Methods, apparatus, equipment and computer-readable media for identifying enterprise relationships

Abstract

The present invention discloses a method, device, equipment and computer-readable medium for identifying corporate relationships, and relates to the field of computer technology. A specific implementation of the method includes: collecting basic association information of an enterprise to determine the explicit relationship members of the enterprise; based on the relationship between the holding members in the enterprise, obtaining the implicit relationship members of the enterprise; among the candidate members related to the enterprise, according to the preset expert rules, obtaining the expert relationship members of the enterprise; using a preset recognition model to identify the model relationship members of the enterprise among the potential corporate relationship members, the preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members; merging the explicit relationship members, the implicit relationship members, the expert relationship members and the model relationship members to obtain the relationship members of the enterprise. This implementation can improve the accuracy and work efficiency of identifying corporate relationships.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and computer-readable medium for identifying enterprise relationships. Background Art

[0002] In the financial industry, especially in the banking industry, the degree of identification of corporate relationships reflects, to a certain extent, the bank's ability to operate and manage risks, and is an important manifestation of the comprehensive competitiveness of the business.

[0003] At present, traditional corporate relationships are mainly identified and maintained manually, which largely depends on the personnel's own level of cognition and experience, and also requires a lot of energy to collect information about customers and groups. With the continuous development of society, the business and social activities between enterprises and individuals are becoming increasingly rich, and the relationships are becoming more and more complex and obscure.

[0004] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art: the error rate of manual identification of corporate relationships is high and the work efficiency is low. Summary of the invention

[0005] In view of this, embodiments of the present invention provide a method, apparatus, device and computer-readable medium for identifying enterprise relationships, which can improve the accuracy and work efficiency of identifying enterprise relationships.

[0006] To achieve the above objective, according to one aspect of an embodiment of the present invention, a method for identifying enterprise relationships is provided, comprising:

[0007] Collect basic related information of the enterprise and determine the explicit related members of the enterprise;

[0008] Based on the relationship between the holding members of the enterprise, the implicit relationship members of the enterprise are obtained;

[0009] Among the candidate members related to the enterprise, according to the preset expert rules, obtaining the expert relationship members of the enterprise;

[0010] Using a preset recognition model to identify model relationship members of the enterprise among potential enterprise relationship members, the preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members;

[0011] The explicit relationship members, the implicit relationship members, the expert relationship members and the model relationship members are combined to obtain the relationship members of the enterprise.

[0012] The basic related information of the enterprise includes one or more of the following: branches, subsidiaries, foreign investments, actual controllers, shareholders, persons acting in concert, senior executives, legal persons and family relationships, where the family relationships are family relationships among actual controllers, shareholders, persons acting in concert, senior executives and legal persons.

[0013] The collecting of basic association information of the enterprise and determining the explicit relationship members of the enterprise include:

[0014] Collect basic related information of enterprises, and screen out the first-order relationships, second-order relationships, and third-order relationships of the enterprises;

[0015] The explicit relationship members of the enterprise are determined according to the first-order relationship of the enterprise, the second-order relationship of the enterprise and the third-order relationship of the enterprise.

[0016] The determining the explicit relationship members of the enterprise according to the first-order relationship of the enterprise, the second-order relationship of the enterprise and the third-order relationship of the enterprise includes:

[0017] Establishing a member relationship graph of the enterprise according to the first-order relationship of the enterprise, the second-order relationship of the enterprise and the third-order relationship of the enterprise;

[0018] According to the member relationship diagram of the enterprise, the explicit relationship members of the enterprise are determined.

[0019] The step of obtaining the implicit relationship members of the enterprise based on the relationship between the holding members of the enterprise includes:

[0020] According to the relationship between persons acting in concert among the holding members of the enterprise, an undirected graph is constructed;

[0021] In combination with the equity ratio of the persons acting in concert, a directed graph with weights is established on the basis of the undirected graph;

[0022] According to the weighted directed graph, the implicit relationship members of the enterprise are obtained.

[0023] It is characterized in that the controlling member in the enterprise is a member that has a controlling relationship with the enterprise.

[0024] The persons acting in concert are enterprises or individuals that meet any of the following relationships: kinship, member enterprises of the same group, enterprises with the same directors, supervisors or senior managers, enterprises that directly or indirectly hold shares in the same enterprise, enterprises controlled by the same person and enterprises that hold shares in the same enterprise.

[0025] The above-mentioned combined with the equity ratio of the persons acting in concert, based on the above-mentioned undirected graph, establishes a directed graph with weights, including:

[0026] Calculate the weight of the edge in the undirected graph based on the equity ratio of the persons acting in concert;

[0027] According to the weights of the edges in the undirected graph, multiple edges in the undirected graph are merged to obtain the weighted directed graph.

[0028] The step of merging the multiple edges in the undirected graph according to the weights of the edges in the undirected graph comprises:

[0029] Merge multiple edges in the undirected graph according to the maximum value of the edge weights in the undirected graph.

[0030] The step of obtaining implicit relationship members of the enterprise according to the weighted directed graph includes:

[0031] Starting from the persons acting in concert, the sum of the equity of the enterprise nodes is calculated according to the directed graph with weights;

[0032] Based on the sum of the equity of the enterprise nodes, the implicit relationship members of the enterprise are obtained.

[0033] The preset expert rules include the expert rules set by the CBRC.

[0034] The candidate members related to the enterprise include enterprises or individuals having relationships with the enterprise in terms of personnel, finance and business.

[0035] Before the method of using the preset identification model to identify the model relationship members of the enterprise among the potential enterprise relationship members, the method further includes:

[0036] The explicit relationship members, the implicit relationship members and the expert relationship members are taken as positive samples, and the relationship members who do not belong to the enterprise are taken as negative samples. The preset recognition model is trained with the positive samples and the negative samples.

[0037] The ratio of the positive samples to the negative samples is between 1 / 20 and 1 / 5.

[0038] The potential corporate relationship members include enterprises that have one or more of the following relationships with the enterprise: a relationship of large-scale operating transactions, a relationship of large-scale non-operating capital occupation and / or being occupied, a relationship of large-scale financing guarantees, being guaranteed or having mutual guarantees, and a relationship of transferring assets and profits at unfair prices and conditions.

[0039] The preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members in combination with the GBDT algorithm.

[0040] The preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members in combination with the GBDT algorithm and the community discovery algorithm.

[0041] According to a second aspect of an embodiment of the present invention, there is provided a device for identifying enterprise relationships, comprising:

[0042] The explicit module is used to collect basic related information of the enterprise and determine the explicit relationship members of the enterprise;

[0043] An implicit module, used for obtaining implicit relationship members of the enterprise based on the relationship between the holding members of the enterprise;

[0044] An expert module, used to obtain expert relationship members of the enterprise from candidate members related to the enterprise according to preset expert rules;

[0045] An identification module, used to identify model relationship members of the enterprise among potential enterprise relationship members by using a preset identification model, wherein the preset identification model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members;

[0046] A merging module is used to merge the explicit relationship members, the implicit relationship members, the expert relationship members and the model relationship members to obtain the relationship members of the enterprise.

[0047] According to a third aspect of an embodiment of the present invention, there is provided an electronic device for identifying enterprise relationships, including:

[0048] one or more processors;

[0049] a storage device for storing one or more programs,

[0050] When the one or more programs are executed by the one or more processors, the one or more processors implement the above method.

[0051] According to a fourth aspect of an embodiment of the present invention, there is provided a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the method as described above is implemented.

[0052] One embodiment of the above invention has the following advantages or beneficial effects: collecting basic association information of an enterprise to determine the explicit relationship members of the enterprise; based on the relationship between the controlling members of the enterprise, obtaining the implicit relationship members of the enterprise; obtaining the expert relationship members of the enterprise in the candidate enterprises related to the enterprise according to the preset expert rules; using a preset recognition model to identify the model relationship members of the enterprise among the potential enterprise relationship members, the preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members; merging the explicit relationship members, the implicit relationship members, the expert relationship members and the model relationship members to obtain the relationship members of the enterprise. Since the relationship members of the enterprise can be automatically identified from multiple aspects, the error rate of identifying enterprise relationships can be reduced while improving work efficiency.

[0053] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0055] Figure 1 is a schematic diagram of the main process of the method for identifying enterprise relationships according to an embodiment of the present invention;

[0056] Figure 2 is a schematic diagram of a process of determining explicit relationship members of an enterprise according to an embodiment of the present invention;

[0057] Figure 3 is a schematic diagram of a process of determining explicit relationship members of an enterprise according to a relationship diagram according to an embodiment of the present invention;

[0058] Figure 4 is a schematic diagram of a member relationship diagram of enterprise A according to an embodiment of the present invention;

[0059] Figure 5 is a schematic diagram of persons acting in concert according to an embodiment of the present invention;

[0060] Figure 6 is a schematic diagram of a process of obtaining implicit relationship members of an enterprise according to an embodiment of the present invention;

[0061] Figure 7 is a schematic diagram of a process of establishing a directed graph with weights according to an embodiment of the present invention;

[0062] Figure 8 It is a schematic diagram of a process of obtaining implicit relationship members of an enterprise according to a directed graph with weights according to an embodiment of the present invention;

[0063] Fig. 9 is a schematic diagram of the main structure of an apparatus for identifying enterprise relationships according to an embodiment of the present invention;

[0064] Fig.10 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0065] Fig.11 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0067] The member companies of a group have cross-regional and cross-industry characteristics. With the continuous development of the market economy, it is difficult to unilaterally investigate the relationship between the member companies of a group. At the same time, the connection and interaction between enterprises are becoming more and more complex. The outbreak of risks of a single customer may trigger risks at the level of the entire group, and in serious cases, it may also trigger regional financial risks. The business behavior of customers and the risks that may be brought to banks are becoming more and more obscure and more and more difficult to locate. Therefore, how to quickly and accurately dig out the member companies of the group; how to do a good job of risk management for customers efficiently and with high quality will be a severe challenge.

[0068] Combined with the definition of group customers in relevant documents, the group is mainly determined from the perspective of control rights, which can be reflected in the following two aspects. On the one hand, it is the direct control relationship between shareholders, actual controllers, subsidiaries and branches; on the other hand, due to the existence of certain hidden control relationships, there are economic transaction characteristics such as large-scale capital occupation and transfer of assets not at fair value.

[0069] However, in the actual operation process, it was found that many hidden potential groups were not identified, resulting in the group size in the actual group relationship being inconsistent with the actual situation, which affected the risk management work to a certain extent.

[0070] Traditional manual identification methods have major limitations such as information asymmetry, small identification scale, and incomplete consideration. Therefore, the error rate of manual identification of corporate relationships is high and the work efficiency is low.

[0071] In order to solve the technical problems of high error rate and low work efficiency in manual identification of enterprise relationships, the following technical solutions in the embodiments of the present invention may be adopted.

[0072] See also Figure 1 , Figure 1 FIG. 1 is a schematic diagram of the main process of the method for identifying enterprise relationships according to an embodiment of the present invention, which determines the relationship members of an enterprise from multiple aspects. Figure 1 As shown, the specific steps include:

[0073] S101. Collect basic related information of the enterprise and determine the explicit relationship members of the enterprise.

[0074] In the embodiment of the present invention, considering that the amount of data of the basic associated information of an enterprise is relatively large, one or more servers may be used to collect the data respectively.

[0075] The basic related information of an enterprise includes one or more of the following: branches, subsidiaries, outbound investments, actual controllers, shareholders, persons acting in concert, senior executives, legal persons and family relationships. Among them, family relationships refer to the family relationships among actual controllers, shareholders, persons acting in concert, senior executives and legal persons.

[0076] In the embodiment of the present invention, the relationship members determined based on the basic association information of the enterprise are called explicit relationship members of the enterprise.

[0077] See also Figure 2 , Figure 2 1 is a flow chart of determining explicit relationship members of an enterprise according to an embodiment of the present invention, which specifically includes the following steps:

[0078] S201. Collect basic related information of enterprises, and screen out first-order relationships, second-order relationships, and third-order relationships of enterprises.

[0079] In the embodiment of the present invention, according to the relationship between entities that reflect the control rights of enterprises, enterprise pairs are formed through the derivation of multi-level relationships, and enterprises form groups through control rights relationships. Based on the existing group's subordinate enterprises, more hidden potential connections between enterprises are discovered.

[0080] The relationships between enterprises include first-order relationships, second-order relationships and third-order relationships.

[0081] The first-order relationship of an enterprise refers to one or more of the following relationships: branches, subsidiaries, foreign investments, actual controllers, shareholders and persons acting in concert. Among them, persons acting in concert refer to persons who, through agreements or other arrangements, jointly increase the number of voting rights of a listed company's shares that they can control with other investors.

[0082] The second-order relationship of enterprises refers to having the same actual controller, such as: the current / former senior executives, legal persons or shareholders of companies A and B overlap.

[0083] The third-order relationship of an enterprise refers to the relationship of employment, relatives and implicit relationships. Among them, implicit relationships refer to the objectively existing and unknown relationships between enterprises or personnel. For example, the third-order relationship of an enterprise includes the relationship between spouses, siblings or parents of children in two companies. The positions can be actual controllers, legal persons, directors, supervisors, senior executives or shareholders.

[0084] S202. Determine the explicit relationship members of the enterprise based on the first-order relationship, the second-order relationship and the third-order relationship of the enterprise.

[0085] In the embodiment of the present invention, members having first-order, second-order, and third-order relationships with an enterprise are determined as explicit relationship members of the enterprise. That is, if an enterprise has a first-order, second-order, or third-order relationship with another enterprise, the enterprise is determined to be an explicit relationship member of the other enterprise.

[0086] It can be understood that the explicit relationship members of an enterprise can be obtained from the first-order relationship members of the enterprise, the second-order relationship members of the enterprise, or the third-order relationship members of the enterprise.

[0087] In one embodiment of the present invention, in order to increase the speed of determining the explicit relationship members of an enterprise, this can be achieved through a member relationship graph.

[0088] See also Figure 3 , Figure 3 The present invention is a flowchart of determining explicit relationship members of an enterprise according to a relationship diagram according to an embodiment of the present invention, which specifically includes the following steps:

[0089] S301. Establish a member relationship diagram of the enterprise based on the first-order relationship, the second-order relationship and the third-order relationship of the enterprise.

[0090] The first-order relationship, second-order relationship and third-order relationship of an enterprise are all relationships with the enterprise. Then, with the enterprise as the center, a member relationship diagram of the enterprise is established.

[0091] See also Figure 4 , Figure 4 It is a schematic diagram of a member relationship diagram of enterprise A according to an embodiment of the present invention. Figure 4 In the figure, Enterprise A is the parent company, and the member relationship diagram of Enterprise A is displayed.

[0092] Taking enterprise A as the center, Figure 4 There are three circles in the figure. The members in the circle with the smallest radius are the first-order members of enterprise A; the members between the circle with the smallest radius and the circle with the middle radius are the second-order members of enterprise A; the members between the circle with the middle radius and the circle with the largest radius are the third-order members of enterprise A. The lines between enterprise A and the members indicate the proportion of shares held.

[0093] S302. Determine the explicit relationship members of the enterprise according to the member relationship diagram of the enterprise.

[0094] The relationship between the enterprise and its members is clearly marked in the enterprise's member relationship diagram, and then the enterprise's displayed relationship members can be determined according to the enterprise's member relationship diagram.

[0095] exist Figure 3 In the embodiment, the member relationship diagram of the enterprise is used to determine the explicit relationship members of the enterprise, which can improve the speed of determination.

[0096] S102. Based on the relationship between the holding members in the enterprise, the implicit relationship members of the enterprise are obtained.

[0097] The holding between enterprises is divided into direct holding and indirect holding. Enterprises hold other enterprises through indirect holding through multi-level relationships or family relationships.

[0098] In the embodiment of the present invention, based on the relationship between the holding members in the enterprise, the implicit relationship members of the enterprise are obtained. The holding members in the enterprise are members that have a holding relationship with the enterprise.

[0099] Specifically, first, the persons acting in concert need to be known. In the embodiment of the present invention, persons acting in concert are enterprises or persons that meet any of the following relationships: kinship, member enterprises in the same group, enterprises with the same directors, supervisors or senior managers, enterprises that directly or indirectly hold shares in the same enterprise, enterprises controlled by the same person, and enterprises that hold shares in the same enterprise.

[0100] That is to say, persons who meet one of the following conditions are considered persons acting in concert.

[0101] Condition 1: Natural person entities with a family relationship are persons acting in concert.

[0102] Condition 2: Member companies in the same group are persons acting in concert

[0103] Condition 3: Meet Figure 6 A and B are persons acting in concert.

[0104] See also Figure 5 , Figure 5 is a schematic diagram of persons acting in concert according to an embodiment of the present invention. Figure 5 The first row on the left shows that the same director, supervisor or senior manager of two companies is a person acting in concert. That is, individual C is a director, supervisor or senior manager of company A, and is also a director, supervisor or senior manager of company B. Then companies A and B are persons acting in concert.

[0105] Figure 5The first row in the figure shows that enterprises directly or indirectly holding shares in the same enterprise are persons acting in concert. That is, if enterprise A and enterprise B are direct holding enterprises of enterprise C, then enterprise A and enterprise B are persons acting in concert.

[0106] Figure 5 The first row on the right shows that enterprises directly or indirectly held by the same enterprise are persons acting in concert. That is, Enterprise A and Enterprise B are directly held by Enterprise C. At the same time, Enterprise D holds a controlling stake in Enterprise A and Enterprise B, so Enterprise A and Enterprise B are persons acting in concert.

[0107] Figure 5 The left picture in the second row shows companies controlled by the same person or companies held by the same company. That is, if individual C is a director, supervisor or senior manager of company A, and individual C controls company B's shareholding in company A, then companies A and B are persons acting in concert.

[0108] Figure 5 The second row of the figure shows that the companies hold shares in the same company, and the companies or individuals held are persons acting in concert. That is, individual A holds shares in company C, company B holds shares in company C, and individual A holds 30% of company B, then individual A and company B are persons acting in concert.

[0109] Figure 5 The second row on the right shows companies holding shares in the same company, and the held companies are persons acting in concert. That is, if company A holds shares in company C, and company B holds shares in company C, and there is an economic interest relationship between company A and company B, then company A and company B are persons acting in concert.

[0110] See also Figure 6 , Figure 6 The flowchart of obtaining implicit relationship members of an enterprise according to an embodiment of the present invention specifically includes the following steps:

[0111] S601. Construct an undirected graph based on the relationships among the persons acting in concert among the holding companies.

[0112] The implicit members of an enterprise generally hold shares in the enterprise indirectly. The controlling members of an enterprise include one or more persons acting in concert. By analyzing the relationship between the persons acting in concert and the enterprise, the implicit members of the enterprise can be known.

[0113] Specifically, according to the relationship between the persons acting in concert among the holding members of the enterprise, the connection between the persons acting in concert and the enterprise is established. An undirected graph is constructed based on the connection between multiple persons acting in concert and the enterprise.

[0114] S602. Based on the undirected graph, a directed graph with weights is established in combination with the equity ratios of persons acting in concert.

[0115] In the constructed undirected graph, the connection between the enterprise and the persons acting in concert has no direction and no weight. However, considering that the implicit relationship members of the enterprise need to be known through the equity relationship between the enterprise and the persons acting in concert, a directed graph with weights can be established based on the undirected graph.

[0116] See also Figure 7 , Figure 7 The present invention is a flowchart of establishing a directed graph with weights according to an embodiment of the present invention, which specifically includes the following steps:

[0117] S701. Calculate the weights of the edges in the undirected graph based on the equity ratios of persons acting in concert.

[0118] Find all connected components in the undirected graph as the entity combination of concerted actors. As an example, find all connected components in the undirected graph by traversing the graph or using a union-find.

[0119] Based on all connected components of the undirected graph, the edge weight represents the equity ratio or investment proportion. For the actual controlling equity, the edge weight is 1. Add a bidirectional relationship between two entities in the concerted action combination, and the edge weight is 1. Entities are nodes in an undirected graph.

[0120] S702: Merge multiple edges in the undirected graph according to the weights of the edges in the undirected graph to obtain a weighted directed graph.

[0121] In an undirected graph, two entities are connected by an edge, and the number of edges is one or more. In order to know the equity relationship between two entities, multiple edges in the undirected graph can be merged. That is, according to the weights of the edges in the undirected graph, multiple edges in the undirected graph are merged to obtain a weighted directed graph.

[0122] In one embodiment of the present invention, in order to improve the accuracy of the directed graph, when there are multiple edges between two entities, the multiple edges in the undirected graph are merged according to the maximum value among the weights of the multiple edges in the undirected graph.

[0123] The weighted directed graph can reflect the control path and decision-making path, and includes actual control, shareholders, equity, personal investment and corporate external investment relations.

[0124] exist Figure 7 In the embodiment, a directed graph with weights is established in combination with the equity ratio of persons acting in concert, laying the foundation for knowing the implicit relationship members of the enterprise.

[0125] S603: Obtain the implicit relationship members of the enterprise according to the directed graph with weights.

[0126] A weighted directed graph includes multiple nodes, each with a weight, and thus the implicit relationship members of the enterprise can be known.

[0127] See also Figure 8 , Figure 8 The present invention is a flowchart of obtaining implicit relationship members of an enterprise according to a directed graph with weights according to an embodiment of the present invention, which specifically includes:

[0128] S801. Starting from the persons acting in concert, calculate the sum of the equity of the enterprise nodes according to the directed graph with weights.

[0129] In a weighted directed graph, starting from any node among the persons acting in concert, calculate the consolidated equity of the person acting in concert in the enterprise node.

[0130] Then, all paths from this node to each enterprise node are calculated. In the above paths, equity is accumulated according to point multiplication, and the combined equity of the concerted actor in each enterprise node is the sum of the equity of all enterprise nodes in the path.

[0131] S802. Obtain the implicit relationship members of the enterprise based on the sum of the equity of the enterprise nodes.

[0132] The indirect equity of a node in a company is the combined equity of the person acting in concert to which the node belongs. Then, based on the sum of the equity of the enterprise nodes and the above-mentioned method of determining indirect equity, the enterprise or individual with indirect equity in the enterprise is regarded as the implicit relationship member of the enterprise.

[0133] exist Figure 8 In the embodiment, implicit relationship members of the enterprise are obtained based on a directed graph with weights.

[0134] exist Figure 6 In the embodiment, the implicit relationship members of the enterprise can be timely and accurately known through the directed value with weight.

[0135] Adopting the technical solution in S102 can improve the accuracy of determining the relationship members of the enterprise. Taking a certain company as an example, the determination of relationship members by using the existing technology and the determination of relationship members by using the solution in S102 are analyzed.

[0136] By manually identifying the enterprise relationship, it is determined that the enterprise has 106 relationship members. By adopting the solution in S102, it is determined that the enterprise has 113 relationship members. It can be seen that the effect of adopting the solution in S102 is more obvious.

[0137] S103. Among the candidate members related to the enterprise, according to the preset expert rules, the expert relationship members of the enterprise are obtained.

[0138] In the embodiment of the present invention, the expert relationship members of the enterprise can also be known according to the preset expert rules. The candidate members related to the enterprise include enterprises or personnel that have a relationship with the enterprise in terms of personnel, finance and business.

[0139] In an embodiment of the present invention, the preset expert rules include expert rules set by the China Banking Regulatory Commission. As an example, the preset expert rules include the following three types of rules:

[0140] First expert rule: Candidate members related to the enterprise must be shareholders of the enterprise and their shareholding is greater than 50%.

[0141] Second expert rule: Among the candidate members related to the enterprise, he / she must be the largest shareholder of the enterprise and his / her shareholding is greater than 25%.

[0142] The third expert rule: Among the candidate members related to the enterprise, the total investment in the enterprise accounts for more than 50% of the registered capital.

[0143] In one embodiment of the present invention, the preset expert rules may include one or more of the following:

[0144] An enterprise controlled by the group's parent company.

[0145] Enterprises controlled by the top 10 natural person shareholders of the group's parent company.

[0146] Directors, supervisors or senior executives of the group's parent company hold controlling stakes in external enterprises.

[0147] An enterprise that is controlled by the group's parent company or after a merger of equity interests.

[0148] A company that can hold a controlling stake after the group's parent company and its top 10 natural person shareholders combine their equity.

[0149] The parent company of the group and its legal persons, directors, supervisors or senior executives can hold a controlling stake in the company after merging their equity.

[0150] A company that can be controlled by the top 10 natural person shareholders of the group's parent company and the top 10 natural person shareholders, as well as the relatives of the above persons, after combining their equity.

[0151] An enterprise controlled by the legal person of a group of member companies or controlled after the combined equity.

[0152] Enterprises controlled by directors, supervisors or senior executives of group member companies.

[0153] All enterprises of which the actual controller of the group's parent company serves as the legal person.

[0154] S104. Using a preset recognition model, model relationship members of the enterprise are identified among potential enterprise relationship members. The preset recognition model is obtained by training explicit relationship members, implicit relationship members and expert relationship members.

[0155] For the relationship members that cannot be determined through S101, S102 and S103, a preset identification model can be used to further determine the relationship members. The relationship members identified by the preset identification model are called model relationship members of the enterprise.

[0156] The following describes how to build a preset recognition model. The preset recognition model is a machine learning model. Machine learning models need to be trained with data.

[0157] In the embodiment of the present invention, the preset recognition model is obtained by training explicit relationship members, implicit relationship members and expert relationship members. It can be understood that after determining the relationship members in S101, S102 and S103, the determined relationship members are used as training data to train the preset recognition model to meet the coverage of the following types of members.

[0158] Category 1 members: Enterprises that have continuous, large-value business transactions with confirmed members and whose operations and credit status are clearly correlated. That is, enterprises that have large-value business transactions.

[0159] Category II members: Enterprises that have large amounts of non-operating funds occupied and / or occupied by confirmed members, and whose operations and credit status are obviously correlated. In other words, enterprises that have large amounts of non-operating funds occupied and / or occupied by confirmed members.

[0160] The third category of members: enterprises that have large financing guarantees or are guaranteed, or have mutual guarantees with confirmed relationship members, and there is an obvious correlation between their operations and credit status. That is, enterprises that have large financing guarantees, are guaranteed, or have mutual guarantees.

[0161] Category 4 members: Enterprises that have other relationships with confirmed members and transfer assets and profits at unfair prices and conditions. In other words, enterprises that have relationships that transfer assets and profits at unfair prices and conditions.

[0162] Specifically, training the preset recognition model requires positive samples and negative samples. Positive samples are relationship members belonging to the enterprise; negative samples are relationship members that do not belong to the enterprise.

[0163] In the embodiment of the present invention, explicit relationship members, implicit relationship members and expert relationship members are taken as positive samples, and relationship members who do not belong to the enterprise are taken as negative samples. The positive samples and negative samples are used to train a preset recognition model.

[0164] In the process of implementing the embodiments of the present invention, under large-scale data, the number of positive samples is much smaller than the number of negative samples, and the number of positive samples is insufficient. Sample sampling can be used to increase the number of samples, and sampling needs to be performed according to the overall sample distribution.

[0165] After many practices, the recognition effect of the preset recognition model obtained by training is better when the ratio of positive samples to negative samples is between 1 / 20 and 1 / 5. In other words, the minimum ratio of positive samples to negative samples is 1:20, and the maximum ratio of positive samples to negative samples is 1:5.

[0166] After determining the positive samples and negative samples, the GBDT algorithm can be used to train the preset recognition model. That is, the preset recognition model is obtained by combining explicit relationship members, implicit relationship members and expert relationship members with the GBDT algorithm training.

[0167] The GBDT algorithm is a decision tree model based on the idea of ​​integration, and its essence is based on residual learning. Its characteristics are: it can process various types of data; it has a high accuracy rate; and it is robust to outliers.

[0168] In one embodiment of the present invention, in order to improve the speed of training the recognition model, the recognition model can also be trained in combination with a community detection algorithm. The community detection algorithm is used to discover the community structure in the network and is a clustering algorithm.

[0169] That is, the preset recognition model is obtained by training explicit relationship members, implicit relationship members and expert relationship members in combination with the GBDT algorithm and the community discovery algorithm.

[0170] Use the GBDT algorithm to train the recognition model, estimate all sample pairs, and get the probability that the samples belong to the same group. Remove the edges with low probability, and then use the modularity-based community discovery algorithm to discover the community.

[0171] Finally, a preset identification model is obtained, and the preset identification model is used to identify the model relationship members of the enterprise among the potential enterprise relationship members.

[0172] S105. Merge explicit relationship members, implicit relationship members, expert relationship members, and model relationship members to obtain the relationship members of the enterprise.

[0173] After determining the explicit relationship members, implicit relationship members, expert relationship members and model relationship members, considering the possibility of duplication of the above-determined relationship members, it is necessary to merge them to obtain the relationship members of the enterprise. It can be understood that the relationship members of the enterprise are composed of explicit relationship members, implicit relationship members, expert relationship members and model relationship members.

[0174] In the above embodiment, the basic association information of the enterprise is collected to determine the explicit relationship members of the enterprise; based on the relationship between the holding members in the enterprise, the implicit relationship members of the enterprise are obtained; among the candidate enterprises related to the enterprise, the expert relationship members of the enterprise are obtained according to the preset expert rules; the model relationship members of the enterprise are identified among the potential enterprise relationship members using the preset recognition model, and the preset recognition model is obtained by training the explicit relationship members, the implicit relationship members, and the expert relationship members; the explicit relationship members, the implicit relationship members, the expert relationship members, and the model relationship members are merged to obtain the relationship members of the enterprise. Since the relationship members of the enterprise can be automatically identified from multiple aspects, the error rate of identifying enterprise relationships can be reduced while improving work efficiency.

[0175] It is understandable that in order to identify corporate relationships, we start from the following aspects.

[0176] The first aspect: identify explicit relationship members.

[0177] By utilizing the direct control relationship between enterprises to conduct equity penetration, based on the members that have been discovered, other related companies controlled by group members are discovered through third-order control derivation.

[0178] The second aspect: identify implicit relationship members.

[0179] For some members who also form enterprise relationships but are not discovered, mining is performed through directed graph calculation.

[0180] The third aspect: determine the model relationship members.

[0181] Using the characteristics of economic transactions and the GBDT algorithm to train a preset recognition model, we can determine whether a pair of companies belong to the same group and conduct mining from the perspective of economic transactions.

[0182] See also Fig. 9 , Fig. 9 is a schematic diagram of the main structure of the device for identifying enterprise relationships according to an embodiment of the present invention. The device for identifying enterprise relationships can implement a method for identifying enterprise relationships, such as Fig. 9 As shown, the device for identifying enterprise relationships specifically includes:

[0183] The explicit module 901 is used to collect basic association information of an enterprise and determine explicit relationship members of the enterprise;

[0184] Implicit module 902, used to obtain implicit relationship members of the enterprise based on the relationship between the holding members of the enterprise;

[0185] An expert module 903 is used to obtain expert relationship members of the enterprise from candidate members related to the enterprise according to preset expert rules;

[0186] The identification module 904 is used to identify the model relationship members of the enterprise among the potential enterprise relationship members by using a preset identification model, wherein the preset identification model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members;

[0187] The merging module 905 is used to merge the explicit relationship members, the implicit relationship members, the expert relationship members and the model relationship members to obtain the relationship members of the enterprise.

[0188] In one embodiment of the present invention, the basic related information of the enterprise includes one or more of the following: branches, subsidiaries, foreign investments, actual controllers, shareholders, persons acting in concert, senior executives, legal persons and kinship relationships, where the kinship relationships are kinship relationships among actual controllers, shareholders, persons acting in concert, senior executives and legal persons.

[0189] In one embodiment of the present invention, the explicit module 901 is specifically used to collect basic association information of an enterprise, and screen out the first-order relationship of the enterprise, the second-order relationship of the enterprise, and the third-order relationship of the enterprise;

[0190] The explicit relationship members of the enterprise are determined according to the first-order relationship of the enterprise, the second-order relationship of the enterprise and the third-order relationship of the enterprise.

[0191] In one embodiment of the present invention, the explicit module 901 is specifically used to establish a member relationship diagram of the enterprise according to the first-order relationship of the enterprise, the second-order relationship of the enterprise and the third-order relationship of the enterprise;

[0192] According to the member relationship diagram of the enterprise, the explicit relationship members of the enterprise are determined.

[0193] In one embodiment of the present invention, the implicit module 902 is specifically used to construct an undirected graph according to the relationship between persons acting in concert among the enterprise holding members;

[0194] In combination with the equity ratio of the persons acting in concert, a directed graph with weights is established on the basis of the undirected graph;

[0195] According to the weighted directed graph, the implicit relationship members of the enterprise are obtained.

[0196] In one embodiment of the present invention, the controlling member of the enterprise is a member having a controlling relationship with the enterprise.

[0197] In one embodiment of the present invention, the persons acting in concert are enterprises or individuals that meet any of the following relationships: kinship, member enterprises in the same group, enterprises with the same directors, supervisors or senior managers, enterprises that directly or indirectly hold shares in the same enterprise, enterprises controlled by the same person and enterprises that hold shares in the same enterprise.

[0198] In one embodiment of the present invention, the implicit module 902 is specifically used to calculate the weight of the edge in the undirected graph based on the equity ratio of the person acting in concert;

[0199] According to the weights of the edges in the undirected graph, multiple edges in the undirected graph are merged to obtain the weighted directed graph.

[0200] In one embodiment of the present invention, the implicit module 902 is specifically configured to merge multiple edges in the undirected graph according to the maximum value of the weights of the edges in the undirected graph.

[0201] In one embodiment of the present invention, the implicit module 902 is specifically used to calculate the sum of the equity of the enterprise nodes starting from the concerted person according to the weighted directed graph;

[0202] Based on the sum of the equity of the enterprise nodes, the implicit relationship members of the enterprise are obtained.

[0203] In one embodiment of the present invention, the preset expert rules include expert rules set by the China Banking Regulatory Commission.

[0204] In one embodiment of the present invention, the candidate members related to the enterprise include enterprises or personnel that have relationships with the enterprise in terms of personnel, finance and business.

[0205] In one embodiment of the present invention, the recognition module 904 is specifically used to use the explicit relationship members, the implicit relationship members and the expert relationship members as positive samples, and the relationship members who do not belong to the enterprise as negative samples, and train the preset recognition model with the positive samples and the negative samples.

[0206] In one embodiment of the present invention, the ratio of the positive samples to the negative samples is between 1 / 20 and 1 / 5.

[0207] In one embodiment of the present invention, the potential corporate relationship members include enterprises that have one or more of the following relationships with the enterprise: a relationship of large-scale operating transactions, a relationship of large-scale non-operating capital occupation and / or being occupied, a relationship of large-scale financing guarantees, being guaranteed or having mutual guarantees, and a relationship of transferring assets and profits at unfair prices and conditions.

[0208] In one embodiment of the present invention, the preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members in combination with the GBDT algorithm.

[0209] In one embodiment of the present invention, the preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members in combination with the GBDT algorithm and the community discovery algorithm.

[0210] Fig.10 An exemplary system architecture 1000 is shown to which the method for identifying enterprise relationships or the apparatus for identifying enterprise relationships according to the embodiments of the present invention can be applied.

[0211] like Fig.10 As shown, the system architecture 1000 may include terminal devices 1001, 1002, 1003, a network 1004, and a server 1005. The network 1004 is used to provide a medium for communication links between the terminal devices 1001, 1002, 1003 and the server 1005. The network 1004 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0212] Users can use terminal devices 1001, 1002, 1003 to interact with server 1005 through network 1004 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 1001, 1002, 1003, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0213] The terminal devices 1001 , 1002 , and 1003 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc.

[0214] The server 1005 may be a server that provides various services, such as a backend management server (only an example) that provides support for shopping websites browsed by users using the terminal devices 1001, 1002, and 1003. The backend management server may analyze and process the received data such as product information query requests, and feed back the processing results (such as target push information, product information - only an example) to the terminal device.

[0215] It should be noted that the method for identifying corporate relationships provided in the embodiment of the present invention is generally executed by the server 1005 , and accordingly, the device for identifying corporate relationships is generally set in the server 1005 .

[0216] It should be understood that Fig.10The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0217] Reference below Fig.11 , which shows a schematic diagram of the structure of a computer system 1100 of a terminal device suitable for implementing an embodiment of the present invention. Fig.11 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0218] like Fig.11 As shown, the computer system 1100 includes a central processing unit (CPU) 1101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1102 or a program loaded from a storage part 1108 into a random access memory (RAM) 1103. In the RAM 1103, various programs and data required for the operation of the system 1100 are also stored. The CPU 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0219] The following components are connected to the I / O interface 1105: an input section 1106 including a keyboard, a mouse, etc.; an output section 1107 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1108 including a hard disk, etc.; and a communication section 1109 including a network interface card such as a LAN card, a modem, etc. The communication section 1109 performs communication processing via a network such as the Internet. A drive 1110 is also connected to the I / O interface 1105 as needed. A removable medium 1111, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1110 as needed, so that a computer program read therefrom is installed into the storage section 1108 as needed.

[0220] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1109, and / or installed from the removable medium 1111. When the computer program is executed by the central processing unit (CPU) 1101, the above-mentioned functions defined in the system of the present invention are executed.

[0221] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0222] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0223] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes an explicit module, an implicit module, an expert module, an identification module and a merging module. The names of these modules do not constitute a limitation on the modules themselves in certain cases, for example, "identification module 904, specifically used to collect basic association information of an enterprise and determine the explicit relationship members of the enterprise".

[0224] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes:

[0225] Collect basic related information of the enterprise and determine the explicit related members of the enterprise;

[0226] Based on the relationship between the holding members of the enterprise, the implicit relationship members of the enterprise are obtained;

[0227] Among the candidate members related to the enterprise, according to the preset expert rules, obtaining the expert relationship members of the enterprise;

[0228] Using a preset recognition model to identify model relationship members of the enterprise among potential enterprise relationship members, the preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members;

[0229] The explicit relationship members, the implicit relationship members, the expert relationship members and the model relationship members are combined to obtain the relationship members of the enterprise.

[0230] According to the technical solution of the embodiment of the present invention, the basic association information of the enterprise is collected to determine the explicit relationship members of the enterprise; based on the relationship between the holding members in the enterprise, the implicit relationship members of the enterprise are obtained; among the candidate enterprises related to the enterprise, the expert relationship members of the enterprise are obtained according to the preset expert rules; the model relationship members of the enterprise are identified among the potential enterprise relationship members using the preset recognition model, and the preset recognition model is obtained by training the explicit relationship members, the implicit relationship members, and the expert relationship members; the explicit relationship members, the implicit relationship members, the expert relationship members, and the model relationship members are merged to obtain the relationship members of the enterprise. Since the relationship members of the enterprise can be automatically identified from multiple aspects, the error rate of identifying enterprise relationships can be reduced while improving work efficiency.

[0231] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for identifying business relationships, characterized in that: include: Collect basic related information of the enterprise and determine the explicit related members of the enterprise; Based on the relationship between the holding members of the enterprise, the implicit relationship members of the enterprise are obtained; Among the candidate members related to the enterprise, according to the preset expert rules, obtaining the expert relationship members of the enterprise; Using a preset recognition model to identify model relationship members of the enterprise among potential enterprise relationship members, the preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members; Merging the explicit relationship members, the implicit relationship members, the expert relationship members and the model relationship members to obtain the relationship members of the enterprise; The step of obtaining the implicit relationship members of the enterprise based on the relationship between the holding members of the enterprise includes: According to the relationship between persons acting in concert among the holding members of the enterprise, an undirected graph is constructed; In combination with the equity ratio of the persons acting in concert, a directed graph with weights is established on the basis of the undirected graph; Obtaining implicit relationship members of the enterprise according to the weighted directed graph; The above-mentioned combined with the equity ratio of the persons acting in concert, based on the above-mentioned undirected graph, establishes a directed graph with weights, including: Calculate the weight of the edge in the undirected graph based on the equity ratio of the persons acting in concert; According to the weights of the edges in the undirected graph, merging multiple edges in the undirected graph to obtain the weighted directed graph; The step of merging the multiple edges in the undirected graph according to the weights of the edges in the undirected graph comprises: Merge multiple edges in the undirected graph according to the maximum value of the weights of the edges in the undirected graph; The preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members in combination with the GBDT algorithm and the community discovery algorithm; The GBDT algorithm is used to train the recognition model, estimate the probability of the samples belonging to the same group, remove the edges with low probability, and then use the modularity-based community discovery algorithm to discover the community.

2. The method for identifying corporate relationships according to claim 1, characterized in that: The basic related information of the enterprise includes one or more of the following: branches, subsidiaries, foreign investments, actual controllers, shareholders, persons acting in concert, senior executives, legal persons and family relationships, where the family relationships are family relationships among actual controllers, shareholders, persons acting in concert, senior executives and legal persons.

3. The method for identifying corporate relationships according to claim 1 or 2, characterized in that: The collecting of basic association information of the enterprise and determining the explicit relationship members of the enterprise include: Collect basic related information of enterprises, and screen out the first-order relationships, second-order relationships, and third-order relationships of the enterprises; The explicit relationship members of the enterprise are determined according to the first-order relationship of the enterprise, the second-order relationship of the enterprise and the third-order relationship of the enterprise.

4. The method for identifying corporate relationships according to claim 3, characterized in that: The determining the explicit relationship members of the enterprise according to the first-order relationship of the enterprise, the second-order relationship of the enterprise and the third-order relationship of the enterprise includes: Establishing a member relationship graph of the enterprise according to the first-order relationship of the enterprise, the second-order relationship of the enterprise and the third-order relationship of the enterprise; According to the member relationship diagram of the enterprise, the explicit relationship members of the enterprise are determined.

5. The method for identifying corporate relationships according to claim 1, characterized in that: The controlling member of the enterprise is a member that has a controlling relationship with the enterprise.

6. The method for identifying corporate relationships according to claim 1, characterized in that: The persons acting in concert are enterprises or individuals that meet any of the following relationships: kinship, member enterprises of the same group, enterprises with the same directors, supervisors or senior managers, enterprises that directly or indirectly hold shares in the same enterprise, enterprises controlled by the same person and enterprises that hold shares in the same enterprise.

7. The method for identifying corporate relationships according to claim 1, characterized in that: The step of obtaining implicit relationship members of the enterprise according to the weighted directed graph includes: Starting from the persons acting in concert, the sum of the equity of the enterprise nodes is calculated according to the directed graph with weights; Based on the sum of the equity of the enterprise nodes, the implicit relationship members of the enterprise are obtained.

8. The method for identifying corporate relationships according to claim 1, characterized in that: The preset expert rules include the expert rules set by the CBRC.

9. The method for identifying corporate relationships according to claim 1 or 8, characterized in that: The candidate members related to the enterprise include enterprises or individuals having relationships with the enterprise in terms of personnel, finance and business.

10. The method for identifying corporate relationships according to claim 1, characterized in that: Before the method of using the preset identification model to identify the model relationship members of the enterprise among the potential enterprise relationship members, the method further includes: The explicit relationship members, the implicit relationship members and the expert relationship members are taken as positive samples, and the relationship members who do not belong to the enterprise are taken as negative samples. The preset recognition model is trained with the positive samples and the negative samples.

11. The method for identifying corporate relationships according to claim 10, characterized in that: The ratio of the positive samples to the negative samples is between 1 / 20 and 1 / 5.

12. The method for identifying corporate relationships according to claim 1 or 10, characterized in that: The potential corporate relationship members include enterprises that have one or more of the following relationships with the enterprise: a relationship of large-scale operating transactions, a relationship of large-scale non-operating capital occupation and / or being occupied, a relationship of large-scale financing guarantees, being guaranteed or having mutual guarantees, and a relationship of transferring assets and profits at unfair prices and conditions.

13. The method for identifying corporate relationships according to claim 1, characterized in that: The preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members in combination with the GBDT algorithm.

14. A device for identifying corporate relationships, characterized in that: include: An identification module is specifically used to collect basic related information of an enterprise and determine explicit related members of the enterprise; An implicit module, used for obtaining implicit relationship members of the enterprise based on the relationship between the holding members of the enterprise; And, construct an undirected graph according to the relationship of the persons acting in concert among the holding members of the enterprise; establish a directed graph with weights based on the undirected graph in combination with the equity ratio of the persons acting in concert; and obtain the implicit relationship members of the enterprise according to the directed graph with weights; and, based on the equity ratio of the persons acting in concert, calculating the weights of the edges in the undirected graph; merging the multiple edges in the undirected graph according to the weights of the edges in the undirected graph to obtain the directed graph with weights; and, merging the multiple edges in the undirected graph according to the maximum value of the weights of the edges in the undirected graph; an expert module, for obtaining expert relationship members of the enterprise from among candidate members related to the enterprise according to preset expert rules; An identification module, used to identify model relationship members of the enterprise among potential enterprise relationship members by using a preset identification model, wherein the preset identification model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members; The preset recognition model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members in combination with the GBDT algorithm and the community discovery algorithm; the recognition model is trained using the GBDT algorithm to estimate the full sample pairs, obtain the probability that the sample items belong to the same group relationship, remove the edges with lower probability, and then use the modularity-based community discovery algorithm to discover the community; A merging module is used to merge the explicit relationship members, the implicit relationship members, the expert relationship members and the model relationship members to obtain the relationship members of the enterprise.

15. An electronic device for identifying corporate relationships, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 13.

16. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 13 is implemented.

Images