Risk-associated enterprise group identification method and system based on fund backflow, and medium

Abstract

The invention discloses a risk-associated enterprise group identification method and system based on fund backflow, and a medium, mainly relates to the technical field of risk-associated enterprise group identification, and is used for solving the problems that an existing scheme depends on public information and data, the data size is huge and complex, the manual analysis efficiency is low, and errors are likely to occur. Comprising the following steps: acquiring a closed-loop transaction quantity, and extracting a closed-loop transaction enterprise group which corresponds to a preset identification parameter and is composed of enterprises with the closed-loop transaction quantity from an enterprise fund transaction network; obtaining a standard deviation of fund transaction scales between every two enterprises in the closed-loop transaction enterprise groups and a fund transaction scale mean value between the closed-loop transaction enterprise groups; calculating a discrete coefficient corresponding to the current closed-loop transaction enterprise group according to the standard deviation and the fund transaction scale mean value; and according to the falling interval of the discrete coefficient, determining a risk level corresponding to a closed-loop transaction enterprise group composed of the closed-loop transaction number of enterprises.

Description

Technical Field

[0001] This application relates to the technical field of risk-related enterprise group identification, and particularly to a method, system, and medium for identifying risk-related enterprise groups based on fund reflux. Background Art

[0002] The associated risk between enterprises is an important form of financial risk. In particular, the identification of implicit associated risks plays an important role in risk prevention and control.

[0003] The existing risk-related enterprise group identification solutions mainly include clarifying aspects such as equity structure, capital changes, and external investments. Specifically, commercial banks can query the registered capital, shareholders, and equity structure of enterprises through information platforms, and then conduct further extended inquiries on the registered capital, equity structure, and investment situations of shareholders to determine whether there are implicit associated situations in the enterprises. At the same time, by understanding the changes in the registered capital of enterprises, the adjustments of shareholders and equity structures, and analyzing the reasons for the changes, potential associated risks can also be discovered. In addition, reviewing the detailed accounts of enterprise investments and understanding the external investments and income situations of enterprises are also important ways to identify associated risks.

[0004] However, the existing risk-related enterprise group identification solutions also have some drawbacks. These solutions mainly rely on public information and data. Due to the large volume and complexity of the data, the efficiency of manual analysis is low and it is prone to errors. Summary of the Invention

[0005] In view of the above deficiencies of the prior art, this application provides a method, system, and medium for identifying risk-related enterprise groups based on fund reflux to solve the problems that the existing solutions rely on public information and data, and due to the large volume and complexity of the data, the efficiency of manual analysis is low and it is prone to errors.

[0006] In a first aspect, this application provides a method for identifying risk-related enterprise groups based on fund reflux. The method includes: Obtaining the number of closed-loop transactions, and then extracting a closed-loop transaction enterprise group composed of the number of enterprises corresponding to the preset identification parameters from the enterprise fund transaction network; obtaining the standard deviation of the fund transaction scale between any two enterprises within the closed-loop transaction enterprise group and the average value of the fund transaction scale between closed-loop transaction enterprise groups; calculating the coefficient of variation corresponding to the current closed-loop transaction enterprise group according to the standard deviation and the average value of the fund transaction scale; and determining the risk level corresponding to the closed-loop transaction enterprise group composed of the number of enterprises corresponding to the number of closed-loop transactions according to the interval in which the coefficient of variation falls.

[0007] In an implementation manner of the present application, before obtaining the number of closed-loop transactions and then extracting a closed-loop transaction enterprise group composed of the number of closed-loop transactions of enterprises corresponding to the preset identification parameters from the enterprise fund transaction network, the method further includes: Retain the data with the transaction debit and credit flag being debit in the bank fund flow data according to the customer name, counterparty name, transaction debit and credit flag, and the total transaction amount within the time range parameter; Import the data including the customer name, counterparty name, and total transaction amount into Neo4j to obtain the enterprise fund transaction network.

[0008] In an implementation manner of the present application, obtaining the number of closed-loop transactions and then extracting a closed-loop transaction enterprise group composed of the number of closed-loop transactions of enterprises corresponding to the preset identification parameters from the enterprise fund transaction network specifically includes: Obtain the number of closed-loop transactions through a preset acquisition interface; Use the knowledge graph method to identify a closed-loop transaction enterprise group that meets the size of the number of closed-loop transactions in the enterprise fund transaction network of Neo4j.

[0009] In an implementation manner of the present application, calculating the coefficient of variation corresponding to the current closed-loop transaction enterprise group according to the standard deviation and the average value of the fund transaction scale specifically includes: Through the formula: CV = σ / μ, calculate the coefficient of variation CV corresponding to the current closed-loop transaction enterprise group; Wherein, σ represents the standard deviation, and μ represents the average value of the fund transaction scale.

[0010] In an implementation manner of the present application, before determining the risk level corresponding to the closed-loop transaction enterprise group composed of the number of closed-loop transactions of enterprises according to the interval in which the coefficient of variation falls, the method further includes: Obtain the relationship between the coefficient of variation and the interval in which it falls through a preset interval acquisition interface; Wherein, the interval in which it falls includes at least low risk, medium risk, and high risk, and the magnitude of the coefficient of variation is proportional to the degree of risk.

[0011] In a second aspect, the present application provides a risk-related enterprise group identification system based on fund reflux. The system includes: An acquisition module for acquiring the number of closed-loop transactions, and then extracting a closed-loop transaction enterprise group consisting of the number of closed-loop transactions of enterprises corresponding to preset identification parameters from the enterprise capital transaction network; a calculation module for obtaining the standard deviation of the capital transaction scale between any two enterprises within the closed-loop transaction enterprise group and the average value of the capital transaction scale between the closed-loop transaction enterprise groups; calculating a coefficient of variation corresponding to the current closed-loop transaction enterprise group according to the standard deviation and the average value of the capital transaction scale; a determination module for determining the risk level corresponding to the closed-loop transaction enterprise group consisting of the number of closed-loop transactions of enterprises according to the interval in which the coefficient of variation falls.

[0012] In an implementation manner of the present application, the acquisition module includes an acquisition unit for retaining the data with the transaction debit flag as debit in the bank capital flow data according to the customer name, the name of the other party, the transaction debit and credit flag, and the total transaction amount within the time range parameter; importing the data including the customer name, the name of the other party, and the total transaction amount into Neo4j to obtain the enterprise capital transaction network.

[0013] In an implementation manner of the present application, the acquisition module includes an identification unit for acquiring the number of closed-loop transactions through a preset acquisition interface; using the knowledge graph method to identify a closed-loop transaction enterprise group that meets the size of the number of closed-loop transactions in the enterprise capital transaction network of Neo4j.

[0014] In an implementation manner of the present application, the calculation module includes a calculation unit for calculating the coefficient of variation CV corresponding to the current closed-loop transaction enterprise group through the formula: CV = σ / μ; where σ represents the standard deviation and μ represents the average value of the capital transaction scale.

[0015] In a third aspect, the present application provides a non-volatile computer storage medium, on which computer instructions are stored, and when the computer instructions are executed, a method for identifying a risk-related enterprise group based on capital reflux as described in any one of the above is implemented.

[0016] Those skilled in the art can understand that the present application has at least the following beneficial effects: 1. Improve the recognition efficiency and accuracy: Through automated algorithms and programs, it can efficiently process and analyze a large amount of enterprise capital transaction data, avoiding the low efficiency and error-proneness of manual analysis. By calculating the standard deviation of the capital transaction scale between any two enterprises within the closed-loop transaction enterprise group and the average value of the capital transaction scale between the groups, and then calculating the coefficient of variation, it can more accurately identify the enterprise groups with risk associations.

[0017] 2. Not relying on public information: Existing risk-related enterprise group identification solutions mainly rely on public information and data. However, this application directly analyzes the fund transaction data of enterprises, avoiding identification errors caused by incomplete or lagging public information. Facing the huge and complex enterprise fund transaction data, this application can effectively process and analyze it, extract key information, and provide strong support for risk identification.

[0018] 3. Providing a scientific basis for risk assessment: By calculating the coefficient of variation, this application can provide a quantitative risk assessment basis for enterprises, making risk identification more objective and scientific. According to the interval in which the coefficient of variation falls, this application can clearly divide the closed-loop transaction enterprise group into different risk levels, facilitating enterprises to take corresponding risk management measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the present invention, the accompanying drawings required in the description will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other accompanying drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 is a flowchart of a method for identifying a risk-related enterprise group based on fund reflux provided by an embodiment of this application.

[0021] Figure 2 is a schematic internal structure diagram of a system for identifying a risk-related enterprise group based on fund reflux provided by an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] Those skilled in the art should understand that the embodiments described below are only the preferred embodiments of the present disclosure, and do not mean that the present disclosure can only be implemented through these preferred embodiments. These preferred embodiments are only used to explain the technical principles of the present disclosure and are not used to limit the protection scope of the present disclosure. Based on the preferred embodiments provided by the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts should still fall within the protection scope of the present disclosure.

[0023] It should also be noted that the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "including one..." does not exclude the presence of additional identical elements in the process, method, commodity or device including the element.

[0024] The technical solutions proposed in the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0025] The embodiment provides a method for identifying a risk-related enterprise group based on capital reflux, as Figure 1 shown. The method provided in the embodiments of the present application mainly includes the following steps: Step 110: Obtain the number of closed-loop transactions, and then extract a closed-loop transaction enterprise group composed of the number of enterprises corresponding to the preset identification parameter from the enterprise capital transaction network.

[0026] It should be noted that the number of closed-loop transactions is the size of the closed-loop transaction, and the size of the closed-loop transaction is the number of enterprises participating in the closed-loop transaction. This step can locate the closed-loop transaction group in the enterprise capital transaction network and provide a basis for subsequent risk analysis. By setting the number of closed-loop transactions as the identification parameter, the accuracy and scope of identification can be flexibly adjusted.

[0027] The specific implementation manner of this step can be: First, summarize the bank capital flow data according to the customer name, counterparty name, transaction debit / credit flag, and time range, and retain the data with the transaction debit / credit flag being "debit". Import the processed data (including customer name, counterparty name, and total transaction amount) into the Neo4j graph database to construct an enterprise capital transaction network. Through a preset acquisition interface, the user can input or select the number of closed-loop transactions as the identification parameter. Use the knowledge graph method to identify a closed-loop transaction enterprise group that meets the size of the number of closed-loop transactions in the enterprise capital transaction network of Neo4j.

[0028] For example: Suppose the user sets the number of closed-loop transactions to 3, then the system will search for a closed-loop transaction group composed of 3 enterprises in the enterprise capital transaction network.

[0029] In addition, before obtaining the number of closed-loop transactions and then extracting a closed-loop transaction enterprise group composed of the number of enterprises corresponding to the preset identification parameter from the enterprise capital transaction network, the present application may further include: Retain the data with the debit transaction flag among the bank fund flow data according to the customer name, the name of the counterparty, the transaction debit / credit flag, and the total transaction amount within the time range parameter. Import the data including the customer name, the name of the counterparty, and the total transaction amount into Neo4j to obtain the enterprise fund transaction network.

[0030] It should be noted that by importing the data into the Neo4j graph database, it is convenient to use graph theory algorithms for subsequent analysis.

[0031] The specific implementation method can be: Use programming languages such as SQL or Python to clean, summarize, and filter the bank fund flow data. Use the Cypher query language of Neo4j or the Neo4j driver library of Python (such as py2neo) to import the data into the Neo4j graph database.

[0032] In some embodiments, obtain the number of closed-loop transactions, and then extract a closed-loop transaction enterprise group composed of the number of enterprises corresponding to the preset identification parameters from the enterprise fund transaction network. Specifically, it can be: Through a preset acquisition interface, obtain the number of closed-loop transactions; use the knowledge graph method to identify a closed-loop transaction enterprise group that meets the size of the number of closed-loop transactions in the enterprise fund transaction network of Neo4j.

[0033] Step 120: Obtain the standard deviation of the fund transaction scale between every two enterprises within the closed-loop transaction enterprise group and the average value of the fund transaction scale between the closed-loop transaction enterprise groups; calculate the corresponding coefficient of variation for the current closed-loop transaction enterprise group according to the standard deviation and the average value of the fund transaction scale.

[0034] It should be noted that in this step, by calculating the standard deviation and the average value, the difference in the fund transaction scale within the closed-loop transaction enterprise group can be quantified. The coefficient of variation (CV) is a dimensionless index, which can conveniently compare the risk levels of different closed-loop transaction enterprise groups.

[0035] The specific implementation method involved in this step can be: For each closed-loop transaction enterprise group, calculate the standard deviation (σ) and the average value (μ) of the fund transaction scale between every two enterprises within it.

[0036] Use the formula CV = σ / μ to calculate the corresponding coefficient of variation for the current closed-loop transaction enterprise group.

[0037] Example: Suppose there are 3 enterprises within a group of closed-loop trading enterprises, and the scale of their capital transactions with each other is 1 million, 2 million, and 3 million respectively. Then, the standard deviation and mean of the capital transaction scale of these 3 enterprises can be calculated, and further the coefficient of variation can be obtained.

[0038] Step 130: Determine the risk level corresponding to the closed-loop trading enterprise group composed of the number of enterprises in the closed-loop transaction according to the interval in which the coefficient of variation falls.

[0039] It should be noted that in this step, by comparing the coefficient of variation with the preset risk level interval, the risk level of the closed-loop trading enterprise group can be quickly determined. The division of the risk level helps enterprises take targeted risk management measures.

[0040] The specific implementation method can be: Obtain the risk level interval: Through the preset interval acquisition interface or configuration file, obtain the corresponding relationship between the coefficient of variation and the risk level interval.

[0041] Determine the risk level: According to the calculated coefficient of variation, find its corresponding risk level interval, and determine the risk level of the closed-loop trading enterprise group. The smaller the CV, the higher the degree of capital return. CV = 0 means complete capital return.

[0042] For example: Suppose the preset risk level interval is: CV < 0.5 is low risk, 0.5 ≤ CV < 1 is medium risk, CV ≥ 1 is high risk. Then, for the closed-loop trading enterprise group with the calculated coefficient of variation CV = 0.7, it can be classified as the medium risk level.

[0043] In some embodiments, before determining the risk level corresponding to the closed-loop trading enterprise group composed of the number of enterprises in the closed-loop transaction according to the interval in which the coefficient of variation falls, this application further includes: Obtain through the preset interval acquisition interface the relationship between the coefficient of variation and the falling interval; wherein, the falling interval at least includes low risk, medium risk, and high risk, and the size of the coefficient of variation is proportional to the degree of risk.

[0044] It should be noted that this application provides a way to obtain the risk level interval here, making the division of the risk level more flexible and configurable.

[0045] In addition, this application Figure 2 is a risk-related enterprise group identification system based on capital return provided by an embodiment of this application. As Figure 2 shown, the system provided by the embodiment of this application mainly includes: An acquisition module 210 is configured to acquire the number of closed-loop transactions, and then extract a closed-loop transaction enterprise group composed of the number of enterprises corresponding to the preset identification parameter from the enterprise fund transaction network.

[0046] It should be noted that the number of closed-loop transactions is the size of the closed-loop transaction, and the size of the closed-loop transaction is the number of enterprises participating in the closed-loop transaction. This step can locate the closed-loop transaction group in the enterprise fund transaction network and provide a basis for subsequent risk analysis. By setting the number of closed-loop transactions as the identification parameter, the accuracy and scope of identification can be flexibly adjusted.

[0047] The specific implementation manner of the acquisition module 210 may be: First, summarize the bank fund flow data according to the customer name, counterparty name, transaction debit / credit flag, and time range, and retain the data with the transaction debit / credit flag being "debit". Import the processed data (including customer name, counterparty name, and total transaction amount) into the Neo4j graph database to construct the enterprise fund transaction network. Through a preset acquisition interface, the user can input or select the number of closed-loop transactions as the identification parameter. Use the knowledge graph method to identify a closed-loop transaction enterprise group that meets the size of the number of closed-loop transactions in the enterprise fund transaction network of Neo4j.

[0048] For example: Suppose the user sets the number of closed-loop transactions to 3, then the system will search for a closed-loop transaction group composed of 3 enterprises in the enterprise fund transaction network.

[0049] The acquisition module 210 includes an acquisition unit, which is configured to summarize the bank fund flow data according to the customer name, counterparty name, transaction debit / credit flag, and total transaction amount within the time range parameter, and retain the data with the transaction debit / credit flag being debit; import the data including customer name, counterparty name, and total transaction amount into Neo4j to obtain the enterprise fund transaction network.

[0050] It should be added that by importing the data into the Neo4j graph database, the acquisition unit can conveniently use graph theory algorithms for subsequent analysis.

[0051] The specific implementation manner of the acquisition unit may be: Use programming languages such as SQL or Python to clean, summarize, and filter the bank fund flow data. Use the Cypher query language of Neo4j or the Neo4j driver library of Python (such as py2neo) to import the data into the Neo4j graph database.

[0052] The acquisition module 210 includes an identification unit for obtaining the number of closed-loop transactions through a preset acquisition interface, and identifying a group of closed-loop transaction enterprises that meet the size of the number of closed-loop transactions in the enterprise fund transaction network of Neo4j using the knowledge graph method.

[0053] The calculation module 220 is used to obtain the standard deviation of the fund transaction scale between every two enterprises within the group of closed-loop transaction enterprises and the average value of the fund transaction scale between the groups of closed-loop transaction enterprises, and calculate the corresponding coefficient of variation for the current group of closed-loop transaction enterprises based on the standard deviation and the average value of the fund transaction scale.

[0054] The calculation module 220 includes a calculation unit for calculating, through the formula: CV = σ / μ, the corresponding coefficient of variation CV for the current group of closed-loop transaction enterprises; where σ represents the standard deviation and μ represents the average value of the fund transaction scale.

[0055] It should be noted that by calculating the standard deviation and the average value, the calculation module 220 can quantify the difference in the fund transaction scale within the group of closed-loop transaction enterprises. The coefficient of variation (CV) is a dimensionless index that can conveniently compare the risk levels of different groups of closed-loop transaction enterprises.

[0056] The specific implementation method involved in the calculation module 220 can be: For each group of closed-loop transaction enterprises, calculate the standard deviation (σ) and the average value (μ) of the fund transaction scale between every two enterprises within it.

[0057] Use the formula CV = σ / μ to calculate the corresponding coefficient of variation for the current group of closed-loop transaction enterprises.

[0058] For example: Suppose there are 3 enterprises within a group of closed-loop transaction enterprises, and the fund transaction scales between them are 1 million, 2 million, and 3 million respectively. Then, the calculation module 220 can calculate the standard deviation and the average value of the fund transaction scales of these 3 enterprises, and further obtain the coefficient of variation.

[0059] The determination module 230 is used to determine the risk level corresponding to the group of closed-loop transaction enterprises composed of the number of enterprises in the closed-loop transaction according to the interval into which the coefficient of variation falls.

[0060] It should be noted that by comparing the coefficient of variation with the preset risk level interval, the determination module 230 can quickly determine the risk level of the group of closed-loop transaction enterprises. The division of the risk level helps enterprises take targeted risk management measures.

[0061] The specific implementation method of the determination module 230 can be: Obtain the risk level interval: Obtain the correspondence between the coefficient of variation and the risk level interval through a preset interval acquisition interface or configuration file.

[0062] Determine the risk level: According to the calculated coefficient of variation, find its corresponding risk level interval and determine the risk level of the closed-loop trading enterprise group. The smaller the CV, the higher the degree of capital return. CV = 0 means complete capital return.

[0063] Example: Suppose the preset risk level intervals are: CV < 0.5 is low risk, 0.5 ≤ CV < 1 is medium risk, and CV ≥ 1 is high risk. Then, for the closed-loop trading enterprise group with the calculated coefficient of variation CV = 0.7, the determination module 230 can classify it as the medium risk level.

[0064] In addition, the embodiment of the present application also provides a non-volatile computer storage medium, on which executable instructions are stored. When the executable instructions are executed, the above-mentioned method for identifying a risk-related enterprise group based on capital return is implemented.

[0065] So far, the technical solutions of the present disclosure have been described in combination with multiple foregoing embodiments. However, it is easy for those skilled in the art to understand that the protection scope of the present disclosure is not limited to these specific embodiments. Without departing from the technical principle of the present disclosure, those skilled in the art can split and combine the technical solutions in the above-mentioned various embodiments, and can also make equivalent changes or replacements to the relevant technical features. Any changes, equivalent replacements, improvements, etc. made within the technical concept and / or technical principle of the present disclosure will fall within the protection scope of the present disclosure.

Claims

1. A method for identifying risk-related enterprise groups based on capital reflow, characterized in that: The method comprises: Obtain the number of closed-loop transactions, and then extract a closed-loop transaction enterprise group consisting of enterprises with the number of closed-loop transactions corresponding to preset identification parameters from the enterprise capital transaction network; Obtain the standard deviation of the fund transaction scale between any two enterprises in the closed-loop transaction enterprise group and the mean fund transaction scale between the closed-loop transaction enterprise group; calculate the dispersion coefficient corresponding to the current closed-loop transaction enterprise group based on the standard deviation and the mean fund transaction scale; According to the interval in which the dispersion coefficient falls, the risk level corresponding to the closed-loop transaction enterprise group consisting of a number of closed-loop transaction enterprises is determined.

2. The method for identifying risk-related enterprise groups based on capital reflow according to claim 1, characterized in that: Before obtaining the number of closed-loop transactions and then extracting a closed-loop transaction enterprise group consisting of the number of closed-loop transaction enterprises corresponding to preset identification parameters from the enterprise capital transaction network, the method further includes: The bank fund flow data is aggregated according to the customer name, the other party's account name, the transaction debit / credit mark, and the transaction amount within the time range parameters, and the data with the transaction debit / credit mark as debit is retained; Import the data including customer name, other party's account name and summary transaction amount into Neo4j to obtain the enterprise capital transaction network.

3. The method for identifying risk-related enterprise groups based on capital reflow according to claim 1, characterized in that: The number of closed-loop transactions is obtained, and then a closed-loop transaction enterprise group consisting of the number of closed-loop transaction enterprises corresponding to the preset identification parameters is extracted from the enterprise capital transaction network, specifically including: Obtain the number of closed-loop transactions through the preset acquisition interface; The knowledge graph method is used to identify closed-loop transaction enterprise groups with the same number of closed-loop transactions in the Neo4j enterprise capital transaction network.

4. The method for identifying risk-related enterprise groups based on capital reflow according to claim 1, characterized in that: According to the standard deviation and the mean value of fund transaction scale, the dispersion coefficient corresponding to the current closed-loop transaction enterprise group is calculated, including: By formula: CV=σ / μ, calculate and obtain the discrete coefficient CV corresponding to the current closed-loop transaction enterprise group; Among them, σ represents the standard deviation and μ represents the mean value of fund transaction size.

5. The method for identifying risk-related enterprise groups based on capital reflow according to claim 1, characterized in that: Before determining the risk level corresponding to the closed-loop transaction enterprise group consisting of the number of closed-loop transaction enterprises according to the interval in which the dispersion coefficient falls, the method further includes: The relationship between the discrete coefficient and the falling interval is obtained through the preset interval acquisition interface; Among them, the interval includes at least low risk, medium risk and high risk, and the size of the dispersion coefficient is proportional to the degree of risk.

6. A risk-related enterprise group identification system based on capital reflow, characterized in that: The system comprises: An acquisition module is used to acquire the number of closed-loop transactions, and then extract a closed-loop transaction enterprise group consisting of enterprises with the number of closed-loop transactions corresponding to preset identification parameters from the enterprise capital transaction network; A calculation module is used to obtain the standard deviation of the fund transaction scale between any two enterprises in the closed-loop transaction enterprise group and the mean fund transaction scale between the closed-loop transaction enterprise group; based on the standard deviation and the mean fund transaction scale, the dispersion coefficient corresponding to the current closed-loop transaction enterprise group is calculated; The determination module is used to determine the risk level corresponding to a closed-loop transaction enterprise group consisting of a number of closed-loop transaction enterprises according to the interval in which the discrete coefficient falls.

7. The risk-related enterprise group identification system based on capital reflow according to claim 6 is characterized in that: The acquisition module includes an acquisition unit, It is used to aggregate bank fund flow data according to customer name, counterparty account name, transaction debit / credit mark and transaction amount within the time range parameters, and retain data with transaction debit / credit mark as debit; Import the data including customer name, other party's account name and summary transaction amount into Neo4j to obtain the enterprise capital transaction network.

8. The risk-related enterprise group identification system based on capital reflow according to claim 6 is characterized in that: The acquisition module includes a recognition unit, Used to obtain the number of closed-loop transactions through the preset acquisition interface; The knowledge graph method is used to identify closed-loop transaction enterprise groups with the same number of closed-loop transactions in the Neo4j enterprise capital transaction network.

9. The risk-related enterprise group identification system based on capital reflow according to claim 6 is characterized in that: The computing module includes a computing unit, Used by the formula: CV=σ / μ, calculate and obtain the discrete coefficient CV corresponding to the current closed-loop transaction enterprise group; Among them, σ represents the standard deviation and μ represents the mean value of fund transaction size.

10. A non-volatile computer storage medium, characterized in that: Computer instructions are stored thereon, and when the computer instructions are executed, the method for identifying a group of risk-associated enterprises based on capital reflow as described in any one of claims 1 to 5 is implemented.

Images