Risk overflow simulation method and system based on equity knowledge graph

Abstract

The invention discloses a risk overflow simulation method based on an equity knowledge graph. Firstly, a user specifies a target enterprise of an equity network, the amount and proportion of assets or liabilities, affected by a risk enterprise, of equity associated enterprises are quantitatively calculated from a risk source enterprise based on the equity holding proportion, and if the proportion of the assets or liabilities affected by the enterprise triggers a risk threshold value, the enterprise also becomes the risk enterprise; and continuing to conduct the risk according to the method until all propagation paths conducted to the target enterprise from the risk enterprise are calculated, calculating the amount and proportion of assets and liabilities affected by the target enterprise, returning whether the target enterprise is the risk enterprise or not, and evaluating the amount and proportion of assets and liabilities affected by the risk enterprise. According to the method, the financial risk assessment result based on stock right path propagation can be obtained, so that enterprise risk assessment work can be better completed, and the comprehensiveness of enterprise financial risk assessment is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of risk simulation, and specifically relates to a risk spillover simulation method and system based on equity knowledge graph. Background Art

[0002] In the financial field, risk spillover refers to the spread of financial risks in a specific field or institution and its impact on other fields or institutions and even the entire financial system. Today, equity network transmission has become an important way for financial risk transmission. How to analyze and simulate equity financial risks from the balance sheet is crucial for enterprises and investors. With the increasing complexity of financial markets, the equity relations between enterprises are intricate, and the risk transmission channels have become more hidden and diversified. Therefore, how to effectively simulate and analyze the impact of other risky enterprises on target enterprises from the perspective of equity has become an important topic in the current financial research field.

[0003] Knowledge graphs have the ability to integrate and analyze information such as equity relations and related transactions between enterprises, thereby providing solid and strong support for financial risk analysis. At present, traditional financial analysis techniques mainly rely on simple quantitative analysis of financial statements, construction of indicator analysis, and construction of statistical models. These methods can indeed evaluate the financial status and risk level of enterprises within a certain range. However, when dealing with risk transmission analysis involving complex equity relationship networks, they expose obvious limitations. In addition, the balance sheet, as a commonly used report reflecting the financial status of enterprises, has not been fully utilized in the analysis of equity risks. Not only that, when faced with large-scale and dynamically changing financial data, traditional methods have room for improvement in processing efficiency and accuracy. The characteristics of knowledge graphs make them naturally suitable for displaying complex equity networks and processing large-scale dynamically changing equity data, thus providing new ideas and methods for solving difficult problems in financial risk analysis. Summary of the invention

[0004] In order to make full use of the existing technical foundation and overcome the shortcomings of existing methods, the present invention aims to provide a risk spillover simulation method and system based on equity knowledge graph. By constructing an equity knowledge graph and using the corporate balance sheet and equity data set, the impact of corporate risk propagation along the equity path on the target enterprise is accurately simulated, the risk exposure amount and proportion are calculated, and the risk transmission situation is judged according to the threshold, thereby providing a more comprehensive and accurate risk assessment basis for financial decision-making.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A risk spillover simulation method based on equity knowledge graph includes the following steps:

[0007] Step 1: Obtain enterprise business data, create an enterprise dataset FIRM, build an equity dataset S based on the enterprise dataset, and build an equity knowledge graph based on the enterprise dataset FIRM and the equity dataset S;

[0008] Step 2: Based on the enterprise data set FIRM, analyze the target enterprise based on factual risk or simulated risk, and select risk enterprise r and target enterprise i for analysis;

[0009] Step 3: Traverse the equity data set S and the enterprise data set FIRM; calculate the affected asset amount and the affected asset amount ratio of enterprise o of risk enterprise r on the liability side of risk enterprise r, and calculate the affected equity investment amount and the affected equity investment amount ratio of equity investment enterprise p of risk enterprise r on the asset side of risk enterprise r;

[0010] Step 4: Determine whether enterprise o and enterprise p are risky enterprises respectively, and repeat steps 2 to 4 until neither the asset-side nor the liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i;

[0011] Step 5: After traversing all equity paths to target enterprise i, calculate the final affected total asset funds and total liability funds of target enterprise i, as well as the proportion of total asset funds affected by target enterprise i and the proportion of total liability funds affected by target enterprise i.

[0012] Furthermore, the enterprise data set FIRM in step 1 is:

[0013] FIRM={firm1,firm2,...,firm m}

[0014] The equity data set S is:

[0015] S={stock1, stock2,…, stock m}

[0016] Among them, firm1, firm2, ..., firm m are the first, second, and mth companies in the FIRM dataset respectively; stock1, stock2, ..., stock m They are the shareholding data of the first, second and nth enterprises in the equity dataset S respectively.

[0017] Furthermore, the shareholding data stock of the jth enterprise in the equity dataset S j It is a two-dimensional tuple, which is a tuple consisting of one or more triples. The form of the triple is (firm j , proportion jk , firmk ), proportion jk represents the shareholding ratio of the j-th enterprise to the k-th enterprise, where firm j and firm k are the j-th and k-th enterprises in the FIRM dataset (1≤j≤m, 1≤k≤m, j≠k) and proportion jk ∈(0,1].

[0018] Furthermore, the step 3 comprises:

[0019] Traverse the S data set to obtain the equity data set Astock of venture enterprise r r , is a two-dimensional tuple, including triples (firm r , proportion rf , firm f ), where f∈[1,m];

[0020] Traverse the S data set to obtain the equity data set Dstock of venture enterprise r r , is a two-dimensional tuple consisting of a triple (firm l , proportion lr , firm r ), where l∈[1,m];

[0021] Traversing the FIRM enterprise dataset and Astock r Get the data set Afirm of companies held by venture firm r r , is a two-dimensional tuple, composed of firm o Composition, where O∈[1,m] and satisfies firm o For Astock r The third element of the middle triple;

[0022] Traversing FIRM enterprise dataset and Dstock r Get the enterprise data set Dfirm of the venture enterprise r r , is a two-dimensional tuple, composed of firm p Composition, where p∈[1,m] and satisfies firm p For Dstock r The first element of the triplet in .

[0023] Furthermore, the amount of assets affected by the risk enterprise r's liability side in step 3 is Arisk or and the proportion of the amount of assets affected Ashare or They are:

[0024] Arisk or =debt r ×proportion or

[0025]

[0026] Among them, debt r is the paid-in capital of venture enterprise r, proportion or is the proportion of equity investment of enterprise o to venture enterprise r in the equity investment received by venture enterprise r, asset o is the total assets of enterprise o.

[0027] Furthermore, the amount of equity investment Drisk of venture r's equity investment enterprise p on the asset side of venture r is affected. pr and the proportion of equity investment affected Dshare pr They are:

[0028] Drisk pr =investment p ×proportion rp

[0029]

[0030] Among them, investment p is the paid-in capital of enterprise p, proportion rp DEBT is the proportion of venture enterprise r’s equity investment in enterprise p to the equity investment received by enterprise p. p is the total liabilities of enterprise p.

[0031] Furthermore, the determination of whether enterprise o and enterprise p are venture enterprises includes:

[0032] If Ashare or >α ao , then enterprise o becomes a venture enterprise, otherwise it does not become a venture enterprise; if Dshare pr >α dp , then enterprise p becomes a venture enterprise, otherwise it does not become a venture enterprise; among them, α ao is the average ratio of investment loss amount to total assets amount when the enterprise triggers risk based on the industry in which enterprise o is located, α dp It is the average value of the proportion of the impairment of investment value caused by the decrease in other comprehensive income of the investee under the equity method to the total liabilities when the enterprise in the industry in which enterprise P is located triggers risk.

[0033] Furthermore, neither the asset-side nor the liability-side enterprise triggers risk or transmits it to the target enterprise i. The stop includes:

[0034] If enterprise o is a risky enterprise, the target enterprise will remain enterprise i, the risky enterprise in step 2 will be set as enterprise o, and steps 2 to 4 will be repeated until neither its asset-side nor its liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i; if enterprise p is a risky enterprise, the target enterprise will remain enterprise i, the risky enterprise in step 2 will be set as enterprise p, and steps 2 to 4 will be repeated until neither its asset-side nor its liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i.

[0035] Furthermore, the total amount of assets and funds affected by the target enterprise i in step 5 is Arisk i Total amount of debt funds affected i They are:

[0036]

[0037] The proportion of total assets and funds affected by target enterprise i Ashare i and the proportion of total liabilities of target enterprise i affected Dshare i They are:

[0038]

[0039] Among them, v and w are the set of enterprises that have risk spillovers on the target enterprise i’s asset side and the set of enterprises that have risk spillovers on the target enterprise i’s liability side, respectively. X and Y are the set of risk enterprises that affect the enterprise i’s funds from the asset side and the liability side, respectively. Arisk iv and Drisk iw are the total asset funds of enterprise i affected by venture enterprise v and the total liability funds of enterprise i affected by venture enterprise w, respectively;

[0040] If Ashare i >α ai or Dshare i >α di , then the result is: Enterprise i is a risk enterprise, and the affected funds and proportions on the asset side are Arisk i and Ashare i The funds and proportions affected on the liability side are Drisk i and Dshare I ; Among them, α ai is the average ratio of investment loss amount to total assets amount when the enterprise triggers risk based on the industry in which enterprise i is located, α diIt is the average value of the proportion of the investment value impairment amount caused by the decrease in other comprehensive income of the investee under the equity method to the total liability amount when the enterprise in the industry in which enterprise i is located triggers risk.

[0041] On the other hand, the present invention provides a risk spillover simulation system based on equity knowledge graph, comprising:

[0042] Data acquisition module: used to acquire enterprise business data, create an enterprise data set FIRM, construct an equity data set S based on the enterprise data set, and construct an equity knowledge graph based on the enterprise data set FIRM and the equity data set S;

[0043] Target enterprise and risk enterprise analysis module: It is used to analyze target enterprises based on the enterprise data set FIRM, based on factual risk or simulated risk, and select risk enterprise r and target enterprise i for analysis;

[0044] Calculation module: It is used to traverse the equity data set S and the enterprise data set FIRM; calculate the affected asset amount and the affected asset amount ratio of enterprise o of risk enterprise r on the liability side of risk enterprise r, and calculate the affected equity investment amount and the affected equity investment amount ratio of equity investment enterprise p of risk enterprise r on the asset side of risk enterprise r;

[0045] Risk enterprise judgment module: It is used to judge whether enterprise o and enterprise p become risk enterprises respectively, until neither the asset-side nor the liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i;

[0046] Total amount calculation module: It is used to calculate the final affected total asset funds and total liability funds of target enterprise i, as well as the proportion of the affected total asset funds of target enterprise i and the proportion of the affected total liability funds of target enterprise i after traversing all equity paths to target enterprise i;

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] First, unlike the existing technology based on creditor-debtor relationships and corporate financial indicators, the present invention uses knowledge graph technology and provides a simulation analysis method for financial risk spillover based on the perspective of equity networks, which is innovative in perspective; second, unlike the existing technology's strict requirements on data form, the present invention simulates and analyzes financial risks based on existing corporate balance sheets and corporate equity data, which is less costly in practice and more operational; third, unlike the existing technology that conducts quantitative analysis and risk assessment on a single company, the present invention can show the path of equity risk transmission, providing a reference for decision makers to cut off the channels for risk transmission and take preventive measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 is a flow chart of an embodiment of the present invention.

[0051] Figure 2 It is a program operation flow chart used in an embodiment of the present invention.

[0052] Figure 3 A schematic diagram of the equity structure relationship in the financial equity knowledge graph according to an embodiment of the present invention.

[0053] Figure 4 It is a result schematic diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present application, so the present application is not limited by the specific embodiments disclosed below.

[0055] Example 1

[0056] In this embodiment, a data set S consisting of 13 equity data and an enterprise data set FIRM consisting of 12 enterprise data are used, which already contain data on the enterprise's name, paid-in capital, long-term equity investment, total liabilities, total assets, asset-side risk triggering thresholds, and liability-side risk triggering thresholds. The user starts running the program.

[0057] like Figure 1 and Figure 2 As shown, the present invention comprises the following steps:

[0058] Step 1: Obtain enterprise business data, create an enterprise dataset FIRM, build an equity dataset S based on the enterprise dataset, and build an equity knowledge graph based on the enterprise dataset FIRM and the equity dataset S;

[0059] In this embodiment, the graph database neo4j is used to construct the equity knowledge graph. Each node in the graph is mapped to an element in the enterprise set FIRM, and each edge in the graph is mapped to an equity relationship in the equity set S;

[0060] Let the given equity data set S = {stock1, stock2, ..., stock m} and enterprise data set FIRM = {firm1, firm2, ..., firm m}, stock j is the shareholding data of the j-th enterprise in the S data set (1≤j≤n), which is a two-dimensional tuple, a tuple consisting of one or more triples, and the form of the triple is (firm j , proportion jk , firm k ), indicating the shareholding ratio of the j-th enterprise to the k-th enterprise jk , where firm j and firm k are the j-th and k-th enterprises in the FIRM dataset (1≤j≤n, 1≤k≤n, j≠k) and proportion jk ∈(0, 1], stock j is the set of all companies held by company j and their proportions; the data in the company set FIRM

[0061] firm j =(name j , debt j , investment j ,DEBT j ,asset j , α aj , α dj ) is the one-dimensional tuple of the j-th enterprise, where

[0062] name j , debt j , investment j ,DEBT j ,asset j , α aj , α dj They are the name, paid-in capital, long-term equity investment, total liabilities, total assets, the threshold for triggering risk on the asset side, and the threshold for triggering risk on the liability side of the j-th enterprise.

[0063] Step 2: Based on the enterprise data set FIRM, analyze the target enterprise based on factual risk or simulated risk, and select risk enterprise r and target enterprise i for analysis;

[0064] Step 3: Traverse the equity data set S and the enterprise data set FIRM; calculate the affected asset amount and the affected asset amount ratio of enterprise o of risk enterprise r on the liability side of risk enterprise r, and calculate the affected equity investment amount and the affected equity investment amount ratio of equity investment enterprise p of risk enterprise r on the asset side of risk enterprise r;

[0065] Traverse the S data set to obtain the equity data set Astock of venture enterprise r r , is a two-dimensional tuple, including triples (firm r , proportion rf , firm f ), where f∈[1,m]. At the same time, traverse the S data set to obtain the equity data set Dstock of venture enterprise r r , is a two-dimensional tuple consisting of a triple (firm l , proportion lr , firm r ), where l∈[1,m];

[0066] On this basis, we traverse the FIRM enterprise dataset and Astock r Get the data set Afirm of companies held by venture firm r r , is a two-dimensional tuple, composed of firm o Constitute, where o∈[1,m] and satisfies firm o For Astock r The third element of the triplet; at the same time, traverse the FIRM enterprise dataset and Dstock r Get the enterprise data set Dfirm of the venture enterprise r r , is a two-dimensional tuple, composed of firm p Composition, where p∈[1,m] and satisfies firm p For Dstock r The first element of the triplet in .

[0067] Based on Dfirm r and Dstock r , calculate the amount of assets affected by all enterprises holding risk enterprise r on the liability side of enterprise r, and the calculation formula is as follows:

[0068] Arisk or =debt r ×proportion or

[0069]

[0070] Of which debtr is the paid-in capital of venture enterprise r, proportion or is the proportion of equity investment of enterprise o to venture enterprise r in the equity investment received by venture enterprise r, Arisk or Ashare is the amount of impact on enterprise o from venture enterprise r. or is the proportion of enterprise o affected by risk enterprise r, asset o is the total assets of enterprise o. Based on Dfirm r and Dstock r , calculate the affected equity investment of venture enterprise r's equity investment enterprise p on the asset side of venture enterprise r, the calculation formula is as follows:

[0071] Drisk pr =investment p ×proportion rp

[0072]

[0073] Among them, investment p is the paid-in capital of enterprise p, proportion rp Drisk is the proportion of venture enterprise r’s equity investment in enterprise p to the equity investment received by enterprise p. pr is the total amount of liabilities of enterprise p affected by risk enterprise r, Dshare pr is the proportion of liabilities of enterprise p affected by risk enterprise r, DEBT p is the total liabilities of enterprise p.

[0074] Step 4: Determine whether enterprise o and enterprise p are risky enterprises respectively, and repeat steps 2 to 4 until neither the asset-side nor the liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i;

[0075] If Ashare or >α ao , then enterprise o becomes a venture enterprise, otherwise it does not become a venture enterprise; if Dshare pr >α dp , then enterprise p becomes a risk enterprise, otherwise it does not become a risk enterprise. If enterprise o is a risk enterprise, the target enterprise remains enterprise i, the risk enterprise in step 2 is set as enterprise o, and steps 2 to 4 are repeated until neither its asset-side nor its liability-side enterprises trigger risks or are transmitted to target enterprise i; if enterprise p is a risk enterprise, the target enterprise remains enterprise i, the risk enterprise in step 2 is set as enterprise p, and steps 2 to 4 are repeated until neither its asset-side nor its liability-side enterprises trigger risks or are transmitted to target enterprise i.

[0076] Step 5: After traversing all equity paths to target enterprise i, calculate the final affected total asset funds and total liability funds of target enterprise i, as well as the proportion of total asset funds affected by target enterprise i and the proportion of total liability funds affected by target enterprise i.

[0077]

[0078] Among them, Arisk i and Drisk i are the total amount of assets and liabilities of target enterprise i that are ultimately affected, respectively. v and w are the set of enterprises that have risk spillovers on the asset side of target enterprise i and the set of enterprises that have risk spillovers on the liability side of target enterprise i, respectively. X and Y are the sets of risk enterprises that affect the funds of enterprise i from the asset side and from the liability side of enterprise i, respectively. Arisk iv and Drisk iw are the total amount of assets of enterprise i affected by risk enterprise v and the total amount of liabilities of enterprise i affected by risk enterprise w. i and Dshare i are the final capital ratio of the affected assets and the capital ratio of the affected liabilities of enterprise i. i >α ai or Dshare i >α di , then the result is: Enterprise i is a risk enterprise, and the affected funds and proportions on the asset side are Arisk i and Ashare i The funds and proportions affected on the liability side are Drisk i and Dshare i .

[0079] In this embodiment, the data set S consisting of 13 equity data and the enterprise data set consisting of 12 enterprise data are as follows: Figure 3 As shown;

[0080] like Figure 4As shown, in this embodiment, the system shows the path from risk enterprise r to target enterprise i, and the system sets the risk threshold of target enterprise i's liability side to 30%. Among them, risk enterprise r causes enterprises A and C holding risk enterprise r to trigger risks, but enterprises A and C will not trigger the risk of enterprise O, so the financial risk will not be transmitted to target enterprise i through enterprise O. Risk enterprise r will trigger the risk of enterprise p, causing enterprises G and enterprise w to trigger risks, thereby affecting the liability funds of target enterprise i from the liability side. It can be calculated that when enterprises G and enterprise w trigger risks, the proportion of affected liability funds of target enterprise i is 30.64%, which is greater than its risk threshold, so target enterprise i triggers risks, and the system reports the results to the user.

[0081] Example 2

[0082] This embodiment provides a risk spillover simulation system based on equity knowledge graph, including:

[0083] Data acquisition module: used to acquire enterprise business data, create an enterprise data set FIRM, construct an equity data set S based on the enterprise data set, and construct an equity knowledge graph based on the enterprise data set FIRM and the equity data set S;

[0084] Target enterprise and risk enterprise analysis module: It is used to analyze target enterprises based on the enterprise data set FIRM, based on factual risk or simulated risk, and select risk enterprise r and target enterprise i for analysis;

[0085] Calculation module: It is used to traverse the equity data set S and the enterprise data set FIRM; calculate the affected asset amount and the affected asset amount ratio of enterprise o of risk enterprise r on the liability side of risk enterprise r, and calculate the affected equity investment amount and the affected equity investment amount ratio of equity investment enterprise p of risk enterprise r on the asset side of risk enterprise r;

[0086] Risk enterprise judgment module: It is used to judge whether enterprise o and enterprise p become risk enterprises respectively, until neither the asset-side nor the liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i;

[0087] Total amount calculation module: It is used to calculate the final affected total asset funds and total liability funds of target enterprise i, as well as the proportion of the affected total asset funds of target enterprise i and the proportion of the affected total liability funds of target enterprise i after traversing all equity paths to target enterprise i;

[0088] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed in the present application should be covered within the protection scope of the present application.

[0089] It should be understood that parts not elaborated in detail in this specification belong to the prior art.

[0090] It should be understood that the above description of the preferred embodiment is relatively detailed and cannot be regarded as limiting the scope of patent protection of the present invention. Under the enlightenment of the present invention, ordinary technicians in this field can also make substitutions or modifications without departing from the scope of protection of the claims of the present invention, which all fall within the scope of protection of the present invention. The scope of protection requested for the present invention shall be based on the attached claims.

Claims

1. A risk spillover simulation method based on equity knowledge graph, characterized in that: The steps include: Step 1: Obtain enterprise business data, create an enterprise dataset FIRM, build an equity dataset S based on the enterprise dataset, and build an equity knowledge graph based on the enterprise dataset FIRM and the equity dataset S; Step 2: Based on the enterprise data set FIRM, analyze the target enterprise based on factual risk or simulated risk, and select risk enterprise r and target enterprise i for analysis; Step 3: Traverse the equity dataset S and the enterprise dataset FIRM; Calculate the affected asset amount and the percentage of the affected asset amount of the company o holding the risk enterprise r on the liability side of the risk enterprise r, and calculate the affected equity investment amount and the percentage of the affected equity investment amount of the equity investment company p of the risk enterprise r on the asset side of the risk enterprise r; Step 4: Determine whether enterprise o and enterprise p are risky enterprises respectively, and repeat steps 2 to 4 until neither the asset-side nor the liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i; Step 5: After traversing all equity paths to target enterprise i, calculate the final affected total asset funds and total liability funds of target enterprise i, as well as the proportion of total asset funds affected by target enterprise i and the proportion of total liability funds affected by target enterprise i.

2. The risk spillover simulation method based on equity knowledge graph according to claim 1 is characterized in that: The enterprise data set FIRM in step 1 is: FIRM={firm1,firm2,…,firm m } The equity data set S is: S={stock1,stock2,…,stock m } Among them, firm1, firm2, …, firm m are the first, second and mth companies in the FIRM dataset respectively; stock1, stock2,…, stock m They are the shareholding data of the first, second and mth enterprises in the equity dataset S respectively.

3. The risk spillover simulation method based on equity knowledge graph according to claim 2 is characterized in that: The stock holding data of the jth enterprise in the equity data set S j It is a two-dimensional tuple, which is a tuple consisting of one or more triples. The form of the triple is (firm j ,proportion jk ,firm k ), proportion jk represents the shareholding ratio of the j-th enterprise to the k-th enterprise, where firm j and firm k are the j-th and k-th enterprises in the FIRM data set (1≤j≤m,1≤k≤m,j≠k) and proportion jk ∈(0,1].

4. The risk spillover simulation method based on equity knowledge graph according to claim 3 is characterized in that: The step 3 comprises: Traverse the S data set to obtain the equity data set Astock of venture enterprise r r , is a two-dimensional tuple, including triples (firm r ,proportion rf ,firm f ), where f∈[1,m]; Traverse the S data set to obtain the equity data set Dstock of venture enterprise r r , is a two-dimensional tuple consisting of a triple (firm l ,proportion lr ,firm r ), where l∈[1,m]; Traversing the FIRM enterprise dataset and Astock r Get the data set Afirm of companies held by venture firm r r , is a two-dimensional tuple, composed of firm o Constitute, where o∈[1,m] and satisfies firm o For Astock r The third element of the middle triple; Traversing FIRM enterprise dataset and Dstock r Get the enterprise data set Dfirm of the venture enterprise r r , is a two-dimensional tuple, composed of firm p Composition, where p∈[1,m] and satisfies firm p For Dstock r The first element of the triplet in .

5. The risk spillover simulation method based on equity knowledge graph according to claim 4 is characterized in that: The amount of assets Arisk of the risk enterprise r's liabilities in the risk enterprise r's enterprise o or and the proportion of the amount of assets affected Ashare or They are: Risk or =debt r ×proportion or Among them, debt r is the paid-in capital of venture enterprise r, proportion or is the proportion of equity investment of enterprise o to venture enterprise r in the equity investment received by venture enterprise r, asset o is the total assets of enterprise o.

6. The risk spillover simulation method based on equity knowledge graph according to claim 4 is characterized in that: The amount of equity investment Drisk of venture r's equity investment company p on the asset side pr and the proportion of equity investment affected Dshare pr They are: Drisk pr =investment p ×proportion rp Among them, investment p is the paid-in capital of enterprise p, proportion rp DEBT is the proportion of venture enterprise r’s equity investment in enterprise p to the equity investment received by enterprise p. p is the total liabilities of enterprise p.

7. A risk spillover simulation method based on equity knowledge graph according to claim 5 or 6, characterized in that: The determination of whether enterprise o and enterprise p are risky enterprises includes: If Ashare or >α ao , then enterprise o becomes a venture enterprise, otherwise it does not become a venture enterprise; if Dshare pr >α dp , then enterprise p becomes a venture enterprise, otherwise it does not become a venture enterprise; among them, α ao is the average ratio of investment loss amount to total assets amount when the enterprise triggers risk based on the industry in which enterprise o is located, α dp It is the average value of the proportion of the impairment of investment value caused by the decrease in other comprehensive income of the investee under the equity method to the total liabilities when the enterprise in the industry in which enterprise P is located triggers risk.

8. The risk spillover simulation method based on equity knowledge graph according to claim 7 is characterized in that: Neither the asset-side nor the liability-side enterprise triggers risk or transmits it to the target enterprise i. Stops include: If enterprise o is a risky enterprise, the target enterprise will remain enterprise i, the risky enterprise in step 2 will be set as enterprise o, and steps 2 to 4 will be repeated until neither its asset-side nor its liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i; if enterprise p is a risky enterprise, the target enterprise will remain enterprise i, the risky enterprise in step 2 will be set as enterprise p, and steps 2 to 4 will be repeated until neither its asset-side nor its liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i.

9. The risk spillover simulation method based on equity knowledge graph according to claim 8 is characterized in that: The final total amount of affected assets and funds of target enterprise i in step 5 Arisk i Total amount of debt funds affected i They are: The proportion of total assets and funds affected by target enterprise i Ashare i and the proportion of total liabilities of target enterprise i affected Dshare i They are: Among them, v and w are the set of enterprises that have risk spillovers on the target enterprise i’s asset side and the set of enterprises that have risk spillovers on the target enterprise i’s liability side, respectively. X and Y are the set of risk enterprises that affect the enterprise i’s funds from the asset side and the liability side, respectively. Arisk iv and Drisk iw are the total asset funds of enterprise i affected by venture enterprise v and the total liability funds of enterprise i affected by venture enterprise w, respectively; If Ashare i >α ai or Dshare i >α di , then the result is: Enterprise i is a risk enterprise, and the affected funds and proportions on the asset side are Arisk i and Ashare i The funds and proportions affected on the liability side are Drisk i and Dshare i ; Among them, α ai is the average ratio of investment loss amount to total assets amount when the enterprise triggers risk based on the industry in which enterprise i is located, α di It is the average value of the proportion of the investment value impairment amount caused by the decrease in other comprehensive income of the investee under the equity method to the total liability amount when the enterprise in the industry in which enterprise i is located triggers risk.

10. A risk spillover simulation system based on equity knowledge graph, characterized in that: include: Data acquisition module: used to acquire enterprise business data, create an enterprise data set FIRM, construct an equity data set S based on the enterprise data set, and construct an equity knowledge graph based on the enterprise data set FIRM and the equity data set S; Target enterprise and risk enterprise analysis module: It is used to analyze target enterprises based on the enterprise data set FIRM, based on factual risk or simulated risk, and select risk enterprise r and target enterprise i for analysis; Calculation module: used to traverse the equity data set S and the enterprise data set FIRM; Calculate the affected asset amount and the percentage of the affected asset amount of the company o holding the risk enterprise r on the liability side of the risk enterprise r, and calculate the affected equity investment amount and the percentage of the affected equity investment amount of the equity investment company p of the risk enterprise r on the asset side of the risk enterprise r; Risk enterprise judgment module: It is used to judge whether enterprise o and enterprise p become risk enterprises respectively, until neither the asset-side nor the liability-side enterprises trigger risks or the risks are transmitted to the target enterprise i; Total amount calculation module: It is used to calculate the final affected total asset funds and total liability funds of target enterprise i, as well as the proportion of the affected total asset funds of target enterprise i and the proportion of the affected total liability funds of target enterprise i after traversing all equity paths to target enterprise i; The risk spillover simulation system based on the equity knowledge graph is used to execute the steps in the risk spillover simulation method based on the equity knowledge graph described in any one of claims 1-9.

Images