A method and system for early warning of enterprise risks based on relationship graph

Abstract

The present invention discloses a method and system for early warning of enterprise risks based on a relationship map, the method comprising the following steps: according to a user request, retrieving the equity map of a target enterprise; the user request includes information of the target enterprise; calling the business data of a local database, the business data including business data of several group subsidiaries; matching the business data with the equity map, and establishing a business data relationship map. Performing risk assessment on the target enterprise information, the business data relationship map, the business data, and historical enterprise risk / project risk data to obtain a risk assessment result of the target enterprise; performing relevant early warning operations according to the risk assessment result. This technical solution has a high degree of fit with the actual business when conducting risk assessment because it expands the external equity map and combines it with the expanded business data relationship map. From a global perspective, it effectively solves the risk assessment needs of group-type enterprises.

Description

Technical Field

[0001] The present invention belongs to the technical field of enterprise risk early warning, and specifically relates to an enterprise risk early warning method and system based on a relationship graph. Background Art

[0002] Enterprise risk warning is achieved by collecting the enterprise's business management information and then conducting risk assessment based on the collected information. When the enterprise is at risk, it is necessary to control the risk factors in a timely manner and make risk decisions in a timely manner to avoid a major impact on the enterprise.

[0003] In the common scenario of risk assessment for external enterprises within a group, the traditional risk assessment method generally only collects information on the enterprise to be warned, and currently only considers the risk factors of the enterprise to be warned. However, at this stage, the equity structure of enterprises is often complex, and there are many other enterprises associated with the enterprise to be warned, such as the upper-level enterprises of the equity relationship and subordinate subsidiaries. The risk factors of the associated enterprises associated with the enterprise to be warned will also bring potential risks to the enterprise to be warned. If only the risks of the enterprise to be warned are considered, the risk assessment of the enterprise to be warned is not comprehensive enough, which leads to the limitation of risk assessment; and it is easy to cause a lag in risk assessment, which leads to a lag in risk decision-making. Summary of the invention

[0004] The purpose of the present invention is to solve the above-mentioned technical problems and provide an enterprise risk early warning method and system based on relationship graph.

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

[0006] In a first aspect, the present invention provides an enterprise risk early warning method based on a relationship graph, characterized in that it comprises the following steps:

[0007] Retrieving the equity map of the target enterprise according to the user request; the user request includes the target enterprise information;

[0008] Calling business data from a local database, wherein the business data includes business data of several group subsidiaries;

[0009] Matching the business data with the equity graph to establish a business data relationship graph;

[0010] Perform risk assessment on the target enterprise information, the business data relationship map, the business data and historical enterprise risk / project risk data to obtain a risk assessment result for the target enterprise;

[0011] Perform relevant early warning operations based on the risk assessment results.

[0012] In combination with the first aspect, the present invention provides a first preferred implementation manner of the first aspect, specifically, performing risk assessment on the target enterprise information, the business data relationship map, the business data and the historical enterprise risk / project risk data to obtain a risk assessment result of the target enterprise, specifically including:

[0013] Conduct risk assessment on the target enterprise and obtain the target enterprise's own assessment results;

[0014] Obtaining the associated enterprises of each node of the business data relationship graph, conducting an impact assessment on each associated enterprise, and obtaining the impact risk results of each associated enterprise on the target enterprise;

[0015] Based on the target enterprise's own assessment results and several risk impact results, the risk assessment results of the target enterprise are comprehensively calculated.

[0016] In combination with the first aspect, the present invention provides a second preferred implementation of the first aspect, specifically, the risk assessment of the target enterprise specifically includes:

[0017] The risk assessment is provided with several first assessment dimensions, which include: corporate background of the target enterprise, corporate development history and current situation, risk information, intellectual property information, operating information, annual report information, financial information, independent information, and business transaction data between the target enterprise and the group.

[0018] In combination with the first aspect, the present invention provides a third preferred implementation manner of the first aspect, specifically, the impact assessment on each associated enterprise specifically includes:

[0019] The impact assessment is provided with several second assessment dimensions, which include: equity ratio, shareholding amount, hierarchical distance of business data relationship map, business type and internal business transactions;

[0020] Each second assessment dimension is set with a different risk impact weight coefficient, and the risk impact result is calculated based on the weight coefficient.

[0021] In combination with the first aspect, the present invention provides a fourth preferred implementation of the first aspect, specifically, comprehensively calculating the risk assessment result of the target enterprise, specifically including:

[0022] The risk assessment result is provided with several third assessment dimensions, and the third assessment dimensions include one or more combinations of credit risk, market risk, liquidity risk, operational risk, industry risk, legal / regulatory / policy risk, personnel risk and natural disasters.

[0023] In combination with the first aspect, the present invention provides a fifth preferred implementation of the first aspect, specifically, matching the business data with the equity graph specifically includes the following steps:

[0024] Extracting all project information contained in the business data, each project information including several key fields;

[0025] Using the key fields of each of the project information as query statements one by one, traversing all nodes matching the equity graph;

[0026] If the key field is the same as the information of any node in the equity graph, the project information is successfully matched, and the association relationship between the current project information and the current node is established;

[0027] If the key field is not the same as the information of all nodes in the equity graph, the project information matching is unsuccessful;

[0028] Stop until all project information is matched.

[0029] In combination with the first aspect, the present invention provides a sixth preferred implementation of the first aspect, specifically, traversing all nodes matching the equity graph, specifically including:

[0030] Perform traversal matching on all nodes of the equity graph according to the set traversal order;

[0031] The traversal order is composed of multiple key fields of a single project information arranged in a preset order.

[0032] In combination with the first aspect, the present invention provides a seventh preferred implementation of the first aspect, specifically, when any key field in the traversal order successfully matches any node of the equity graph, the traversal matching of the current project information is terminated;

[0033] Perform traversal matching of the next item information.

[0034] In combination with the first aspect, the present invention provides an eighth preferred implementation scheme of the first aspect. Specifically, the traversal order includes the direct transaction counterparties of the project, the group to which the direct transaction counterparties of the project belong, the actual controllers of the direct transaction counterparties of the project and the project guarantors arranged in order of priority.

[0035] In a second aspect, the present invention provides an enterprise risk early warning system based on a relationship graph, the enterprise risk early warning system comprising:

[0036] A retrieval module, which is used to retrieve the equity map of the target enterprise according to the user request; the user request includes the target enterprise information;

[0037] A calling module, which is used to call the business data of the local database, wherein the business data includes the business data of several group subsidiaries;

[0038] A matching module, which is used to match the business data with the equity graph to establish a business data relationship graph;

[0039] An assessment module, which is used to conduct risk assessment on the target enterprise information, the business data relationship map, the business data and historical enterprise risk / project risk data to obtain a risk assessment result of the target enterprise;

[0040] The early warning module is used to perform relevant early warning operations according to the risk assessment results.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention provides an enterprise risk early warning method based on a relationship map, comprising the following steps: according to a user request, retrieve the equity map of the target enterprise; the user request includes the target enterprise information; call the business data of the local database, the business data includes the business data of several group subsidiaries; match the business data with the equity map, and establish a business data relationship map. Perform risk assessment on the target enterprise information, the business data relationship map, the business data and the historical enterprise risk / project risk data to obtain the risk assessment result of the target enterprise; and perform relevant early warning operations according to the risk assessment result.

[0043] On the basis of the equity map, the present invention introduces the business systems of the subsidiaries under the group for data docking, associates the business data of all subsidiaries with the equity map, and forms a multi-dimensionally associated equity map; when conducting risk assessment, it not only includes the self-assessment of the external enterprise, but also can consider the risk factors of the target enterprise in multiple dimensions based on the business data relationship picture, and solves the problems of delayed and inaccurate assessment due to the lack of reference dimensions for risk assessment. The technical solution expands the external equity map and combines it with the expanded business data relationship map, so that the risk assessment is highly consistent with the actual business, and from a global perspective, it effectively solves the risk assessment needs of group-type enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings, wherein:

[0045] Figure 1 It is a schematic diagram of an application scenario of an enterprise risk early warning system based on a relationship graph of the present invention;

[0046] Figure 2 It is a flow chart of an enterprise risk early warning method based on a relationship map of the present invention;

[0047] Figure 3 It is a schematic diagram of the equity map of the target enterprise of the present invention;

[0048] Figure 4 is an example diagram of project information of business data of the present invention;

[0049] Figure 5 It is the marked equity map in Embodiment 2 of the present invention;

[0050] Figure 6 is the identification relationship list in Embodiment 2 of the present invention;

[0051] Figure 7 It is the business data relationship map established by the present invention;

[0052] Figure 8 is the application diagram in Example 3 of the present invention;

[0053] Fig. 9 It is a description list of association relationships that have been successfully matched in Example 3 of the present invention. DETAILED DESCRIPTION

[0054] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0057] In order to enable those skilled in the art to use the contents of this application, the following implementation method is given in combination with the specific application scenario "In the field of financial business, when a group cooperates with an external enterprise, it is usually necessary to obtain the equity relationship map of the external enterprise and conduct an investigation on cooperation with the external enterprise". For those skilled in the art, the general principles defined here can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application. Although this application is mainly described around the theme of "In the field of financial business, when a group cooperates with an external enterprise, it is usually necessary to obtain the equity relationship map of the external enterprise and conduct an investigation on cooperation with the external enterprise", it should be understood that this is only an exemplary embodiment.

[0058] In the existing technology, the enterprise equity map has the following defects:

[0059] Group-affiliated companies are common in market economy, and the group includes multiple different types of financial institutions. However, due to the relative independence of each financial institution, and the problems of information asymmetry and lack of intercommunication between financial institutions, it is difficult for each financial institution to effectively use group-level information data in the risk management of external corporate customers. On the other hand, when each financial institution combines the equity relationship map of external enterprises with internal business data for investigation, there is a problem of single dimension.

[0060] To this end, the present invention is committed to providing an enterprise risk early warning method and system based on relationship graph.

[0061] Example 1

[0062] Figure 1 A schematic diagram of an application scenario of an enterprise risk early warning system based on a relationship graph provided by the present invention, such as Figure 1 As shown, it includes the enterprise risk early warning system based on relationship graph of the present invention, several group subsidiary business systems, a local database server and a third-party service system.

[0063] The group subsidiary business system stores the business data of each subsidiary. A subsidiary is a subsidiary of the group, such as a financial group that includes several financial service institutions. The local database server stores business data and historical enterprise risk / project risk data. The business data includes business data of several group subsidiaries, and the local database server is also used to store equity maps and relevant data of the enterprise risk early warning system.

[0064] In a preferred implementation, the enterprise risk early warning system is connected to the business systems of the subsidiaries of the group, and the business data of all subsidiaries is extracted regularly, and the business data is uniformly maintained in the system and then fed back to the business systems of each subsidiary. Specifically, the local database is connected to the business systems of several group subsidiaries, and the local database obtains all business data of all group subsidiaries according to a preset period to update the business data of the local database server.

[0065] Specifically, the third-party service system is a third-party platform that provides enterprise information query services. For example, when creating an equity map for an external enterprise for the first time, the equity map of the target enterprise is retrieved through a third-party interface provided by the third-party service system. For example, third-party platforms such as Qichacha, Tianyancha, and Qixinbao.

[0066] The enterprise risk early warning system based on the relationship map includes a retrieval module, a call module and a matching module. The retrieval module is used to retrieve the equity map of the target enterprise according to the user request; the user request includes the target enterprise information. The call module is used to call the business data of the local database; the matching module is used to match the business data with the equity map to establish a business data relationship map.

[0067] On the basis of the equity map, the present invention introduces the business systems of the subsidiaries under the group to carry out data docking, associates the business data of all subsidiaries with the equity map, forms a multi-dimensionally associated equity map, and effectively solves the needs of group-type enterprises for external enterprise cooperation inspections.

[0068] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0069] Example 2

[0070] Figure 2 This is a flow chart of an enterprise risk warning method based on a relationship graph provided by the present invention. The executor of this embodiment can be the enterprise risk warning system based on the relationship graph in Example 1. Specifically, the enterprise risk warning system can execute and implement the enterprise risk warning method based on the relationship graph in the form of software or hardware.

[0071] like Figure 2 As shown, the enterprise risk early warning method of this embodiment may include:

[0072] S100: Retrieving the equity map of the target enterprise according to the user request; the user request includes the target enterprise information.

[0073] S200: calling the business data of the local database, wherein the business data includes the business data of several group subsidiaries;

[0074] S300: Matching the business data with the equity graph to establish a business data relationship graph;

[0075] S400: Perform risk assessment on the target enterprise information, the business data relationship map, the business data and historical enterprise risk / project risk data to obtain a risk assessment result of the target enterprise;

[0076] S500: Execute relevant warning operations based on the risk assessment results.

[0077] Specifically, the present invention describes in detail each step of the enterprise risk early warning method as follows:

[0078] S100: Retrieving the equity map of the target enterprise according to the user request; the user request includes the target enterprise information.

[0079] In one implementation, the enterprise risk early warning system of the present invention can be installed and deployed in a terminal, such as various personal computers, laptops, smart phones, tablet computers, etc. The user enters the enterprise risk early warning system to obtain the equity map of the target enterprise. The terminal generates a user request for obtaining the equity map of the target enterprise by responding to the user's input operation on the corresponding UI interface of the enterprise risk early warning system. The user request carries the target enterprise information of the enterprise to be queried input by the user, such as the enterprise name, enterprise credit code, etc. The terminal system parses the user request and uses it to access a third-party system or access a local database server to obtain the equity map of the target enterprise.

[0080] In a specific implementation, the enterprise risk warning system can be configured in the form of a scheduled task to regularly trigger the entire risk assessment process, or it can be manually triggered by a user to generate a user request.

[0081] In another implementation, the user can also enter the enterprise risk warning interface based on the relationship map by logging into the browser running in the terminal. The user enters the target enterprise information based on the constructed interface, and the user request carrying the target enterprise information is sent through the terminal to the local database server or third-party service system to request the corresponding equity map.

[0082] To this end, the present invention also provides a specific implementation process in S100, which specifically includes:

[0083] S110: According to the user request, query whether the local database stores the equity map of the target enterprise.

[0084] In the present invention, according to the target enterprise identifier in the user request, it is first queried in the local database whether there is a stored equity map of the target enterprise.

[0085] S111: If yes, retrieve the equity map stored in the local database and return it;

[0086] If not, then, based on the target enterprise information, obtain the target enterprise's equity map through a third-party interface;

[0087] S120: Return the equity map and store the equity map in the local database.

[0088] In a preferred implementation, in step S120, the equity map is stored in a local database, and an effective period is set for the storage of the equity map. When the user queries again within a certain period, the local data is directly displayed. When the effective period expires, the equity map is obtained from the third-party service system again for update, that is, as long as the user queries, searches or requests to establish an external enterprise again, the process of obtaining the equity relationship map of the external enterprise and matching it with the internal business data association relationship will be automatically triggered again.

[0089] In one example, each time a user requests to retrieve the equity map of an external enterprise, the validity period of the equity map of the external enterprise is pre-identified. When the validity period of the equity map of the target enterprise exceeds, the equity map of the enterprise is retrieved and stored again from the third-party service system according to the current user request, and an enterprise risk warning method based on the relationship map is executed to update the business data relationship map of the local database.

[0090] For example, the validity period can be set to 30 days, 60 days, etc.

[0091] In the existing technology, when the group conducts cooperative inspections on external enterprises, there are the following problems: 1. Business personnel need to manually obtain equity penetration maps from websites such as Tianyancha, and each search will be charged. 2. Even if this search process becomes an automated program that provides a unified query interface, when the equity of external enterprises changes or the business data in the group system changes, the previous inspection results will be inaccurate, and manual data association will be labor-intensive.

[0092] To this end, in response to the above-mentioned problems, the present invention provides a targeted solution through steps S110 to S120, and each time a new external enterprise's equity relationship map is obtained, it will be saved in the local database, and the local data will be directly displayed when the user queries again within a certain period of time. On the other hand, as long as the user searches for the same external enterprise again, the process of obtaining the external enterprise's equity relationship map and matching it with the internal business data association relationship will be automatically retriggered. And through this design, when the equity of an external enterprise changes and the business data in the group system changes, it will also be updated regularly, thereby improving the accuracy of the inspection results, and there is no need for manual data association, saving human capital.

[0093] In a preferred implementation, in step S120, the equity map is stored in a local database, and a validity period is set for the storage of the equity map. If the user queries again within a certain period, the local data is directly displayed. Otherwise, each time a user requests to retrieve the equity map of an external enterprise, the validity period of the equity map of the external enterprise is pre-identified. When the validity period of the target enterprise exceeds the validity period, the equity map of the enterprise is retrieved from the third-party service system and stored again according to the user's request.

[0094] In a specific example, taking "Qixinbao" as an example of a third-party service system, the equity map of the target enterprise is obtained.

[0095] (1) When a user searches for target enterprise A, the system first obtains the suspected actual controller F of target enterprise A on Qixinbao through the third-party interface "suspected actual controller" provided by Qixinbao, and then uses the third-party interface "six-layer family tree" provided by Qixinbao to realize the equity relationship map of external enterprises from F downward. The resulting equity map of external enterprises is as follows: Figure 3 ;

[0096] (2) When a user searches for any enterprise in the six layers below F, he will obtain the same equity map, which is then stored locally;

[0097] (3) When a user actively searches for a target enterprise, the external interface is called to generate a new equity map. However, if the target enterprise searched by the user already has a generated equity map or business data relationship map in the local database, the local map is directly displayed (without incurring additional call fees).

[0098] (4) If F contains both companies and people, the user is allowed to select one of the companies to form an external relationship map, and the person's name is saved for subsequent relationship matching.

[0099] (5) The third-party interface "six-layer family tree" provided by Qixinbao is used to realize the external equity relationship map from F downward. Since there are many hierarchical relationships in the map, consider folding some nodes and displaying a total of 4 layers including F;

[0100] (6) After obtaining the equity map of the target enterprise, target enterprises A and F are connected separately with a dotted line when displaying. The line indicates that the relationship between target enterprises A and F is "suspected actual controller". There are several situations between target enterprises A and F: target enterprise A is in the equity penetration of F downward (six layers); the map relationship between target enterprises A and F exceeds six layers; target enterprises A and F are not established on the basis of equity relationship (may be through a partnership, etc.)

[0101] S200: calling business data from a local database, where the business data includes business data of several group subsidiaries.

[0102] In the present invention, the business data includes all project information of each group subsidiary. In one example, the business data includes multiple projects, and the project information of each project includes the direct counterparty of the project, the group to which the direct counterparty of the project belongs, the actual controller of the direct counterparty of the project, and the guarantor of the project.

[0103] In the present invention, each project information includes several key fields, which are used to associate and match business data with equity maps. The object information of the direct counterparty of each project, the group to which the direct counterparty belongs, the actual controller of the direct counterparty, and the guarantor of the project can be used as key fields.

[0104] like Figure 4 As shown in the following screenshot, the local database contains a lot of project information (several thousand projects). Assuming the project information of business data, there are 17 projects, which are the projects of subsidiaries M1 to M6 under the group. The key fields include the project direct counterparty 001A of project 001, the group 001B to which the project direct counterparty of project 001 belongs, and so on.

[0105] In a preferred implementation, the key fields are divided into regular fields and manual judgment fields. Figure 4 The "direct counterparty", "group", "actual controller" and "project guarantor" fields in the manual judgment field are used to match the external enterprise equity relationship map. The manual judgment field is used to identify matching relationships that require manual judgment. If a matching relationship that requires manual judgment is identified, manual intervention judgment and submission will be performed on another editing page.

[0106] In a specific implementation, the matching relationships that require manual judgment include "affiliated group" / "actual controller" / "project guarantor".

[0107] S300: Match the business data with the equity graph to establish a business data relationship graph.

[0108] In the present invention, matching the business data with the equity graph specifically includes the following steps:

[0109] S310: extracting all project information contained in the business data, each project information including several key fields.

[0110] S320: Using the key fields of each of the project information as query statements one by one, traverse all nodes matching the equity graph.

[0111] In one example, the first key field of project 001 is the project's direct counterparty 001A, and 001A is used as a query statement to traverse all nodes that match the equity graph; the second key field of project 001 is the group 001B to which the project's direct counterparty belongs, and 001B is used as a query statement... and so on.

[0112] In a preferred implementation, traversal matching is performed on all nodes of the equity map according to a set traversal order; wherein the traversal order is composed of multiple key fields of a single project information arranged in a preset order.

[0113] In a specific implementation, the traversal order includes the direct transaction counterparties of the project, the group to which the direct transaction counterparties of the project belong, the actual controllers of the direct transaction counterparties of the project, and the project guarantors arranged in order of priority.

[0114] S321: If the key field is the same as the information of any node of the equity map, the project information is successfully matched, and an association relationship between the current project information and the current node is established.

[0115] In a specific implementation, the first key field 001A of item 001 is used as a query statement to traverse all nodes matching the equity graph. For example, when Figure 4 The information of the J6 node in is also 001A, that is, the key field is the same as the information of any node in the equity graph, the project information is matched successfully, and the association relationship between the current project information and the current node is established. At the same time, the subsequent key field matching of 001B, 001C, etc. is terminated. Switch to the traversal matching of the next project 002.

[0116] S322: If the key field is not the same as the information of all nodes in the equity graph, the project information matching fails.

[0117] In steps S321-S322, when any key field in the traversal order successfully matches any node of the equity graph, the traversal matching of the current project information is terminated; and the traversal matching of the next project information is executed.

[0118] S330: Stop after all project information is matched.

[0119] Through this design, all past business data in the local database will be matched when the association relationship is matched. The number of business data items is comprehensive and the time span is long, and both active and passive management items will be covered.

[0120] In another preferred implementation, when the key fields are divided into regular fields and manual judgment fields, in one example, the matching relationships that require manual judgment include "group" / "actual controller" / "project guarantor".

[0121] S320: Using the key fields of each of the project information as query statements one by one, traversing all nodes matching the equity graph, specifically including:

[0122] S321: If the general field is the same as the information of any node of the equity graph, the project information is successfully matched, and an association relationship between the current project information and the current node is established;

[0123] S322: If the manual judgment field is the same as the information of any node of the equity graph, it is identified as a matching relationship that requires manual judgment, and the user is notified to intervene manually; when the user instruction matches successfully, the project information matches successfully, and the association relationship between the current project information and the current node is established; when the user instruction matches unsuccessfully, the manual judgment field matches unsuccessfully, and the traversal matching of the next key field is performed;

[0124] S323: If the manual judgment field of the regular field is not consistent with the information of all nodes of the equity map, the project information matching is unsuccessful.

[0125] Among them, the business data relationship map is established, and the specific example process is as follows:

[0126] (1) All items in the database (thousands) need to be Figure 3 Match the equity map of the target enterprise;

[0127] When matching a single project (for example, project 001), first traverse to see whether the direct counterparty 001A can be matched successfully, that is, whether the enterprise code of 001A = a certain enterprise code in {F, H1, H2, H3…H5, I1, I2…I8, J1…J8…} is established; if 001A can be matched successfully, there is no need to use the "Group / Actual Controller / Project Guarantor" field of project 001 for matching.

[0128] If the direct counterparty 002A of project 002 fails to match successfully, but the company code of the group 002B to which it belongs = one of the company codes in {F, H1, H2, H3…H5, I1, I2…I8, J1…J8…}, there is no need to use the "actual controller / project guarantor" field of project 002 for matching.

[0129] If the direct counterparty 003A of project 003 and its group 003B cannot be traversed to obtain the same enterprise code, but the enterprise code of the actual controller 003C = one of the enterprise codes in {F, H1, H2, H3…H5, I1, I2…I8, J1…J8…}, there is no need to use the "Project Guarantor" field of project 003 for matching.

[0130] If the direct counterparty 004A of project 004, the group 004B to which it belongs, and the actual controller 004C cannot be traversed to obtain equal enterprise codes, then the enterprise code of the project guarantor 004D = one of the enterprise codes in {F, H1, H2, H3…H5, I1, I2…I8, J1…J8…} will be used for matching.

[0131] Assume that by traversing the items 001 to 017 in the above example, the final result is:

[0132] 001A enterprise code = H1 enterprise code (direct counterparty matching);

[0133] 002B enterprise code = H2 enterprise code (matching by the group to which it belongs);

[0134] 003C enterprise code = H3 enterprise code (matching of actual controller);

[0135] 004D enterprise code = I1 enterprise code (matching of project guarantor);

[0136] 005A Enterprise Code = I2 Enterprise Code (direct counterparty matching);

[0137] 006B Enterprise Code = H4 Enterprise Code (matching by the group to which it belongs);

[0138] 007C Enterprise Code = H5 Enterprise Code (matching of actual controller);

[0139] 008D enterprise code = I7 enterprise code (matching of project guarantor);

[0140] Among them, even if the enterprise codes of "Group" / "Actual Controller" / "Project Guarantor" are equal, it is still necessary to manually judge whether these three associations are matched successfully. The manual judgment interface is as follows: Figure 5 The identified relationship graph and Figure 6Identify the relationship list. Enterprises that require users to manually determine the relationship are Figure 5 The relationship map is marked with colors. Users can select a single option in the "Is the match successful" column in the identification relationship list. Checking "Yes" means the match is successful. In this example, it is assumed that the user checked "Yes" for items 002 to 004 and "No" for the others.

[0141] like Figure 7 As shown, the established business data relationship map is composed of the equity map of external enterprises + the successfully matched group business data; when the mouse hovers over the marked node, the text describing the relationship is displayed.

[0142] S400: Perform risk assessment on the target enterprise information, the business data relationship map, the business data and historical enterprise risk / project risk data to obtain a risk assessment result for the target enterprise.

[0143] In the present invention, risk assessment can be performed by inputting relevant data into a risk assessment model to obtain a risk assessment result of the target enterprise.

[0144] In one example, the enterprise risk warning system of the present invention obtains the downward equity relationship map of the group to which the target enterprise belongs (the upper-level enterprise in the equity relationship), expands the relationship map in multiple dimensions such as business data, and displays the expanded identification map and application map from different angles. It uses the relationship map and business data as input and uses the risk model to output relevant risk indicator values.

[0145] In order to achieve the above functions, the system associates the group's business data with the external enterprise equity relationship map to form an expanded business data relationship map. The business data relationship map, the group's business data, historical enterprise risk data, and historical project risk data are used as input to conduct risk assessment using the risk model.

[0146] Specifically, the target enterprise information, the business data relationship map, the business data and historical enterprise risk / project risk data are subjected to risk assessment to obtain a risk assessment result of the target enterprise, which specifically includes:

[0147] S410: Conduct risk assessment on the target enterprise to obtain the target enterprise's own assessment result.

[0148] In one example, the risk assessment is set with several first assessment dimensions, which include: the corporate background of the target enterprise, corporate development history and current situation, risk information, intellectual property information, operating information, annual report information, financial information, autonomous information, and business transaction data between the target enterprise and the group.

[0149] S420: Obtain the associated enterprises of each node of the business data relationship graph, conduct an impact assessment on each associated enterprise, and obtain the impact risk results of each associated enterprise on the target enterprise.

[0150] In one example, the impact assessment is set with several second assessment dimensions, which include: equity ratio, shareholding amount, hierarchical distance of the business data relationship map, business type and internal business transactions; each second assessment dimension is set with a different risk impact weight coefficient, and the risk impact result is calculated based on the weight coefficient.

[0151] S430: Based on the target enterprise's own assessment results and risk impact results, a risk assessment result of the target enterprise is comprehensively calculated.

[0152] In one example, the risk assessment result is provided with several third assessment dimensions, and the third assessment dimensions include one or more combinations of credit risk, market risk, liquidity risk, operational risk, industry risk, legal / regulatory / policy risk, personnel risk and natural disasters.

[0153] In a specific example, the risk assessment of the present invention can obtain the risk assessment result of the target enterprise by inputting relevant data into the risk assessment model. The input data of the risk assessment includes: the business data relationship map after expansion, the current group (company) business data, and the enterprise risk / project risk data maintained in the past;

[0154] The risk assessment model mainly analyzes the risks of the target enterprise from two aspects:

[0155] (a) The risk assessment results of the risk assessment model on the target enterprise itself;

[0156] (b) The risk assessment model structures the risk impact of external enterprises associated with the target enterprise on the target enterprise. Specifically, the risk impact of the associated companies on the business data relationship map on the target enterprise can be determined by analyzing the associated companies associated with the target enterprise (the group (the upper-level enterprise of the equity) and its subordinate companies to which the target enterprise belongs) according to the equity ratio, the hierarchical distance weight in the map, the business type, the internal business transactions, etc.

[0157] Finally, based on the assessment results and risk impact structure of the target enterprise itself, the current risk assessment results of the target enterprise are comprehensively calculated;

[0158] Specifically, the target enterprise's own risk assessment includes enterprise assessment and project assessment. When conducting risk assessment on the target enterprise itself, the following are considered: the target enterprise's corporate background, corporate development history and current situation, risk information (debtor information, dishonesty information, legal proceedings, administrative penalties, etc.), intellectual property information, business information (bidding, bond information, tax rating, qualification certificates, etc.), annual report information, financial information, independent information (business labels, product services, etc.); and the target enterprise's business dealings data with the group (the company).

[0159] Specifically, when the risk assessment model evaluates the risk impact of affiliated companies on the business data relationship map on the target enterprise, it can set the equity ratio and amount of shares, the hierarchical distance in the business data relationship map, the business type, and the internal business transactions. Different risk impact weight coefficients are obtained based on the above factors, and the risk impact results are calculated based on the weight coefficients.

[0160] Furthermore, when designing a risk assessment model, in order to analyze and evaluate the access business transactions, relevant business fields can be extracted from the business data. The relevant business fields include: business type, business amount, business balance, quick cash value of collateral, risk exposure, amount recovered, business start date, risk / concern date and other basic business fields. After extracting these basic business fields, weighted operations are performed on the basic business fields to obtain different types of risk indicators, which are divided into risk performance indicators, key risk indicators, management indicators, etc.; the extracted business fields + risk indicators constitute the basic data for evaluating business transactions.

[0161] It should be noted that the purpose of the present invention is to provide an enterprise analysis and early warning strategy based on a relationship map, and the design of the relevant risk assessment model is designed and specified by those skilled in the art according to actual business needs. For example, a risk assessment model for the target enterprise can be constructed based on the enterprise characteristic data, relationship map, business data, and other external data; the risk assessment result of the target enterprise can be determined using the risk assessment model.

[0162] Among them, the risk assessment model is used to assess the risk situation of the target enterprise. According to specific needs, the target model can be selected to train the risk assessment model. The target model can be a decision tree model, a logistic regression model, an SVM (support vector machines) model, etc. The invention does not make any relevant limitations on the risk assessment model, and those skilled in the art can implement the technical solution of the present invention based on the above records.

[0163] S500: Execute relevant warning operations based on the risk assessment results.

[0164] In one implementation, the risk assessment results can be displayed through a risk dashboard. The risk assessment results are a combination of qualitative analysis and quantitative analysis. Therefore, when displaying the risk assessment results, some assessment dimensions are displayed quantitatively and some assessment dimensions are displayed in text descriptions. Users can download the complete risk assessment result report.

[0165] On the other hand, users can configure thresholds for relevant indicators in advance, and display the risk assessment results in the risk dashboard to issue early warnings. In a specific implementation, when assessing the credit risk of partner company B, the system uses the overdue amount of loans from other banks. Company B has a holding group company A, a subsidiary company C, and a grandson company D.

[0166] Set the hierarchical distance weights to a1 = 0.3 for the first level and a2 = 0.2 for the second level. Assuming that the overdue amount of loans of Company A is 10 million, the overdue amount of loans of Company B is 5 million, the overdue amount of loans of Company C is 3 million, and the overdue amount of loans of Company D is 2 million, then the overdue amount risk index of loans of Company B = 10 million * 0.3 + 5 million + 3 million * 0.3 + 2 million * 0.2 = 9.3 million. Assuming that the credit risk threshold is 8 million, the risk assessment result of Company B triggers a risk warning, reminding or notifying the user.

[0167] Example 3

[0168] The execution subject of this embodiment 3 may be the enterprise risk warning system based on the relationship graph in embodiment 1. Specifically, the enterprise risk warning method of this embodiment 3 is exactly the same as that of embodiment 2. This embodiment 3 may also be executed and implemented by the enterprise risk warning system in the form of software or hardware.

[0169] The enterprise risk early warning method of the third embodiment further expands the application of the business data relationship map and further includes the following steps:

[0170] S400: Generate an application graph and an association relationship description list according to the business data relationship graph.

[0171] Among them, Figure 8 As shown, the application map is a tree map with several levels, the top level of the tree map is the target enterprise or an enterprise or individual associated with the target enterprise, and the lower levels of the tree map are associated business data information;

[0172] like Fig. 9 As shown, the association relationship description list displays the data association relationship information of the business data and the equity map that are successfully matched in a tabular form.

[0173] In one example, the application map is the result extracted from the identification map, which is more concise and intuitive. The application map is divided into only two layers, the upper layer is the search for the "suspected actual controller" F of company A, and the lower layer is multiple direct trading counterparties in the system project library. The connection between the two layers uses text to identify the matching relationship category and details.

[0174] In S400: the target enterprise information, the business data relationship map, the application map, the business data and the historical enterprise risk / project risk data are subjected to risk assessment to obtain a risk assessment result of the target enterprise.

[0175] In a specific example, the risk assessment of the present invention can obtain the risk assessment result of the target enterprise by inputting relevant data into the risk assessment model. The input data of the risk assessment includes: the expanded business data relationship map, application map, current group (company) business data, and previously maintained enterprise risk / project risk data.

[0176] For other structures of the enterprise risk early warning method and system based on relationship graph described in this embodiment, refer to the prior art.

[0177] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Therefore, any modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

[0178] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Therefore, any modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. An enterprise risk early warning method based on relationship graph, characterized in that: The following steps are involved: Retrieving the equity map of the target enterprise according to the user request; the user request includes the target enterprise information; Calling business data from a local database, wherein the business data includes business data of several group subsidiaries; Matching the business data with the equity graph to establish a business data relationship graph; Perform risk assessment on the target enterprise information, the business data relationship map, the business data and historical enterprise risk / project risk data to obtain a risk assessment result for the target enterprise; Execute relevant early warning operations based on risk assessment results; The matching of the business data with the equity graph specifically includes the following steps: Extracting all project information contained in the business data, each project information including several key fields; Using the key fields of each of the project information as query statements one by one, traversing and matching all nodes of the equity graph, specifically comprising: performing traversal matching on all nodes of the equity graph according to a set traversal order; wherein the traversal order is composed of multiple key fields of a single project information arranged in a preset order; If the key field is the same as the information of any node in the equity graph, the project information is successfully matched, and the association relationship between the current project information and the current node is established; If the key field is not the same as the information of all nodes in the equity graph, the project information matching is unsuccessful; Stop until all project information is matched; When any key field in the traversal order successfully matches any node of the equity graph, the traversal matching of the current project information is terminated; the traversal order includes the direct transaction counterparties of the project, the group to which the direct transaction counterparties of the project belong, the actual controller of the direct transaction counterparties of the project and the project guarantor arranged in order of precedence; Execute traversal matching of the next item information; Among them, the key fields are divided into regular fields and manual judgment fields; the key fields of each project information are used as query statements one by one to traverse all nodes matching the equity graph, specifically including: S321: If the general field is the same as the information of any node in the equity graph, then the project information matches Success, establish the association between the current project information and the current node; S322: If the manual judgment field is the same as the information of any node of the equity graph, it is identified as a matching relationship that requires manual judgment, and the user is notified to intervene manually; when the user instruction matches successfully, the project information matches successfully, and the association relationship between the current project information and the current node is established; when the user instruction matches unsuccessfully, the manual judgment field matches unsuccessfully, and the traversal matching of the next key field is performed; S323: If the manual judgment field of the regular field is not consistent with the information of all nodes of the equity map, the project information matching is unsuccessful.

2. The enterprise risk early warning method based on relationship graph according to claim 1 is characterized in that: The target enterprise information, the business data relationship map, the business data and historical enterprise risk / project risk data are subjected to risk assessment to obtain a risk assessment result of the target enterprise, specifically including: Conduct risk assessment on the target enterprise and obtain the target enterprise's own assessment results; Obtaining the associated enterprises of each node of the business data relationship graph, conducting an impact assessment on each associated enterprise, and obtaining the impact risk results of each associated enterprise on the target enterprise; The risk assessment result of the target enterprise is comprehensively calculated based on the target enterprise's own assessment result and the risk impact result.

3. The enterprise risk early warning method based on relationship graph according to claim 2 is characterized in that: The risk assessment of the target enterprise specifically includes: The risk assessment is provided with several first assessment dimensions, which include: corporate background of the target enterprise, corporate development history and current situation, risk information, intellectual property information, operating information, annual report information, financial information, independent information, and business transaction data between the target enterprise and the group.

4. The enterprise risk early warning method based on relationship graph according to claim 2 is characterized in that: The impact assessment of each affiliated enterprise specifically includes: The impact assessment is provided with several second assessment dimensions, which include: equity ratio, shareholding amount, hierarchical distance of business data relationship map, business type and internal business transactions; Each second assessment dimension is set with a different risk impact weight coefficient, and the risk impact result is calculated based on the weight coefficient.

5. The enterprise risk early warning method based on relationship graph according to claim 2 is characterized in that: The risk assessment results of the target enterprise are calculated comprehensively, including: The risk assessment result is provided with several third assessment dimensions, and the third assessment dimensions include one or more combinations of credit risk, market risk, liquidity risk, operational risk, industry risk, legal / regulatory / policy risk, personnel risk and natural disasters.

6. An enterprise risk early warning system based on relationship graph, characterized in that: The enterprise risk early warning system includes: A retrieval module, which is used to retrieve the equity map of the target enterprise according to the user request; the user request includes the target enterprise information; A calling module, which is used to call the business data of the local database, wherein the business data includes the business data of several group subsidiaries; A matching module, which is used to match the business data with the equity graph to establish a business data relationship graph; An assessment module, which is used to conduct risk assessment on the target enterprise information, the business data relationship map, the business data and historical enterprise risk / project risk data to obtain a risk assessment result of the target enterprise; An early warning module is used to perform relevant early warning operations according to the risk assessment results; The matching of the business data with the equity graph specifically includes the following steps: Extracting all project information contained in the business data, each project information including several key fields; Using the key fields of each of the project information as query statements one by one, traversing and matching all nodes of the equity graph, specifically comprising: performing traversal matching on all nodes of the equity graph according to a set traversal order; wherein the traversal order is composed of multiple key fields of a single project information arranged in a preset order; If the key field is the same as the information of any node in the equity graph, the project information is successfully matched, and the association relationship between the current project information and the current node is established; If the key field is not the same as the information of all nodes in the equity graph, the project information matching is unsuccessful; Stop until all project information is matched; When any key field in the traversal order successfully matches any node of the equity graph, the traversal matching of the current project information is terminated; the traversal order includes the direct transaction counterparties of the project, the group to which the direct transaction counterparties of the project belong, the actual controller of the direct transaction counterparties of the project and the project guarantor arranged in order of precedence; Execute traversal matching of the next item information; Among them, the key fields are divided into regular fields and manual judgment fields; the key fields of each project information are used as query statements one by one to traverse all nodes matching the equity graph, specifically including: S321: If the general field is the same as the information of any node in the equity graph, then the project information matches Success, establish the association between the current project information and the current node; S322: If the manual judgment field is the same as the information of any node of the equity graph, it is identified as a matching relationship that requires manual judgment, and the user is notified to intervene manually; when the user instruction matches successfully, the project information matches successfully, and the association relationship between the current project information and the current node is established; when the user instruction matches unsuccessfully, the manual judgment field matches unsuccessfully, and the traversal matching of the next key field is performed; S323: If the manual judgment field of the regular field is not consistent with the information of all nodes of the equity map, the project information matching is unsuccessful.

Images