CN112417060A

CN112417060A - Method, device, equipment and computer readable medium for identifying enterprise relationship

Info

Publication number: CN112417060A
Application number: CN202011307613.1A
Authority: CN
Inventors: 夏成扬; 袁进威; 吴超荣; 关健
Original assignee: China Construction Bank Corp
Current assignee: China Construction Bank Corp
Priority date: 2020-11-19
Filing date: 2020-11-19
Publication date: 2021-02-26

Abstract

The invention discloses a method, a device, equipment and a computer readable medium for identifying enterprise relations, and relates to the technical field of computers. One embodiment of the method comprises: acquiring basic associated information of an enterprise, and determining an explicit relation member of the enterprise; acquiring implicit relation members of the enterprise based on the relation between the central stock members of the enterprise; acquiring expert relation members of the enterprise according to a preset expert rule from the candidate members related to the enterprise; adopting a preset identification model to identify model relation members of the enterprise from potential enterprise relation members, wherein the preset identification model is obtained by training the dominant relation member, the recessive relation member and the expert relation member; and combining the dominant relationship member, the recessive relationship member, the expert relationship member and the model relationship member to obtain the relationship member of the enterprise. The method and the device can improve the accuracy and the working efficiency of identifying the enterprise relationship.

Description

Method, device, equipment and computer readable medium for identifying enterprise relationship

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a computer-readable medium for identifying an enterprise relationship.

Background

In the financial industry, especially the banking industry, aiming at the identification degree of enterprise relations, the capacity of bank operation and risk management is reflected to a certain extent, and the method is an important embodiment of commercial comprehensive competitiveness.

At present, the traditional enterprise relationship is mainly determined and maintained manually, depends on the cognitive degree and experience of personnel, and needs to expend a great deal of energy to collect information of customers and groups. With the continuous development of society, the business and social activities between enterprises and individuals are increasingly abundant, and the relationship is more and more complicated and obscure.

In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art: the error rate of manually identifying the enterprise relationship is high and the working efficiency is low.

Disclosure of Invention

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

To achieve the above object, according to an aspect of an embodiment of the present invention, there is provided a method for identifying a business relationship, including:

acquiring basic associated information of an enterprise, and determining an explicit relation member of the enterprise;

acquiring implicit relation members of the enterprise based on the relation between the central stock members of the enterprise;

acquiring expert relation members of the enterprise according to a preset expert rule from the candidate members related to the enterprise;

adopting a preset identification model to identify model relation members of the enterprise from potential enterprise relation members, wherein the preset identification model is obtained by training the dominant relation member, the recessive relation member and the expert relation member;

and combining the dominant relationship member, the recessive relationship member, the expert relationship member and the model relationship member to obtain the relationship member of the enterprise.

The basic associated information of the enterprise comprises one or more of branch institutions, subsidiary companies, external investment, actual control persons, stockholders, consistent actors, high governance, legal persons and relative relations, wherein the relative relations are the relative relations of the actual control persons, the stockholders, the consistent actors, the high governance and the legal persons.

The collecting basic associated information of an enterprise and determining an explicit relationship member of the enterprise comprises:

acquiring basic association information of an enterprise, and screening out a first-order relation of the enterprise, a second-order relation of the enterprise and a third-order relation of the enterprise;

and determining an dominant relationship member 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.

Determining an explicit relationship member 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, including:

establishing a membership 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;

and determining the dominant relation members of the enterprise according to the membership map of the enterprise.

The acquiring the implicit relationship members of the enterprise based on the relationship between the central stock members of the enterprise comprises the following steps:

constructing an undirected graph according to the relation of consistent actors in the enterprise stock control members;

establishing a directed graph with weight on the basis of the undirected graph by combining the share weight ratio of the consistent actor;

and acquiring the implicit relation members of the enterprise according to the weighted directed graph.

Wherein the enterprise central stock member is a member having a stock-controlling relationship with the enterprise.

The consistency actors are enterprises or persons in any relationship, relationship of relatives, member enterprises in the same group, enterprises of the same director, prison or high-level manager, enterprises holding shares directly or indirectly for the same enterprise, enterprises controlled by the same person and enterprises holding shares for the same enterprise.

The establishing of the directed graph with the weight on the basis of the undirected graph by combining the share weight ratio of the consistent actor comprises the following steps:

calculating the weight of the edge in the undirected graph based on the share weight ratio of the consistent actor;

and combining a plurality of edges in the undirected graph according to the weight of the edges in the undirected graph to obtain the weighted directed graph.

The merging the multiple edges in the undirected graph according to the weight of the edges in the undirected graph includes:

and combining a plurality of edges in the undirected graph according to the maximum value of the weights of the edges in the undirected graph.

The acquiring the implicit relationship members of the enterprise according to the weighted directed graph comprises the following steps:

starting from the consistent actor, calculating the sum of the share weights of the enterprise nodes according to the directed graph with the weights;

and acquiring the implicit relation members of the enterprise based on the sum of the equity of the enterprise nodes.

The preset expert rules comprise expert rules set by the bank prison.

The candidate members associated with the enterprise include enterprises or persons having a personal, financial, and business relationship with the enterprise.

Before the model relationship member of the enterprise is identified in the potential enterprise relationship members by adopting the preset identification model, the method further comprises the following steps:

and taking the dominant relationship member, the recessive relationship member and the expert relationship member as positive samples, taking the relationship members not belonging to the enterprise as negative samples, and training the positive samples and the negative samples to obtain the preset identification model.

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

The potential business relationship members include businesses that have one or more of a relationship with the business, a relationship that has a large amount of business transactions, a relationship that has a large amount of non-business funds occupancy and/or occupancy, a relationship that has a large amount of financing guarantees, a relationship that is guaranteed or has mutual guarantees, and a relationship that has a non-fair price and conditional transfer of assets and profits.

The preset identification model is obtained by training the explicit relation member, the implicit relation member and the expert relation member by combining a GBDT algorithm.

The preset identification model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members in combination with a GBDT algorithm and a community discovery algorithm.

According to a second aspect of the embodiments of the present invention, there is provided an apparatus for identifying an enterprise relationship, including:

the dominant module is used for collecting basic associated information of an enterprise and determining dominant relation members of the enterprise;

the implicit module is used for acquiring implicit relation members of the enterprise based on the relation between the central stock control members of the enterprise;

the expert module is used for acquiring the expert relation members of the enterprise according to a preset expert rule from the candidate members related to the enterprise;

the identification module is used for identifying the model relation members of the enterprise from potential enterprise relation members by adopting a preset identification model, wherein the preset identification model is obtained by training the dominant relation member, the recessive relation member and the expert relation member;

and the merging module is used for merging the dominant relationship member, the recessive relationship member, the expert relationship member and the model relationship member to obtain the relationship member of the enterprise.

According to a third aspect of the embodiments of the present invention, there is provided an electronic device for identifying an enterprise relationship, including:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the method as described above.

According to a fourth aspect of embodiments of the present invention, there is provided a computer readable medium, on which a computer program is stored, which when executed by a processor, implements the method as described above.

One embodiment of the above invention has the following advantages or benefits: acquiring basic associated information of an enterprise, and determining an explicit relation member of the enterprise; acquiring implicit relation members of the enterprise based on the relation between the central stock members of the enterprise; acquiring expert relation members of the enterprise according to a preset expert rule in candidate enterprises related to the enterprise; adopting a preset identification model to identify model relation members of the enterprise from potential enterprise relation members, wherein the preset identification model is obtained by training the dominant relation member, the recessive relation member and the expert relation member; and combining the dominant relationship member, the recessive relationship member, the expert relationship member and the model relationship member to obtain the relationship member of the enterprise. Because the relation members of the enterprise can be automatically identified from multiple aspects, the working efficiency is improved, and the error rate of identifying the enterprise relation is reduced.

Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.

Drawings

The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:

FIG. 1 is a schematic diagram of a main flow of a method of identifying business relationships, according to an embodiment of the invention;

FIG. 2 is a flow diagram illustrating a process for determining dominant relationship members of an enterprise according to an embodiment of the invention;

FIG. 3 is a schematic flow diagram for determining dominant relationship members of an enterprise according to a relationship diagram, according to an embodiment of the invention;

FIG. 4 is a schematic diagram of a membership map for Enterprise A, according to an embodiment of the present invention;

FIG. 5 is a schematic illustration of an actor according to an embodiment of the invention;

FIG. 6 is a flow diagram of learning implicit relationship members of an enterprise, according to an embodiment of the invention;

FIG. 7 is a flow diagram illustrating the process of building a directed graph with weights according to an embodiment of the present invention;

FIG. 8 is a flow diagram illustrating learning implicit relationship members of an enterprise from a weighted directed graph, according to an embodiment of the invention;

fig. 9 is a schematic diagram of a main structure of an apparatus for identifying business relationships according to an embodiment of the present invention;

FIG. 10 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;

fig. 11 is a schematic structural diagram of a computer system suitable for implementing a terminal device or a server according to an embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

The group member enterprises have the characteristics of cross-region, cross-industry and the like, and along with the continuous development of market economy, the relationship of the group member enterprises is difficult to be investigated and clarified unilaterally. Meanwhile, the association and interaction between enterprises is increasingly complex, and the outbreak of the risk of a single client may cause the risk of the whole group level and the regional financial risk may be caused seriously. The business behavior of the customer, the risks that may be brought to the bank, is becoming more and more obscure and difficult to locate. Therefore, how to rapidly and accurately excavate group member enterprises; how to make the risk management work of customers efficient and high-quality is a serious challenge.

The group judgment is mainly performed from the control right point of view by combining the definition of the group client in the related file, and the control right can be embodied in the following two aspects. On one hand, the control relationship among stockholders, actual control persons, subsidiaries, branches and the like is direct; another aspect is that some hidden control relations exist, so that economic features of large fund occupation, property transfer not according to the fair value and the like occur.

However, in the actual operation process, it is found that many hidden potential groups cannot be identified, so that the scale of the group in the actual group relationship is not consistent with the actual scale, and the risk management work is influenced to a certain extent.

The traditional manual identification mode has the disadvantages of information asymmetry, small identification scale and the great limitations of incomplete consideration and the like, so that the error rate of the enterprise relationship is high and the working efficiency is low.

In order to solve the technical problems of high error rate and low working efficiency of manually identifying enterprise relationships, the following technical scheme in the embodiment of the invention can be adopted.

Referring to fig. 1, fig. 1 is a schematic diagram of a main flow of a method for identifying enterprise relationships according to an embodiment of the present invention, and relationship members of an enterprise are determined from multiple aspects. As shown in fig. 1, the method specifically comprises the following steps:

s101, collecting basic associated information of the enterprise, and determining an explicit relation member of the enterprise.

In the embodiment of the invention, one or more servers can be adopted to collect the basic associated information of the enterprise respectively in consideration of large data volume of the basic associated information of the enterprise.

The basic associated information of the enterprise comprises one or more of branch organizations, subsidiaries, external investments, actual control persons, stockholders, consistent actors, high governance, legal persons and relative relations. Wherein the relationship is that of actual control people, stockholders, uniform actors, high-head governors and legal people.

In the embodiment of the invention, the determined relation member is called an dominant relation member of the enterprise based on the basic association information of the enterprise.

Referring to fig. 2, fig. 2 is a schematic flowchart illustrating a process of determining an explicit relationship member of an enterprise according to an embodiment of the present invention, which specifically includes the following steps:

s201, collecting basic associated information of the enterprises, and screening out first-order relations of the enterprises, second-order relations of the enterprises and third-order relations of the enterprises.

In the embodiment of the invention, the enterprise pair is formed by derivation of the multi-order relation according to the relation between the entities reflecting the enterprise control right, and the enterprise forms the group through the control right relation. Based on the existing subordinate enterprises of the group, more hidden potential connections among the enterprises are explored.

The relationships between enterprises include first order relationships of the enterprises, second order relationships of the enterprises, and third order relationships of the enterprises.

The first order relationships of an enterprise refer to one or more of a branch office, a subsidiary, an investor, an actual controller, a stockholder, and a coherent actor. The action-consistent person refers to an action or fact that an investor expands the number of the votes of the stock shares of the listed company which the investor can control with other investors through agreements and other arrangements.

The second order relationship of an enterprise means having the same actual controller, such as: A. business B business incumbent/incumbent high-tube, legal, or stockholder coincided.

The third-order relationship of the enterprise means having the relationship of job, relative and recessive. The implicit relationship refers to an objectively existing and unknown relationship between enterprises or people. Such as: the third-order relationship of an enterprise includes the existence of a spouse, sibling or child-parent relationship between the positions of two companies, which may be actual controls, jurisdictions, directors, prisoners, high governments or stockholders.

S202, determining an explicit relation member of the enterprise according to the first-order relation of the enterprise, the second-order relation of the enterprise and the third-order relation of the enterprise.

In the embodiment of the invention, the members having the first-order relationship, the second-order relationship and the third-order relationship with the enterprise are determined as the dominant 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 dominant relationship member of the other enterprise.

It is to be appreciated that the dominant relationship member of an enterprise can be obtained from a first order relationship member of the enterprise, a second order relationship member of the enterprise, or a third order relationship member of the enterprise.

In one embodiment of the invention, in order to improve the speed of determining the dominant relationship members of an enterprise, the determination can be realized through a membership graph.

Referring to fig. 3, fig. 3 is a schematic flowchart of determining an explicit relationship member of an enterprise according to a relationship diagram according to an embodiment of the present invention, which specifically includes the following steps:

s301, establishing a membership 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.

The first order relationship of the enterprise, the second order relationship of the enterprise and the third order relationship of the enterprise are all relationships with the enterprise. And further, taking the enterprise as a center, and establishing a membership graph of the enterprise.

Referring to fig. 4, fig. 4 is a schematic diagram of a membership map of enterprise a, according to an embodiment of the invention. In fig. 4, enterprise a is shown as a parent company, and a membership map of enterprise a is shown.

Centered on business a, there are three circles in fig. 4. Wherein, the members in the circle with the minimum radius are the members belonging to the first-order relationship of the enterprise A; members between the minimum radius circle and the intermediate length radius circle, i.e., members belonging to the second order relationship of enterprise a; members between the middle-length radius circle and the maximum radius circle, i.e., members of a third-order relationship belonging to enterprise a. The connection between enterprise A and the member marks the share proportion.

S302, determining an explicit relation member of the enterprise according to the member relation graph of the enterprise.

The relationships of the enterprise and the members thereof are clearly identified in the member relationship diagram of the enterprise, and then the display relationship members of the enterprise can be determined according to the member relationship diagram of the enterprise.

In the embodiment of fig. 3, the determination speed can be improved by determining the dominant relationship members of the enterprise by using the membership map of the enterprise.

S102, acquiring implicit relation members of the enterprise based on the relation between the enterprise central control stock members.

The control of stocks among enterprises is divided into direct control and indirect control. Enterprises control other enterprises in a multilayer relationship or a family relationship through indirect control.

In the embodiment of the invention, based on the relationship between the enterprise central stock members, the implicit relationship members of the enterprise are acquired. An enterprise central stock member is a member that has a stock-keeping relationship with the enterprise.

Specifically, a consistent actor needs to be known first. In the embodiment of the invention, the uniform actors are enterprises or persons in any relationship, relationship of relatives, member enterprises in the same group, enterprises of the same director, supervisor or high-level manager, enterprises holding shares directly or indirectly for the same enterprise, enterprises controlled by the same person and enterprises holding shares for the same enterprise.

That is, a person who meets one of the following conditions belongs to a uniform actor.

Condition 1: the natural human entities of the relatives belong to a uniform actor.

Condition 2: member enterprises in the same group belong to a uniform actor

Condition 3: according to A and B in figure 6, belonging to an uniform actor.

Referring to fig. 5, fig. 5 is a schematic diagram of an actor in accordance with an embodiment of the invention. The first left line of the diagram in fig. 5 is two businesses belonging to a same board of directors, proctor, or higher level manager as an actor. That is, person C is the director, supervisor, or high-level manager of enterprise a, and is also the director, supervisor, or high-level manager of enterprise B, then enterprise a and enterprise B are congruent actors.

The first line of FIG. 5 illustrates that businesses that are directly or indirectly held by the same business are coherent actors. That is, enterprise a and enterprise B are direct holdings of enterprise C, and enterprise a and enterprise B are congruent actors.

The first right-hand row of fig. 5 shows that the same business, directly or indirectly holding stock, is a coherent actor. That is, Enterprise A and Enterprise B are direct holdings of Enterprise C. Meanwhile, enterprise D controls the stock of enterprise A and enterprise B, and then enterprise A and enterprise B are consistent actors.

The second left row of fig. 5 is for a business controlled by the same person or held by the same business. That is, person C is the director, supervisor, or high-level manager of enterprise a, while person C controls that enterprise B holds shares for enterprise a, then enterprise a and enterprise B are congruent actors.

The diagram in the second row of FIG. 5 is for a holding business for the same business, with the held business or individual being a consistent actor. That is, person A holds stock for business C, business B holds stock for business C, and person A holds 30% of stock for business B, then person A and business B are congruent actors.

The second row, right, of FIG. 5 is for a holding enterprise for the same enterprise, the held enterprise being a conforming actor. That is, enterprise a holds shares for enterprise C, enterprise B holds shares for enterprise C, and there is an economic interest relationship between enterprise a and enterprise B, so enterprise a and enterprise B are an actor.

Referring to fig. 6, fig. 6 is a schematic flowchart of a process for learning implicit relationship members of an enterprise according to an embodiment of the present invention, which specifically includes the following steps:

s601, constructing an undirected graph according to the relation of the consistent actors in the enterprise stock control members.

Members of an enterprise's implicit relationship typically hold stock indirectly to the enterprise. The enterprise control members comprise one or more consistency actors, and the implicit relation members of the enterprise can be obtained by analyzing the relation between the consistency actors and the enterprise.

Specifically, the connection between the actor and the enterprise is established according to the relationship of the actor in the enterprise stock control members. And constructing an undirected graph by connecting a plurality of consistent actors with the enterprise.

And S602, establishing a directed graph with weight on the basis of the undirected graph by combining the share weight ratio of the uniform actor.

In the constructed undirected graph, the connection of a business with a consistent actor is undirected and without weight. But the implicit relationship members of the enterprise are acquired by considering the stock right relationship between the enterprise and the uniform actor. Then a directed graph with weights can be built on the basis of an undirected graph.

Referring to fig. 7, fig. 7 is a schematic flowchart of establishing a directed graph with weights according to an embodiment of the present invention, which specifically includes the following steps:

s701, calculating the weight of the edge in the undirected graph based on the share weight ratio of the consistent actor.

All connected components are found in the undirected graph as consistent actor entity combinations. As one example, all connected components are solved in an undirected graph using traversal of the graph or using a union set.

The weight value of the edge represents the share proportion or the investment proportion on the basis of all connected components of the undirected graph. For the actual control stock, the weight of the edge is 1. Adding two-way relation to every two entities in the consistent action person combination, wherein the weight of the edge is 1. Entities are nodes in an undirected graph.

S702, combining a plurality of edges in the undirected graph according to the weight of the edges in the undirected graph to obtain the weighted directed graph.

In the undirected graph, two entities are connected through edges, the number of the edges is one or more, and in order to obtain the share right relationship of the two entities, a plurality of edges in the undirected graph can be merged. Namely, according to the weight of the edge in the undirected graph, combining a plurality of edges in the undirected graph to obtain the weighted directed graph.

In an 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.

The weighted directed graph can embody a control path and a decision path and comprises actual control, stockholders, equity, individual investment and enterprise external investment relation.

In the embodiment of fig. 7, a directed graph with weights is established by combining the share right ratios of the actors, so that a foundation is laid for learning the recessive relationship members of the enterprise.

And S603, acquiring the recessive relation members of the enterprise according to the directed graph with the weight.

The weighted directed graph comprises a plurality of nodes, and each node has a weight. And further, the implicit relation members of the enterprise can be known.

Referring to fig. 8, fig. 8 is a schematic flowchart of acquiring implicit relationship members of an enterprise according to a weighted directed graph according to an embodiment of the present invention, which specifically includes:

and S801, starting from a uniform actor, calculating the sum of the share rights of the enterprise nodes according to the directed graph with the weight.

In the weighted directed graph, starting from any node in the consistent actor, the combined share right of the consistent actor to the enterprise node is calculated.

Then, the full path of the node to each enterprise node is computed. In the above path, the stock right is accumulated according to the point multiplication, and the combined stock right of the actor for each enterprise node is the sum of the stock rights of the enterprise nodes of all paths.

S802, acquiring the recessive relation members of the enterprise based on the sum of the equity of the enterprise nodes.

The indirect share right of a certain node to a certain enterprise is the combined share right of the consistent actor belonging to the node to the enterprise. Then, based on the sum of the equity of the enterprise nodes and the manner in which indirect equity is determined, the enterprise or individual having indirect equity to the enterprise will be a member of the enterprise's implicit relationship.

In the embodiment of fig. 8, the implicit relationship members of the enterprise are learned from the weighted directed graph.

In the embodiment of fig. 6, the implicit relationship members of the enterprise can be timely and accurately known through the directed values with weights.

By adopting the technical scheme in the S102, the accuracy of determining the relation members of the enterprise can be improved. Taking a company as a case, the analysis determines the relationship members by using the prior art, and determines the relationship members by using the scheme in S102.

And manually identifying the relationship of the enterprise, and determining that the number of the relationship members of the enterprise is 106. And determining 113 relation members of the enterprise by adopting the scheme in the S102. It can be seen that the effect of adopting the scheme in S102 is obvious.

S103, acquiring the expert relation members of the enterprise according to a preset expert rule from the candidate members related to the enterprise.

In the embodiment of the invention, the expert relation members of the enterprise can be acquired according to the preset expert rules. Wherein the candidate members associated with the enterprise include enterprises or persons having a personal, financial, and business relationship with the enterprise.

In an embodiment of the present invention, the preset expert rules include expert rules set by the bank prison. As an example, the preset expert rules include the following three types of rules:

a first expert rule: and the candidate members related to the enterprise are stockholders of the enterprise, and share accounts are more than 50%.

Second expert rules: among the candidate members associated with the enterprise, the first big stockholder of the enterprise is selected, and the share ratio is more than 25%.

A third expert rule: among candidate members related to the enterprise, the total investment amount of the enterprise accounts for more than 50% of the registered capital.

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

the group mother company controls the stock of the enterprise.

The former 10 natural stocks of the group mother company control the enterprises externally.

The group mother company director, supervise or high control the stock enterprise.

The corporate of the group parent company controls the stock or controls the stock after the stock right is merged.

The group parent company and the former 10 nature holders can control the stock after merging the stock right.

Group mother company and its legal person, director, supervisor or high manager, and company capable of controlling stock after combining stock right.

The former 10 natural persons stockholders and the former 10 natural persons stockholders of the group parent company, and the company which can control stocks after the relatives of the persons merge the stock right.

The corporate of the group member company controls the stock or controls the stock after combining the stock right.

The directors, supervisors or high administration of group member companies control the enterprises with stock externally.

The actual controller of the group parent company acts as all the enterprises of the legal person.

And S104, identifying the model relation members of the enterprise in the potential enterprise relation members by adopting a preset identification model, wherein the preset identification model is obtained by training of dominant relation members, recessive relation members and expert relation members.

For the relationship members which cannot be determined through S101, S102 and S103, the relationship members can be further determined by adopting a preset identification model. And the relation members identified by the preset identification model are called model relation members of the enterprise.

The establishment of the predetermined recognition model is described below. The preset recognition model belongs to a machine learning model. For machine learning models, training with data is required.

In the embodiment of the invention, the preset identification model is obtained by training of an explicit relation member, an implicit relation member and an expert relation member. It is to be understood that after determining the relationship members in S101, S102, and S103, the predetermined recognition model is trained using the determined relationship members as training data to satisfy coverage of the following classes of members.

Members of the first class: and enterprises with continuous and large-volume business transaction with the confirmed relation members and obvious correlation between business and credit conditions exist. That is, enterprises in a relationship of large business transactions.

Members of the second class: enterprises with large non-operational fund occupation and/or occupation and obvious correlation between operation and credit conditions with the confirmed relation members exist. That is, businesses that have large non-operational capital occupancies and/or occupied relationships.

Members of the third class: the enterprise has a large financing guarantee or a guarantee with the confirmed relation members, or has a mutual guarantee and obvious correlation between mutual operation and credit conditions. That is, there are large financing guarantees, and enterprises that are guaranteed or have a mutual guarantee relationship.

Members of the fourth class: enterprises that have other associations with identified relationship members, transfer assets and profits according to non-fair prices and conditions. That is, there are businesses that shift assets and profits at non-fair prices versus conditions.

Specifically, training the pre-set recognition model requires positive and negative examples. Positive samples are the members of the relationship belonging to the enterprise; negative examples are relationship members that do not belong to the enterprise.

In the embodiment of the invention, the dominant relation member, the recessive relation member and the expert relation member are used as positive samples, the relation members which do not belong to the enterprise are used as negative samples, and the positive samples and the negative samples are used for training to obtain the preset recognition model.

In the implementation process of the embodiment of the invention, under large-scale data, the number of positive samples is far smaller than that of negative samples, and the number of positive samples is insufficient. Sample sampling may be employed to increase the number of samples that need to be taken based on the overall sample distribution.

Through practice for many times, under the condition that the ratio of the positive sample to the negative sample is between 1/20 and 1/5, the recognition effect of the trained preset recognition model is good. That is, the ratio of positive samples to negative samples is at least 1: 20, and the ratio of positive samples to negative samples is at most 1: 5.

After the positive samples and the negative samples are determined, a preset recognition model can be obtained by adopting GBDT algorithm training. Namely, the preset identification model is obtained by combining the training of the dominant relationship member, the recessive relationship member and the expert relationship member with the GBDT algorithm.

The GBDT algorithm is a decision tree model based on an integrated idea and is essentially based on residual error learning. Is characterized in that: various types of data can be processed; the accuracy is higher; and robustness to abnormal values is strong.

In one embodiment of the invention, in order to improve the speed of training the recognition model, the recognition model can be trained by combining a community discovery algorithm. The Community discovery (Community Detection) algorithm is used for discovering Community structures in a network, and is a clustering algorithm.

Namely, the preset identification model is obtained by training the explicit relation member, the implicit relation member and the expert relation member in combination with the GBDT algorithm and the community discovery algorithm.

And training the recognition model by using a GBDT algorithm, and estimating the total sample pairs to obtain the probability that the sample pieces belong to the same group relation. Edges with lower probability are removed, and then community discovery algorithm based on modularity is used to discover communities.

And finally, obtaining a preset identification model, and identifying the model relation members of the enterprise in the potential enterprise relation members by adopting the preset identification model.

And S105, combining the dominant relationship member, the recessive relationship member, the expert relationship member and the model relationship member to obtain the relationship member of the enterprise.

After determining the dominant relationship member, the recessive relationship member, the expert relationship member and the model relationship member, considering the possibility that the determined relationship members are repeated, the relationship members of the enterprise need to be obtained after combination. It is understood that the relationship members of the enterprise are composed of explicit relationship members, implicit relationship members, expert relationship members, and model relationship members.

In the above embodiment, basic association information of an enterprise is collected, and an explicit relationship member of the enterprise is determined; acquiring implicit relation members of the enterprise based on the relation between the central stock members of the enterprise; acquiring expert relation members of the enterprise according to a preset expert rule in candidate enterprises related to the enterprise; adopting a preset identification model to identify model relation members of the enterprise from potential enterprise relation members, wherein the preset identification model is obtained by training the dominant relation member, the recessive relation member and the expert relation member; and combining the dominant relationship member, the recessive relationship member, the expert relationship member and the model relationship member to obtain the relationship member of the enterprise. Because the relation members of the enterprise can be automatically identified from multiple aspects, the working efficiency is improved, and the error rate of identifying the enterprise relation is reduced.

It is to be appreciated that in order to identify business relationships, one proceeds from several aspects as follows.

In a first aspect: an explicit relationship member is determined.

And carrying out stock right penetration by utilizing a direct control relationship among enterprises, and discovering other related companies controlled by group members through three-order control right derivation based on the discovered members.

In a second aspect: and determining the recessive relation member.

And mining is carried out on partial members which also form the enterprise relationship but are not found through directed graph calculation.

In a third aspect: and determining the model relation members.

And (4) using the economic traffic characteristics, using GBDT algorithm training to obtain a preset recognition model, judging whether the enterprise pairs belong to the same group, and mining from the economic traffic aspect.

Referring to fig. 9, fig. 9 is a schematic diagram of a main structure of an apparatus for identifying business relationships according to an embodiment of the present invention, where the apparatus for identifying business relationships may implement a method for identifying business relationships, and as shown in fig. 9, the apparatus for identifying business relationships specifically includes:

an explicit module 901, configured to collect basic association information of an enterprise, and determine an explicit relationship member of the enterprise;

a recessive module 902, configured to obtain a recessive relationship member of the enterprise based on a relationship between the central stock members of the enterprise;

an expert module 903, configured to obtain, from the candidate members related to the enterprise, an expert relation member of the enterprise according to a preset expert rule;

an identifying module 904, configured to identify a model relationship member of the enterprise among potential enterprise relationship members by using a preset identifying model, where the preset identifying model is obtained by training of the explicit relationship member, the implicit relationship member, and the expert relationship member;

a merging module 905, configured to merge the explicit relationship member, the implicit relationship member, the expert relationship member, and the model relationship member to obtain the relationship member of the enterprise.

In one embodiment of the invention, the basic associated information of the enterprise comprises one or more of branches, subsidiaries, investments, actual control persons, stockholders, uniform actors, high governance, legal persons and relatives, and the relatives are the relatives of the actual control persons, stockholders, uniform actors, high governance and legal persons.

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

In an embodiment of the present invention, the explicit module 901 is specifically configured to establish a membership 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;

In an embodiment of the present invention, the implicit module 902 is specifically configured to construct an undirected graph according to a relationship between actors in the enterprise stock control members;

In one embodiment of the invention, the enterprise member having a holdings relationship with the enterprise.

In one embodiment of the invention, the actors are enterprises or persons in any relationship, member enterprises in the same group, enterprises of the same director, supervisor or high-level manager, enterprises holding shares directly or indirectly to the same enterprise, enterprises controlled by the same person and enterprises holding shares to the same enterprise.

In an embodiment of the present invention, the implicit module 902 is specifically configured to calculate a weight of an edge in the undirected graph based on the share ratio of the actor;

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

In an embodiment of the present invention, the recessive module 902 is specifically configured to, starting from the actor, calculate a sum of share weights of the enterprise nodes according to the weighted directed graph;

In an embodiment of the present invention, the preset expert rules include expert rules set by the bank prison.

In one embodiment of the invention, the candidate members associated with the business include businesses or persons having a personal, financial and business relationship with the business.

In an embodiment of the present invention, the identifying module 904 is specifically configured to train the explicit relationship member, the implicit relationship member, and the expert relationship member as positive samples, and the relationship member not belonging to the enterprise as negative samples to obtain the preset identifying model.

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

In one embodiment of the invention, the potential business relationship members include businesses having one or more of the following relationships with the business, having a large amount of an operational transaction, having a large amount of a non-operational fund occupancy and/or occupancy relationship, having a large amount of a financing guarantee, being guaranteed or having a mutual guarantee relationship, and having a non-fair price and conditional transfer asset and profit relationship.

In one embodiment of the 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 a GBDT algorithm.

In one embodiment of the 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 a GBDT algorithm and a community discovery algorithm.

Fig. 10 illustrates an exemplary system architecture 1000 to which the method of identifying business relationships or the apparatus for identifying business relationships of embodiments of the invention may be applied.

As shown in fig. 10, 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. Network 1004 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

A user may use the

terminal devices

1001, 1002, 1003 to interact with a server 1005 via a network 1004 to receive or transmit messages or the like. The

terminal devices

1001, 1002, 1003 may have installed thereon various messenger client applications such as shopping applications, web browser applications, search applications, instant messenger, mailbox clients, social platform software, etc. (by way of example only).

The

terminal devices

1001, 1002, 1003 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 1005 may be a server that provides various services, such as a backend management server (for example only) that supports shopping websites browsed by users using the

terminal devices

1001, 1002, 1003. The backend management server may analyze and perform other processing on the received data such as the product information query request, and feed back a processing result (for example, target push information, product information — just an example) to the terminal device.

It should be noted that the method for identifying the business relationship provided by the embodiment of the present invention is generally executed by the server 1005, and accordingly, the apparatus for identifying the business relationship is generally disposed in the server 1005.

It should be understood that the number of terminal devices, networks, and servers in fig. 10 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Referring now to FIG. 11, shown is a block diagram of a computer system 1100 suitable for use with a terminal device implementing an embodiment of the present invention. The terminal device shown in fig. 11 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.

As shown in fig. 11, the computer system 1100 includes a Central Processing Unit (CPU)1101, which can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)1102 or a program loaded from a storage section 1108 into a Random Access Memory (RAM) 1103. In the RAM 1103, various programs and data necessary for the operation of the system 1100 are also stored. The CPU 1101, ROM 1102, and RAM 1103 are connected to each other by a bus 1104. An input/output (I/O) interface 1105 is also connected to bus 1104.

The following components are connected to the I/O interface 1105: an input portion 1106 including a keyboard, mouse, and the like; an output portion 1107 including a signal output unit such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and a speaker; a storage section 1108 including a hard disk and the like; and a communication section 1109 including a network interface card such as a LAN card, a modem, or the like. The communication section 1109 performs communication processing via a network such as the internet. A driver 1110 is also connected to the I/O interface 1105 as necessary. A removable medium 1111 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 1110 as necessary, so that a computer program read out therefrom is mounted into the storage section 1108 as necessary.

In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication portion 1109 and/or installed from the removable medium 1111. The above-described functions defined in the system of the present invention are executed when the computer program is executed by a Central Processing Unit (CPU) 1101.

It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes an explicit module, a implicit module, an expert module, an identification module, and a merge module. The names of these modules do not form a limitation on the modules themselves in some cases, for example, "identify module 904, specifically used for collecting basic association information of an enterprise and determining an explicit relationship member of the enterprise".

As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise:

According to the technical scheme of the embodiment of the invention, basic associated information of an enterprise is collected, and an explicit relation member of the enterprise is determined; acquiring implicit relation members of the enterprise based on the relation between the central stock members of the enterprise; acquiring expert relation members of the enterprise according to a preset expert rule in candidate enterprises related to the enterprise; adopting a preset identification model to identify model relation members of the enterprise from potential enterprise relation members, wherein the preset identification model is obtained by training the dominant relation member, the recessive relation member and the expert relation member; and combining the dominant relationship member, the recessive relationship member, the expert relationship member and the model relationship member to obtain the relationship member of the enterprise. Because the relation members of the enterprise can be automatically identified from multiple aspects, the working efficiency is improved, and the error rate of identifying the enterprise relation is reduced.

The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method of identifying business relationships, comprising:

2. The method for identifying enterprise relations of claim 1, wherein the basic associated information of the enterprise comprises one or more of branches, subsidiaries, investments, actual controllers, stockholders, consensus actors, high governance, legal persons and relatives, and the relatives are the relatives of the actual controllers, stockholders, consensus actors, high governance and legal persons.

3. The method for identifying enterprise relations according to claim 1 or 2, wherein the collecting basic association information of an enterprise and determining an explicit relation member of the enterprise comprises:

4. The method of identifying relationships between businesses as claimed in claim 3, wherein said determining an explicit relationship member of said business based on said first order relationship of said business, said second order relationship of said business and said third order relationship of said business comprises:

5. The method for identifying enterprise relationships according to claim 1, wherein learning implicit relationship members of the enterprise based on the relationships between the enterprise central stock members comprises:

6. The method of identifying enterprise relationships of claim 1 or 5, wherein the enterprise member having a holdings relationship with the enterprise.

7. The method of identifying business relationships of claim 5, wherein the corporate actors are businesses or persons in any relationship, member businesses in the same group, businesses on the same board of director, supervisor, or high-level manager, businesses holding shares directly or indirectly to the same business, businesses controlled by the same person, and businesses holding shares to the same business.

8. The method for identifying business relationships of claim 5, wherein said building a weighted directed graph based on said undirected graph in combination with said share right ratio of said consistent actor comprises:

9. The method of claim 8, wherein the merging the plurality of edges in the undirected graph according to the weights of the edges in the undirected graph comprises:

10. The method for identifying enterprise relationships according to claim 5, wherein learning the implicit relationship members of the enterprise from the weighted directed graph comprises:

11. The method for identifying business relationships of claim 1, wherein the pre-set expert rules comprise expert rules set by the bank council.

12. The method for identifying business relationships of claim 1 or 11, wherein the candidate members related to the business comprise businesses or people having a personal, financial and business relationship with the business.

13. The method for identifying business relationships of claim 1, wherein prior to identifying the model relationship members of the business among the potential business relationship members using the predetermined identification model, further comprising:

14. The method of identifying business relationships of claim 13, wherein the ratio of the positive examples to the negative examples is between 1/20 and 1/5.

15. The method of identifying business relationships of claim 1 or claim 13, wherein the potential business relationship members include businesses having one or more of the following relationships with the business, having a relationship of large business transactions, having a relationship of large non-business funds usage and/or usage, having a large financing guarantee, being guaranteed or having a mutually guaranteed relationship, and having a relationship of transferring assets and profits at a non-fair price and condition.

16. The method of identifying business relationships of claim 1, wherein the pre-defined identification model is obtained by training the explicit relationship members, the implicit relationship members and the expert relationship members in conjunction with a GBDT algorithm.

17. The method of identifying business relationships of claim 1, wherein the pre-defined identification 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.

18. An apparatus for identifying business relationships, comprising:

an identifying module 904, configured to specifically collect basic association information of an enterprise, and determine an explicit relationship member of the enterprise;

19. An electronic device that identifies business relationships, comprising:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-17.

20. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-17.