A method and device for assessing enterprise risk, electronic equipment, and storage medium
By obtaining corporate characteristic information, identifying its own risks, calculating risk infection values and integrating evaluation, the applicability and accuracy of credit risk infection assessment in the existing technology has been solved, and an effective assessment of bank credit risk infection and non-listed enterprises has been achieved.
Patent Information
- Application Number
- CN202210615371.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-01
- Publication Date
- 2025-05-02
- Estimated Expiration
- 2042-06-01
AI Technical Summary
The prior art has poor applicability when evaluating bank credit risk contagion, low accuracy of results, and cannot effectively evaluate the risks of non-listed companies.
By obtaining the enterprise characteristic information of the target enterprise, input the enterprise's own risk identification model to obtain its own risk values, determine the continuous and discrete risk infection coefficients based on the equity and guarantee relationship in the enterprise map, calculate the continuous and discrete risk infection values of the target enterprise, and input these values into the fusion model to obtain the overall risk value.
It has achieved an accurate assessment of corporate credit risk contagion, expanded the scope of assessment application, effectively evaluated the risks of non-listed companies, and improved the accuracy of results.
Smart Images

Figure CN114819742B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of risk assessment, and in particular to an enterprise risk assessment method and device, electronic equipment, and storage medium. Background Art
[0002] Risk contagion is the core concept of systemic risk. When a company is severely impacted and risks arise, the risk factors will quickly spread to other related companies through the interconnected relationships between companies, eventually causing a serious impact on the industry and even the entire economy.
[0003] Therefore, when conducting risk assessment on enterprises, the assessment will be based on risk contagion. The current method of assessing risk contagion is mainly to analyze and model based on market yield data, and then conduct risk assessment through the model.
[0004] However, banks mainly undertake the function of credit intermediary, that is, they are mainly faced with the contagion of credit risk, so the applicability of existing technical analysis is poor and the accuracy of the results is low. In addition, many of the bank's customers are non-listed companies, and it is impossible to analyze and model them through yield data. It can be seen that the scope of application of existing technologies is relatively small, and the accuracy of the results cannot be effectively guaranteed in other scenarios. Summary of the invention
[0005] Based on the above-mentioned deficiencies of the prior art, the present application provides a method and device for assessing enterprise risk, an electronic device, and a storage medium to solve the problem that the prior art has a small scope of application and cannot guarantee the accuracy of the results.
[0006] In order to achieve the above objectives, this application provides the following technical solutions:
[0007] The first aspect of the present application provides a method for assessing enterprise risk, comprising:
[0008] Obtaining the enterprise characteristic information of the target enterprise;
[0009] Inputting the enterprise characteristic information of the target enterprise into a pre-built enterprise risk identification model to obtain the target enterprise's own risk value;
[0010] Determine the continuous risk contagion coefficient based on the equity relationship in the enterprise graph, and calculate the target continuous risk contagion value from the risk source enterprise to the target enterprise, and determine the discrete risk contagion coefficient based on the guarantee relationship in the enterprise graph, and calculate the target discrete risk contagion value from the risk source enterprise to the target enterprise; wherein each node in the enterprise graph corresponds to an enterprise, and the directed edge between each two nodes represents the equity relationship or guarantee relationship between the corresponding two enterprises;
[0011] The target enterprise's own risk value, the target continuous risk contagion value and the target discrete risk contagion value are input into a pre-built fusion model for fusion to obtain the overall risk value of the target enterprise.
[0012] Optionally, in the enterprise risk assessment method provided above, determining the continuous risk contagion coefficient based on the equity relationship in the enterprise map, and calculating the target continuous risk contagion value from the risk source enterprise to the target enterprise, includes:
[0013] Determine a first target subgraph from the enterprise graph; wherein the first target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to various enterprises directly or indirectly affected by the risk source enterprise through equity relations, and includes nodes corresponding to the target enterprise;
[0014] Determine a first direct impact matrix; wherein the first direct impact matrix is composed of continuous risk contagion coefficients between the risk source enterprise and each enterprise in the first target subgraph; the continuous risk contagion coefficient between two enterprises that are not directly connected through equity relations is zero;
[0015] Determine a second direct impact matrix; wherein the second direct impact matrix is composed of the continuous risk contagion coefficients between every two enterprises in the first target subgraph except the risk source enterprise;
[0016] Using the first direct impact matrix and the second direct impact matrix, calculate the continuous risk contagion value corresponding to each enterprise in the target subgraph;
[0017] The continuous risk contagion values corresponding to the paths from the risk source enterprise to the target enterprise in the first target subgraph are summed to obtain the target continuous risk contagion value.
[0018] Optionally, in the enterprise risk assessment method provided above, determining the discrete risk contagion coefficient based on the guarantee relationship in the enterprise map, and calculating the target discrete risk contagion value from the risk source enterprise to the target enterprise, includes:
[0019] Determine a second target subgraph from the enterprise graph; wherein the second target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to various enterprises directly or indirectly affected by the risk source enterprise through guarantee relationships, and includes nodes corresponding to the target enterprise;
[0020] Determine a third direct impact matrix; wherein the third direct impact matrix is composed of discrete risk contagion coefficients between the risk source enterprise and each enterprise in the second target subgraph; the discrete risk contagion coefficient between two enterprises that are not directly connected through a guarantee relationship is zero;
[0021] Determine a fourth direct impact matrix; wherein the fourth direct impact matrix is composed of the discrete risk contagion coefficients between every two enterprises in the second target subgraph except the risk source enterprise;
[0022] Determine all target paths from the risk source enterprise to the target enterprise in the target subgraph; wherein the target path refers to a path without a closed loop;
[0023] Based on the third direct impact matrix and the fourth direct impact matrix, a discrete risk contagion value corresponding to each target path is calculated;
[0024] The discrete risk infection values corresponding to each of the target paths are summed to obtain a target discrete risk infection value.
[0025] Optionally, in the enterprise risk assessment method provided above, after inputting the target enterprise's own risk value, the target continuous risk contagion value and the target discrete risk contagion value into a pre-built fusion model for fusion to obtain the overall risk value of the target enterprise, the method further includes:
[0026] Determine whether the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise;
[0027] If it is determined that the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise, it is determined that the target enterprise has risk contagion.
[0028] A second aspect of the present application provides an enterprise risk assessment device, comprising:
[0029] An acquisition unit, used for acquiring enterprise characteristic information of a target enterprise;
[0030] An own risk determination unit, used to input the enterprise characteristic information of the target enterprise into a pre-built enterprise own risk identification model to obtain the own risk value of the target enterprise;
[0031] A continuous risk determination unit, used to determine a continuous risk contagion coefficient based on the equity relationship in the enterprise map, and calculate a target continuous risk contagion value from the risk source enterprise to the target enterprise;
[0032] A discrete risk determination unit, used to determine a discrete risk contagion coefficient based on the guarantee relationship in the enterprise graph, and calculate a target discrete risk contagion value from the risk source enterprise to the target enterprise; wherein each node in the enterprise graph corresponds to an enterprise, and a directed edge between each two nodes represents an equity relationship or a guarantee relationship between the corresponding two enterprises;
[0033] The fusion unit is used to input the target enterprise's own risk value, the target continuous risk contagion value and the target discrete risk contagion value into a pre-built fusion model for fusion to obtain the overall risk value of the target enterprise.
[0034] Optionally, in the enterprise risk assessment method provided above, the continuous risk determination unit includes:
[0035] A first subgraph determination unit is used to determine a first target subgraph from the enterprise graph; wherein the first target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to various enterprises directly or indirectly affected by the risk source enterprise through equity relations, and includes nodes corresponding to the target enterprise;
[0036] A first matrix determination unit is used to determine a first direct impact matrix; wherein the first direct impact matrix is composed of continuous risk contagion coefficients between the risk source enterprise and each enterprise in the first target subgraph; the continuous risk contagion coefficient between two enterprises that are not directly connected through equity relations is zero;
[0037] A second matrix determination unit is used to determine a second direct impact matrix; wherein the second direct impact matrix is composed of the continuous risk contagion coefficients between every two enterprises in the first target subgraph except the risk source enterprise;
[0038] A first calculation unit, configured to calculate a continuous risk contagion value corresponding to each enterprise in the target subgraph by using the first direct impact matrix and the second direct impact matrix;
[0039] The second calculation unit is used to sum up the continuous risk contagion values corresponding to the paths from the risk source enterprise to the target enterprise in the first target subgraph to obtain the target continuous risk contagion value.
[0040] Optionally, in the enterprise risk assessment method provided above, the discrete risk determination unit includes:
[0041] A second subgraph determination unit is used to determine a second target subgraph from the enterprise graph; wherein the second target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to various enterprises directly or indirectly affected by the risk source enterprise through guarantee relationships, and includes nodes corresponding to the target enterprise;
[0042] A third matrix determination unit is used to determine a third direct impact matrix; wherein the third direct impact matrix is composed of discrete risk contagion coefficients between the risk source enterprise and each enterprise in the second target subgraph; the discrete risk contagion coefficient between two enterprises that are not directly connected through a guarantee relationship is zero;
[0043] A fourth matrix determination unit, used to determine a fourth direct impact matrix; wherein the fourth direct impact matrix is composed of the discrete risk contagion coefficients between every two enterprises in the second target subgraph except the risk source enterprise;
[0044] A path determination unit, configured to determine all target paths from the risk source enterprise to the target enterprise in the target subgraph; wherein the target path refers to a path without a closed loop;
[0045] A third calculation unit, configured to calculate a discrete risk contagion value corresponding to each of the target paths based on the third direct impact matrix and the fourth direct impact matrix;
[0046] The fourth calculation unit is used to sum the discrete risk infection values corresponding to each of the target paths to obtain a target discrete risk infection value.
[0047] Optionally, in the enterprise risk assessment method provided above, it also includes:
[0048] A judgment unit, used to judge whether the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise;
[0049] The contagiousness determination unit is used to determine that the target enterprise has risk contagiousness when it is determined that the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise.
[0050] A third aspect of the present application provides an electronic device, including:
[0051] Memory and processor;
[0052] Wherein, the memory is used to store programs;
[0053] The processor is used to execute the program, and when the program is executed, it is specifically used to implement the enterprise risk assessment method as described in any one of the above.
[0054] A fourth aspect of the present application provides a computer storage medium for storing a computer program, which, when executed, is used to implement the enterprise risk assessment method as described in any one of the above.
[0055] The present application provides a method for assessing enterprise risk, which obtains the enterprise characteristic information of the target enterprise, and inputs the enterprise characteristic information of the target enterprise into a pre-built enterprise risk identification model to obtain the target enterprise's own risk value. Based on the equity relationship in the enterprise map, the continuous risk contagion coefficient is determined, and the target continuous risk contagion value from the risk source enterprise to the target enterprise is calculated, and the discrete risk contagion coefficient is determined based on the guarantee relationship in the enterprise map, and the target discrete risk contagion value from the risk source enterprise to the target enterprise is calculated. Among them, each node in the enterprise map corresponds to an enterprise, and the directed edge between each two nodes represents the equity relationship or guarantee relationship between the corresponding two enterprises. Finally, the target enterprise's own risk value, the target continuous risk contagion value, and the target discrete risk contagion value are input into the pre-built fusion model for fusion to obtain the overall risk value of the target enterprise. Thus, not only the enterprise's own risk is taken into account, but also the enterprise's credit risk contagion risk is accurately analyzed without relying on the yield data, and is no longer limited by whether the enterprise is listed or not, which effectively ensures the accuracy of the results. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art description are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without creative work.
[0057] Figure 1 A flowchart of a method for assessing enterprise risk provided in an embodiment of the present application;
[0058] Figure 2 A flowchart of a method for determining a target discrete risk infection value provided in an embodiment of the present application;
[0059] Figure 3 A flow chart of a method for determining a target continuous risk infection value provided in an embodiment of the present application;
[0060] Figure 4 A schematic diagram of the structure of an enterprise risk assessment device provided in an embodiment of the present application;
[0061] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0062] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0063] In this application, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.
[0064] The present application provides a method for assessing enterprise risk, such as Figure 1 As shown, the following steps are included:
[0065] S101. Obtain enterprise characteristic information of the target enterprise.
[0066] It should be noted that in the embodiments of the present application, for the risks existing in an enterprise, not only the risks brought to the enterprise by the risk transmission from other enterprises are taken into consideration, but also the risks existing in the enterprise itself are taken into consideration.
[0067] The risk of the enterprise itself can be obtained by analyzing the enterprise characteristic information of the enterprise, such as the debt repayment capacity, profitability information and operation capacity information. Therefore, in the embodiment of the present application, the enterprise characteristic information of the target enterprise is used to determine the target enterprise's own risk value in step S102 through the target enterprise's enterprise characteristic information.
[0068] S102: Input the target enterprise's enterprise characteristic information into a pre-built enterprise risk identification model to obtain the target enterprise's own risk value.
[0069] It should be noted that in the embodiment of the present application, the enterprise's own risk identification model has been trained in advance through multiple sample data, so the enterprise characteristic information of the target enterprise can be directly input into the pre-built enterprise's own risk identification model. After the enterprise characteristic information of the target enterprise is feature processed by the enterprise's own risk identification model, the target enterprise's own risk value is obtained based on the processed feature vector analysis.
[0070] S103. Determine the continuous risk contagion coefficient based on the equity relationship in the enterprise map, and calculate the target continuous risk contagion value from the risk source enterprise to the target enterprise; and determine the discrete risk contagion coefficient based on the guarantee relationship in the enterprise map, and calculate the target discrete risk contagion value from the risk source enterprise to the target enterprise.
[0071] It should be noted that in the transmission of credit risk, the risk is mainly transmitted between enterprises with equity relations and guarantee relations. Therefore, in the embodiment of the present application, an enterprise map is constructed in advance based on the equity relations and guarantee relationship information between each enterprise. Among them, each node in the enterprise map corresponds to an enterprise, and the directed edge between each two nodes represents the equity relationship or guarantee relationship between the corresponding two enterprises. In addition, each edge may specifically include a corresponding contagion coefficient, that is, the coefficient of risk transmission through the relationship corresponding to the directed edge, for example, the probability of guarantee risk transmission and the probability of equity risk transmission. Therefore, when an enterprise is at risk, by connecting a directed edge with the enterprise, it is possible to determine the enterprise it will affect, and through the contagion coefficient, it is possible to determine the magnitude of the impact on the pointed enterprise, that is, the magnitude of the risk transmitted.
[0072] Among them, since the risk of guarantee relationship will be transmitted only when the guaranteed enterprise defaults, it belongs to discrete risk transmission, while for equity relationship, when one enterprise fluctuates, it will have an impact on the enterprises it controls, so it is continuous risk transmission. Therefore, in the embodiment of this application, the risk transmission of equity relationship and the risk transmission of guarantee relationship are analyzed respectively.
[0073] Specifically, the continuous risk contagion coefficient is determined based on the equity relationship in the enterprise map, and then based on the continuous risk contagion coefficient, the risk transmission analysis can be carried out from the risk source enterprise through the equity relationship until the target enterprise, so as to obtain the magnitude of the impact on the target enterprise when it is transmitted to the target enterprise, that is, to obtain the target continuous risk contagion value. Similarly, the discrete risk contagion coefficient is determined based on the guarantee relationship in the enterprise map, and then based on the discrete risk contagion coefficient, the risk transmission analysis can be carried out from the risk source enterprise through the guarantee relationship until the target enterprise, so as to obtain the magnitude of the impact on the target enterprise when it is transmitted to the target enterprise, that is, to obtain the target discrete risk contagion value.
[0074] Optionally, in an embodiment of the present application, a specific implementation method is provided for determining a continuous contagion coefficient based on the equity relationship in the enterprise map, and calculating a target continuous risk contagion value from the risk source enterprise to the target enterprise, such as Figure 2 As shown, including:
[0075] S201. Determine a first target subgraph from the enterprise graph.
[0076] The first target subgraph is composed of the nodes corresponding to the risk source enterprise and the nodes corresponding to each enterprise directly or indirectly affected by the risk source enterprise through equity relations, and includes the nodes corresponding to the target enterprise. If the first target subgraph including the target enterprise cannot be determined, it means that the risk caused by the risk source enterprise will not be transmitted to the target enterprise, so the target continuous risk contagion value of the target enterprise can be considered to be zero.
[0077] It should be noted that direct impact refers to the direct transmission of risks from one of two adjacent enterprises to the other enterprise, while indirect impact refers to the transmission of risks from one of two non-adjacent enterprises to the other enterprise through multiple other enterprises.
[0078] Since for two companies with equity relations, one company has a stronger risk-bearing capacity, so the risk it suffers may not affect the next-level company with equity relations. Therefore, in the embodiment of the present application, when determining the first target subgraph, only the companies with influence are moved to. Even if there is an equity relationship, when there is no influence, the company is not included in the first target subgraph. And because usually when analyzing risk transmission, not only one company is analyzed, that is, not only the target company is analyzed, so in the embodiment of the present application, the first target subgraph consisting of all companies with influence is directly determined, so that all companies can be processed.
[0079] S202: Determine a first direct impact matrix.
[0080] Among them, the first direct impact matrix is composed of the continuous risk contagion coefficients between the risk source enterprise and each enterprise in the first target subgraph, and the continuous risk contagion coefficient between two enterprises that are not directly connected by equity relations is zero. That is to say, in the first direct impact matrix, except for the enterprises that have direct equity relations with the risk source enterprise, the continuous risk contagion coefficients corresponding to other enterprises are all zero, that is, the continuous risk contagion coefficient of the first-layer enterprises directly related to the risk source enterprise is greater than zero, and the continuous risk contagion coefficients of other non-first-layer enterprises are all zero. Therefore, it is necessary to determine the connection risk contagion coefficients between enterprises, and then form them into the first direct impact matrix. Optionally, the continuous risk contagion coefficient of the risk source enterprise and the first-layer enterprises directly related to it can be obtained by processing the enterprise characteristic information and specific equity information between the two enterprises using the model.
[0081] S203: Determine a second direct impact matrix.
[0082] Among them, the second direct impact matrix is composed of the continuous risk contagion coefficients between every two enterprises in the first target subgraph except the risk source enterprise.
[0083] Similarly, if two enterprises are directly connected by a directed edge representing the equity relationship, the continuous risk contagion coefficient between the two enterprises is greater than zero. If they are not directly connected by a directed edge, the continuous risk contagion coefficient between the two enterprises is zero.
[0084] S204: Calculate the continuous risk contagion value corresponding to each enterprise in the target subgraph by using the first direct impact matrix and the second direct impact matrix.
[0085] Among them, the continuous risk contagion value corresponding to each enterprise in the target subgraph is calculated by using the first direct impact matrix and the second direct impact matrix, which can be specifically expressed as:
[0086] C=(1-M) -1 α T
[0087] Among them, C represents the matrix of continuous risk contagion values corresponding to each enterprise; α represents the first direct impact matrix; M represents the second direct impact matrix. Solving the above expression, we can get the continuous risk contagion value corresponding to each enterprise.
[0088] It can also be expressed as:
[0089]
[0090] S205 , summing up the continuous risk contagion values corresponding to the paths from the risk source enterprise to the target enterprise in the first target subgraph to obtain a target continuous risk contagion value.
[0091] It should be noted that, in addition to the direct impact of the risk source enterprise on the target enterprise, it will also indirectly affect the target enterprise through the risk contagion of other enterprises. For example, the risk source enterprise is A, the target enterprise is B, and A and B have a direct equity relationship, then A will have a direct impact on B, that is, a continuous risk contagion value can be calculated. At the same time, A indirectly holds B's shares through C, then when A affects C, C will also affect B, so through this path, A will also have an indirect impact on B, so this path can calculate another continuous risk contagion value. Obviously, the final impact on enterprise B should be the sum of the continuous risk contagion values of the two paths of direct impact and indirect impact. Therefore, in the embodiment of the present application, each path from the risk source enterprise to the target enterprise is determined in the first target subgraph, and then based on the continuous risk contagion value corresponding to each enterprise obtained in step S204, the continuous risk contagion value corresponding to each path is determined, and then these continuous risk contagion values are summed to obtain the total continuous risk contagion value of the target enterprise.
[0092] Optionally, in another embodiment of the present application, a specific implementation method is to determine the discrete risk contagion coefficient based on the guarantee relationship in the enterprise map, and calculate and determine the target discrete risk contagion value from the risk source enterprise to the target enterprise, such as Figure 3 As shown, including:
[0093] S301. Determine a second target subgraph from the enterprise graph.
[0094] The second target subgraph is composed of the nodes corresponding to the risk source enterprise and the nodes corresponding to each enterprise directly or indirectly affected by the risk source enterprise through the guarantee relationship, and includes the nodes corresponding to the target enterprise. Similarly, if the second target subgraph including the target enterprise cannot be determined, it means that the risk caused by the risk source enterprise will not be transmitted to the target enterprise, so the target discrete risk contagion value of the target enterprise can also be considered to be zero.
[0095] S302: Determine a third direct impact matrix.
[0096] The third direct impact matrix is composed of discrete risk contagion coefficients between the risk source enterprise and each enterprise in the second target subgraph. The discrete risk contagion coefficient between two enterprises that are not directly connected by a guarantee relationship is zero.
[0097] S303: Determine a fourth direct impact matrix.
[0098] Among them, the fourth direct impact matrix is composed of the discrete risk contagion coefficients between every two enterprises in the second target subgraph except the risk source enterprise.
[0099] It should be noted that the specific implementation of steps S301 to S303 is consistent with the principles of steps S201 to S203 and is only based on different connection relationships. Therefore, the specific implementation of steps S301 to S303 can refer to the specific implementation of steps S201 to S203 accordingly and will not be repeated here.
[0100] S304: Determine all target paths from the risk source enterprise to the target enterprise in the target subgraph.
[0101] Among them, the target path refers to the path without a closed loop. Since risk contagion will only occur when the guaranteed enterprise defaults and have an impact on the guaranteeing enterprise, and the change in the risk status of the guaranteeing enterprise will not affect the guaranteed person in turn, that is, there is no positive feedback in the discrete risk contagion relationship, so the determined path from the risk source enterprise to the target enterprise does not include a path with a closed loop.
[0102] S305. Based on the third direct impact matrix and the fourth direct impact matrix, a discrete risk contagion value corresponding to each target path is calculated.
[0103] Specifically, for each target path, the discrete risk contagion coefficient between every two adjacent enterprises on the target path is found in the third direct impact matrix and the fourth direct impact matrix, and then these discrete risk contagion coefficients are multiplied in sequence to obtain the discrete risk contagion value corresponding to the target path.
[0104] S306: summing up the discrete risk contagion values corresponding to each target path to obtain a target discrete risk contagion value.
[0105] S104: Input the target enterprise's own risk value, the target continuous risk contagion value, and the target discrete risk contagion value into a pre-built fusion model for fusion to obtain the target enterprise's overall risk value.
[0106] Since the impacts of these three types of risks on the enterprise may be different in different scenarios, a fusion model can be built in advance according to the scenario requirements, so that these three types of risks can be integrated through the fusion model to obtain an overall risk value that meets the requirements.
[0107] It should be noted that if the risk contagion analysis is performed sequentially, after executing step S104, it may further include: determining whether the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise.
[0108] If the overall risk value of the target enterprise is judged to be greater than the corresponding risk threshold of the target enterprise, the target enterprise is determined to have risk contagion, which means that the risk of the target enterprise will be transmitted to the next enterprise. This makes it easier to determine whether to consider the impact of the target enterprise when conducting risk assessment on the next enterprise.
[0109] The embodiment of the present application provides a method for assessing enterprise risk, obtains the enterprise characteristic information of the target enterprise, and inputs the enterprise characteristic information of the target enterprise into a pre-built enterprise risk identification model to obtain the target enterprise's own risk value. Based on the equity relationship in the enterprise map, the continuous risk contagion coefficient is determined, and the target continuous risk contagion value from the risk source enterprise to the target enterprise is calculated. The discrete risk contagion coefficient is determined based on the guarantee relationship in the enterprise map, and the target discrete risk contagion value from the risk source enterprise to the target enterprise is calculated. Among them, each node in the enterprise map corresponds to an enterprise, and the directed edge between each two nodes represents the equity relationship or guarantee relationship between the corresponding two enterprises. Finally, the target enterprise's own risk value, the target continuous risk contagion value, and the target discrete risk contagion value are input into the pre-built fusion model for fusion to obtain the overall risk value of the target enterprise. Thus, not only the enterprise's own risk is taken into account, but also the enterprise's risk in credit risk contagion is accurately analyzed without relying on the yield data, and is no longer limited to listed companies, and the accuracy of the results is effectively guaranteed.
[0110] Another embodiment of the present application provides an enterprise risk assessment device, such as Figure 4 As shown, including:
[0111] The acquisition unit 401 is used to acquire enterprise characteristic information of a target enterprise.
[0112] The own risk determination unit 402 is used to input the enterprise characteristic information of the target enterprise into a pre-built enterprise own risk identification model to obtain the own risk value of the target enterprise.
[0113] The continuous risk determination unit 403 is used to determine the target continuous risk contagion value from the risk source enterprise to the target enterprise based on the contagion coefficient transmitted through the equity relationship in the enterprise map.
[0114] The discrete risk determination unit 404 is used to determine the target discrete risk contagion value from the risk source enterprise to the target enterprise based on the contagion coefficient transmitted through the guarantee relationship in the enterprise map.
[0115] Among them, each node in the enterprise graph corresponds to an enterprise, and the directed edge between every two nodes represents the equity relationship or guarantee relationship between the corresponding two enterprises.
[0116] The fusion unit 405 is used to input the target enterprise's own risk value, the target continuous risk contagion value and the target discrete risk contagion value into a pre-built fusion model for fusion to obtain the overall risk value of the target enterprise.
[0117] Optionally, in the enterprise risk assessment method provided in another embodiment of the present application, the continuous risk determination unit includes:
[0118] The first subgraph determining unit is used to determine a first target subgraph from the enterprise graph.
[0119] The first target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to each enterprise directly or indirectly affected by the risk source enterprise through equity relations, and includes nodes corresponding to the target enterprise.
[0120] The first matrix determining unit is used to determine a first direct impact matrix.
[0121] The first direct impact matrix consists of the continuous risk contagion coefficients between the risk source enterprise and each enterprise in the first target subgraph. The continuous risk contagion coefficient between two enterprises that are not directly connected by equity relations is zero.
[0122] The second matrix determination unit is used to determine a second direct impact matrix, wherein the second direct impact matrix is composed of continuous risk contagion coefficients between every two enterprises except the risk source enterprise in the first target subgraph.
[0123] The first calculation unit is used to calculate the continuous risk contagion value corresponding to each enterprise in the target subgraph by using the first direct impact matrix and the second direct impact matrix.
[0124] The second calculation unit is used to sum up the continuous risk contagion values corresponding to the paths from the risk source enterprise to the target enterprise in the first target subgraph to obtain a target continuous risk contagion value.
[0125] Optionally, in the enterprise risk assessment method provided in another embodiment of the present application, the discrete risk determination unit includes:
[0126] The second subgraph determination unit is used to determine a second target subgraph from the enterprise graph, wherein the second target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to each enterprise directly or indirectly affected by the risk source enterprise through guarantee relationships, and includes nodes corresponding to the target enterprise.
[0127] The third matrix determination unit is used to determine a third direct impact matrix.
[0128] The third direct impact matrix is composed of the discrete risk contagion coefficients between the risk source enterprise and each enterprise in the second target subgraph. The discrete risk contagion coefficient between two enterprises that are not directly connected by a guarantee relationship is zero.
[0129] The fourth matrix determination unit is used to determine a fourth direct impact matrix.
[0130] Among them, the fourth direct impact matrix is composed of the discrete risk contagion coefficients between every two enterprises in the second target subgraph except the risk source enterprise.
[0131] The path determination unit is used to determine all target paths from the risk source enterprise to the target enterprise in the target subgraph.
[0132] The target path refers to a path without a closed loop.
[0133] The third calculation unit is used to calculate the discrete risk contagion value corresponding to each target path based on the third direct impact matrix and the fourth direct impact matrix.
[0134] The fourth calculation unit is used to sum the discrete risk contagion values corresponding to each target path to obtain a target discrete risk contagion value.
[0135] Optionally, in another embodiment of the present application, the enterprise risk assessment method further includes:
[0136] The judgment unit is used to judge whether the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise.
[0137] The contagiousness determination unit is used to determine that the target enterprise has risk contagiousness when it is determined that the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise.
[0138] It should be noted that the specific working process of each unit provided in the above embodiments of the present application can refer to the corresponding steps in the above method embodiments, which will not be repeated here.
[0139] Another aspect of the present application provides an electronic device, such as Figure 5 As shown, including:
[0140] Memory 501 and processor 502 .
[0141] Wherein, the memory 501 is used to store programs;
[0142] The processor 502 is used to execute the program stored in the memory 501. When the program is executed, it is specifically used to implement the enterprise risk assessment method provided by any one of the above embodiments.
[0143] Another embodiment of the present application provides a computer storage medium for storing a computer program. When the computer program is executed, it is used to implement the enterprise risk assessment method provided in any of the above embodiments.
[0144] Computer storage media includes permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.
[0145] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0146] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for assessing enterprise risk, characterized in that: include: Obtaining the enterprise characteristic information of the target enterprise; Inputting the enterprise characteristic information of the target enterprise into a pre-built enterprise risk identification model to obtain the target enterprise's own risk value; Determine the continuous risk contagion coefficient based on the equity relationship in the enterprise graph, and calculate the target continuous risk contagion value from the risk source enterprise to the target enterprise, and determine the discrete risk contagion coefficient based on the guarantee relationship in the enterprise graph, and calculate the target discrete risk contagion value from the risk source enterprise to the target enterprise; wherein each node in the enterprise graph corresponds to an enterprise, and the directed edge between each two nodes represents the equity relationship or guarantee relationship between the corresponding two enterprises; The target enterprise's own risk value, the target continuous risk contagion value and the target discrete risk contagion value are input into a pre-built fusion model for fusion to obtain the overall risk value of the target enterprise.
2. The method according to claim 1, characterized in that The step of determining the continuous risk contagion coefficient based on the equity relationship in the enterprise map and calculating the target continuous risk contagion value from the risk source enterprise to the target enterprise includes: Determine a first target subgraph from the enterprise graph; wherein the first target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to various enterprises directly or indirectly affected by the risk source enterprise through equity relations, and includes nodes corresponding to the target enterprise; Determine a first direct impact matrix; wherein the first direct impact matrix is composed of the continuous risk contagion coefficients between the risk source enterprise and each enterprise in the first target subgraph; the continuous risk contagion coefficient between two enterprises that are not directly connected through equity relations is zero; Determine a second direct impact matrix; wherein the second direct impact matrix is composed of the continuous risk contagion coefficients between every two enterprises in the first target subgraph except the risk source enterprise; Using the first direct impact matrix and the second direct impact matrix, calculate the continuous risk contagion value corresponding to each enterprise in the target subgraph; The continuous risk contagion values corresponding to the paths from the risk source enterprise to the target enterprise in the first target subgraph are summed to obtain the target continuous risk contagion value.
3. The method according to claim 1, characterized in that The step of determining the discrete risk contagion coefficient based on the guarantee relationship in the enterprise map and calculating the target discrete risk contagion value from the risk source enterprise to the target enterprise includes: Determine a second target subgraph from the enterprise graph; wherein the second target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to various enterprises directly or indirectly affected by the risk source enterprise through guarantee relationships, and includes nodes corresponding to the target enterprise; Determine a third direct impact matrix; wherein the third direct impact matrix is composed of discrete risk contagion coefficients between the risk source enterprise and each enterprise in the second target subgraph; the discrete risk contagion coefficient between two enterprises that are not directly connected through a guarantee relationship is zero; Determine a fourth direct impact matrix; wherein the fourth direct impact matrix is composed of the discrete risk contagion coefficients between every two enterprises in the second target subgraph except the risk source enterprise; Determine all target paths from the risk source enterprise to the target enterprise in the target subgraph; wherein the target path refers to a path without a closed loop; Based on the third direct impact matrix and the fourth direct impact matrix, a discrete risk contagion value corresponding to each target path is calculated; The discrete risk infection values corresponding to each of the target paths are summed to obtain a target discrete risk infection value.
4. The method according to claim 1, characterized in that: After inputting the target enterprise's own risk value, the target continuous risk contagion value, and the target discrete risk contagion value into a pre-built fusion model for fusion to obtain the overall risk value of the target enterprise, the method further includes: Determine whether the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise; If it is determined that the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise, it is determined that the target enterprise has risk contagion.
5. An enterprise risk assessment device, characterized in that: include: An acquisition unit, used for acquiring enterprise characteristic information of a target enterprise; An own risk determination unit, used to input the enterprise characteristic information of the target enterprise into a pre-built enterprise own risk identification model to obtain the own risk value of the target enterprise; A continuous risk determination unit, used to determine a continuous risk contagion coefficient based on the equity relationship in the enterprise map, and calculate a target continuous risk contagion value from the risk source enterprise to the target enterprise; A discrete risk determination unit, used to determine a discrete risk contagion coefficient based on the guarantee relationship in the enterprise graph, and calculate a target discrete risk contagion value from the risk source enterprise to the target enterprise; wherein each node in the enterprise graph corresponds to an enterprise, and a directed edge between each two nodes represents an equity relationship or a guarantee relationship between the corresponding two enterprises; The fusion unit is used to input the target enterprise's own risk value, the target continuous risk contagion value and the target discrete risk contagion value into a pre-built fusion model for fusion to obtain the overall risk value of the target enterprise.
6. The device according to claim 5, characterized in that The continuous risk determination unit comprises: A first subgraph determination unit is used to determine a first target subgraph from the enterprise graph; wherein the first target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to various enterprises directly or indirectly affected by the risk source enterprise through equity relations, and includes nodes corresponding to the target enterprise; A first matrix determination unit is used to determine a first direct impact matrix; wherein the first direct impact matrix is composed of continuous risk contagion coefficients between the risk source enterprise and each enterprise in the first target subgraph; the continuous risk contagion coefficient between two enterprises that are not directly connected through equity relations is zero; A second matrix determination unit is used to determine a second direct impact matrix; wherein the second direct impact matrix is composed of the continuous risk contagion coefficients between every two enterprises in the first target subgraph except the risk source enterprise; A first calculation unit, configured to calculate a continuous risk contagion value corresponding to each enterprise in the target subgraph by using the first direct impact matrix and the second direct impact matrix; The second calculation unit is used to sum up the continuous risk contagion values corresponding to the paths from the risk source enterprise to the target enterprise in the first target subgraph to obtain the target continuous risk contagion value.
7. The device according to claim 5, characterized in that The discrete risk determination unit comprises: A second subgraph determination unit is used to determine a second target subgraph from the enterprise graph; wherein the second target subgraph is composed of nodes corresponding to the risk source enterprise and nodes corresponding to various enterprises directly or indirectly affected by the risk source enterprise through guarantee relationships, and includes nodes corresponding to the target enterprise; A third matrix determination unit is used to determine a third direct impact matrix; wherein the third direct impact matrix is composed of discrete risk contagion coefficients between the risk source enterprise and each enterprise in the second target subgraph; the discrete risk contagion coefficient between two enterprises that are not directly connected through a guarantee relationship is zero; A fourth matrix determination unit, used to determine a fourth direct impact matrix; wherein the fourth direct impact matrix is composed of the discrete risk contagion coefficients between every two enterprises in the second target subgraph except the risk source enterprise; A path determination unit, configured to determine all target paths from the risk source enterprise to the target enterprise in the target subgraph; wherein the target path refers to a path without a closed loop; A third calculation unit, configured to calculate a discrete risk contagion value corresponding to each of the target paths based on the third direct impact matrix and the fourth direct impact matrix; The fourth calculation unit is used to sum the discrete risk infection values corresponding to each of the target paths to obtain a target discrete risk infection value.
8. The device according to claim 5, characterized in that Also includes: A judgment unit, used to judge whether the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise; The contagiousness determination unit is used to determine that the target enterprise has risk contagiousness when it is determined that the overall risk value of the target enterprise is greater than the risk bearing threshold corresponding to the target enterprise.
9. An electronic device, characterized in that: include: Memory and processor; Wherein, the memory is used to store programs; The processor is used to execute the program, and when the program is executed, it is specifically used to implement the enterprise risk assessment method as described in any one of claims 1 to 4.
10. A computer storage medium, characterized in that: Used to store a computer program, which, when executed, is used to implement the enterprise risk assessment method as described in any one of claims 1 to 4.
Citation Information
Patent Citations
Risk value evaluation method and device
CN111402064A
Enterprise risk assessment method and device and electronic equipment
CN111784508A