity to the business terminal entity in the scenario sub-graph; The scene impact strength unit is used to extract the loss value of the affected node in the scene sub-graph to calculate the scene impact strength; The risk absorption capacity unit is used to extract risk buffer nodes and calculate risk absorption capacity.

[0040] The risk decision module includes a decoupling engine building unit and a constraint unit; The decoupling engine construction unit is used to construct a risk decision decoupling engine to determine whether the risk source data is acceptable; The constraint unit is used to set constraint conditions. When the constraint conditions are met, the risk source data is directly rejected and the risk decision decoupling engine is not executed.

[0041] Example: The business scenario is: electronic product procurement, assessing the impact of suppliers' "environmental penalties" on electronic product procurement; A scenario-based sub-graph is constructed based on the company's historical data and "environmental protection penalty" risk data; In this business scenario, the user detects the supplier's risk data and extracts the risk source data as "environmental protection penalties" and the business terminal node as "electronic product procurement"; Extract the path from the risk source data node to the business terminal node in the scenario sub-graph as 2, and calculate the risk trigger probability P=0.2×e -0.4×2 =0.09; The affected entity nodes are factory shutdown and order delay, and the corresponding loss values ​​are standardized to 0.5 and 0.6 respectively. The impact intensity of the scenario is calculated as I = 0.5 × 0.6 + 0.6 × 0.3 = 0.48. There is only one risk buffer node in the business scenario. The buffer degree of the risk buffer node and the risk intensity of the risk source data are 0.5 and 0.66 respectively. The calculated risk absorption capacity A=0.75; Calculate the scenario coupling degree of risk source data SAC = 0.09 + 0.48 + (1-0.75) = 0.82; The scenario coupling threshold is set to 0.9 to determine the acceptance of risk source data. It is suggested that "environmental protection penalties" will not have a significant impact on electronic product procurement and are acceptable.

[0042] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A risk data management method based on data analysis, characterized by: The method comprises the following steps: S100. Collect all risk data under different business scenarios from the enterprise's historical information, the risk data including risk type, risk trigger conditions, and risk impact path; and define a risk transmission rule base using the risk data; S200: Extract business entities as nodes from the risk transmission rule base. Business entities include risk source entities, business terminal entities, and risk buffer entities. Extract risk impact paths as relationship edges to construct a scenario-based subgraph. S300: When risk detection is performed in a real-time business scenario and risk source data is collected, the risk trigger probability is calculated based on the path length from the risk source entity to the business terminal entity in the scenario-based sub-graph. The loss value of the affected node in the scenario-based sub-graph is extracted to calculate the scenario impact intensity. The risk buffer node is extracted to calculate the risk absorption capacity. The scenario coupling degree is calculated using the risk trigger probability, scenario impact intensity, and risk absorption capacity. S400: Build a risk decision decoupling engine to determine the calculated scenario coupling degree and filter the risk source data collected in real-time business scenarios. S500. Based on the screened risk source data, extract the risk source data transmission path in the scenario-based sub-graph, construct the risk impact range, and generate a risk report output.

2. The risk data management method based on data analysis according to claim 1, characterized in that: The specific steps of defining the risk transmission rule base using risk data in S100 are: S101. Collect all risk data under different business scenarios in the enterprise's historical information. The risk data includes risk type, risk trigger conditions and risk impact path, and define risk transmission rules as Rule s ={RiskType,TriggerCondition,ImpactPath}, where Rule s Indicates the risk transmission rule for the s-th business scenario, RiskType indicates the risk type, TriggerCondition indicates the risk trigger condition, and ImpactPath indicates the risk impact path; S102. Define risk transmission rules for each business scenario, and integrate the risk transmission rules of all business scenarios to build a risk transmission rule library.

3. The risk data management method based on data analysis according to claim 2, characterized in that: The specific steps of constructing the scenario-based sub-graph in S200 are: S201. For each business scenario, extract the business subject as a node in the risk transmission rule library. The specific business entities include Entities={E risk , E biz , E buffer }, where E risk Represents the risk source entity, E biz Indicates the business terminal entity, E buffer represents a risk buffer entity; S202, extract the risk impact path in the risk transmission rule base as the relationship edge between the corresponding business entities Edges={(e i , e j , r ij )|e i →r ij →e j }, where e i Indicates the i-th business entity node, e j Represents the jth business entity node, r ij Represents the conduction relationship between the i-th business entity node and the j-th business entity node, e i →r ij →e j Indicates the direction of conduction; S203, connect all business entity nodes and relationship edges to build a scenario sub-graph G s = (V, E), where G s Represents the scenario-based subgraph of the s-th business scenario, V represents the relationship edge set, and E represents the business entity node set; set the hard constraint as follows: all business entity nodes in the constructed scenario-based subgraph are connected to the business terminal entity nodes.

4. The risk data management method based on data analysis according to claim 3, characterized in that: The specific steps of calculating the scenario coupling degree using the risk trigger probability, scenario impact intensity, and risk absorption capacity in S300 are as follows: S301. When risk source data is collected in a real-time business scenario, the shortest transmission path length d from the risk source node to the business terminal node corresponding to the collected risk source data is extracted in the scenario-based sub-graph. The number of times the risk source data in the business scenario causes a risk to occur within a historical time T is collected. The basic risk probability of the risk source data is obtained by dividing the number of risk occurrences by the time T. The risk trigger probability is calculated using the transmission path. The formula is: ； In the formula, P represents the risk trigger probability, α represents the basic risk probability, β represents the attenuation coefficient, d represents the shortest conduction path length; e represents the natural constant; d is the number of edges between the risk source node and the business terminal node in the scenario-based sub-graph, dimensionless; S302: Extract the number of business entity nodes affected by the risk source node and the loss value of the business entity corresponding to the node from the scenario-based sub-graph. Calculate the scenario impact intensity using the business entity nodes affected by the risk source node and the loss value. The formula is: ； In the formula, I represents the impact of the scenario, H represents the loss value of the business entity corresponding to the node, and the loss values ​​of the business entities corresponding to different nodes are standardized before participating in the calculation; N represents the number of business entity nodes affected by the risk source node, and w a Indicates the weight of the a-th business entity node affected by the risk source node. The weight is set by the business personnel; S303. Extract all risk buffer nodes in the scenario-based sub-graph. Business personnel conduct professional assessments on the entities and risk source data corresponding to the risk buffer nodes to obtain the buffering degree of the risk buffer nodes and the risk intensity of the risk source data. The risk absorption capacity of the business scenario is calculated using the standardized buffering degree and risk intensity. The formula is: ； In the formula, A represents the risk absorption capacity of the business scenario, Bu m represents the buffering degree of the mth risk buffer node, Ri represents the risk intensity of the risk source data, and M represents the number of risk buffer nodes in the scenario sub-graph; S304. Calculate the scenario coupling degree using the risk trigger probability, scenario impact intensity, and risk absorption capacity. Standardize the three data to eliminate the dimension. The formula is: ； In the formula, SAC represents the scenario coupling degree of risk source data collected in real time in the business scenario.

5. The risk data management method based on data analysis according to claim 4, characterized in that: The specific steps of screening the risk source data collected in the real-time business scenario in S400 are: S401. Build a risk decision decoupling engine, specifically: ； In the formula, Decision represents risk decision, Accept represents acceptance of risk source data, Reject represents rejection of risk source data, and θ represents the scenario coupling threshold, which is manually configured by professionals. S402. Set constraints, specifically: collect the average loss value caused by the impact of historical risk source data on business entities, standardize it and use it as the basic risk score Iy. When I>Iy, directly reject the risk source data and do not execute the risk decision decoupling engine.

6. The risk data management method based on data analysis according to claim 5, characterized in that: The specific steps of generating the risk report output in S500 are: S501. Based on the filtered risk source data, extract the risk source data transmission path in the scenario-based sub-graph. Nodes that have a transmission path to the risk source node are identified as affected nodes. All affected nodes are marked with different colors, and the different-colored areas are identified as the risk impact range. S502. Divide the risk report into a three-tier structure, specifically original risk, scenario analysis, and decision-making recommendations; Integrate all risk source data detected in business scenarios to generate original risks; Display the calculated scenario coupling degree corresponding to different risk source data; After the risk source data is filtered by the risk decision decoupling engine, if all risk source data are accepted and there is no rejected risk source data, a suggestion is generated: the risk does not need to be considered in the business scenario; if there is rejected risk source data, a suggestion is generated: the risk needs to be considered in the business scenario, and manual speculation is recommended; The generated risk report will be output and displayed to the staff in the business scenario.

7. A risk data management system based on data analysis, characterized by: The risk data management system includes a data acquisition module, a rule base construction module, a scenario sub-graph module, a scenario coupling degree calculation module, a risk decision module and a visualization display module; The data collection module is used to collect all risk data in different business scenarios from the enterprise's historical information; standardize the collected data; The rule base construction module is used to define risk transmission rules for each business scenario, and integrate the risk transmission rules of all business scenarios to build a risk transmission rule base; The scenario-based sub-graph module is used to extract business entities as nodes in the risk transmission rule library to construct a scenario-based sub-graph; The scenario coupling degree calculation module is used to calculate the risk trigger probability based on the path length from the risk source entity to the business terminal entity in the scenario sub-graph, extract the loss value of the affected node in the scenario sub-graph to calculate the scenario impact intensity, extract the risk buffer node to calculate the risk absorption capacity, and calculate the scenario coupling degree using the risk trigger probability, scenario impact intensity and risk absorption capacity; The risk decision module is used to build a risk decision decoupling engine, judge the calculated scenario coupling degree, and screen the risk source data collected in real-time business scenarios; The visualization display module is used to extract the risk source data transmission path in the scenario sub-graph based on the screened risk source data, construct the risk impact range, and generate a risk report output.

8. The risk data management system based on data analysis according to claim 7, characterized in that: The scenario sub-graph module includes entity node units and relationship edge units; The entity node unit is used to extract business entities as nodes in the risk transmission rule library for each business scenario. Business entities include risk source entities, business terminal entities and risk buffer entities. The relationship edge unit is used to extract risk impact paths in the risk transmission rule library as relationship edges between corresponding business entities.

9. The risk data management system based on data analysis according to claim 7, characterized in that: The scenario coupling degree calculation module includes a risk trigger probability unit, a scenario impact intensity unit and a risk absorption capacity unit; The risk trigger probability unit is used to calculate the risk trigger probability according to the path length from the risk source entity to the business terminal entity in the scenario sub-graph; The scene impact strength unit is used to extract the loss value of the affected node in the scene sub-graph to calculate the scene impact strength; The risk absorption capacity unit is used to extract risk buffer nodes and calculate risk absorption capacity.

10. The risk data management system based on data analysis according to claim 7, characterized in that: The risk decision module includes a decoupling engine construction unit and a constraint unit; The decoupling engine construction unit is used to construct a risk decision decoupling engine to determine whether the risk source data is acceptable; The constraint unit is used to set constraint conditions. When the constraint conditions are met, the risk source data is directly rejected and the risk decision decoupling engine is not executed.

Images