CN111382956A

CN111382956A - Enterprise group relationship mining method and device

Info

Publication number: CN111382956A
Application number: CN202010230555.0A
Authority: CN
Inventors: 郑宇瀚; 陈青山; 李晓敦; 万光明; 赵世辉; 邓杨; 高宏华; 章晖; 刘冰冰
Original assignee: China Construction Bank Corp; CCB Finetech Co Ltd
Current assignee: China Construction Bank Corp
Priority date: 2020-03-27
Filing date: 2020-03-27
Publication date: 2020-07-07

Abstract

The invention discloses a method and a device for mining enterprise group relations, and relates to the technical field of computers. One embodiment of the method comprises: acquiring enterprise related data; processing the enterprise related data to obtain key indexes, wherein the key indexes comprise: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises; and mining the enterprise group relationship according to the key indexes. The embodiment can ensure that the mined enterprise group relation data is fuller, and the number of groups and the enterprise coverage are greatly improved; and the method does not need to rely on manpower, saves manpower and time, and is convenient for improving the timeliness of data mining.

Description

Enterprise group relationship mining method and device

Technical Field

The invention relates to the technical field of computers, in particular to a method and a device for mining enterprise group relations.

Background

The traditional enterprise group relationship is mainly identified and maintained manually, and a great deal of energy is required to collect information of customers and groups. The problem is seriously amplified in the current high-speed development era, investment behaviors of enterprises and individuals and social interaction relations change all the time, the traditional group identification mode can cause serious information asymmetry, and meanwhile, the manually-judged group has the problems of small scale, difficulty in considering comprehensiveness and the like.

In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art:

the existing enterprise group relation mining is manually determined and arranged, so that the defects of insufficient group quantity, insufficient enterprise coverage and insufficient timeliness exist.

Disclosure of Invention

In view of this, embodiments of the present invention provide a method and an apparatus for mining enterprise group relationships, so that the mined enterprise group relationship data is fuller, and the number of groups and the coverage of enterprises are both greatly improved; and the method does not need to rely on manpower, saves manpower and time, and is convenient for improving the timeliness of data mining.

To achieve the above object, according to an aspect of the embodiments of the present invention, a method for mining enterprise group relationships is provided.

A mining method of enterprise group relations comprises the following steps: acquiring enterprise related data; processing the enterprise-related data to obtain key indicators, wherein the key indicators include: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises; and mining the enterprise group relationship according to the key indexes.

Optionally, mining the enterprise group relationship according to the key index includes: determining enterprises brought into the group according to the consistent action relationship of the natural people or the enterprises, the stock holding ratio of stock holders of the enterprises and the actual controllers of the enterprises respectively; calculating the importance score of each enterprise according to the relationship among the enterprises in the group; and determining the core enterprises of the clique according to the importance score of each enterprise so as to establish enterprise clique relationship.

Optionally, before processing the enterprise-related data, the method further includes: uploading enterprise-related data to a distributed system framework for processing, wherein the enterprise-related data comprises an enterprise name, a stock right relationship of an enterprise and an enterprise state; removing the weight of the enterprise according to the enterprise name and the share right relation of the enterprise; and filtering the enterprise according to the enterprise state and the enterprise naming rule so as to preprocess enterprise related data.

Optionally, processing the enterprise-related data to obtain a key indicator includes: performing first processing on the enterprise related data to obtain a fusible relationship of natural people and a relationship of relatives of the natural people; and performing second processing on the enterprise related data to obtain the key index based on the fusible relationship of the natural person and the relative relationship of the natural person.

Optionally, the fusible relationship of the natural person is determined by: if two enterprises have a plurality of high-rate managers or natural person shareholders with the same name, every two high-rate managers or natural person shareholders with the same name are in a interfusible relationship; if two enterprises have an investment relationship and the two enterprises have the same-name high-rate manager or natural human shareholder, the same-name high-rate manager or natural human shareholder is in a confldable relationship; if the names of the two enterprises are similar and the two enterprises have the same-name high-speed manager or natural person shareholder, the same-name high-speed manager or natural person shareholder is in a fusible relationship.

Optionally, the natural person's relativity is determined by: obtaining an enterprise range according to the invested relation of the enterprise stockholders or enterprises; taking a high-management person or a stock holder included in each enterprise in the enterprise range as a selected natural person; and judging the relationship of the natural person according to the name of the natural person.

Optionally, the consistent action relationship of the natural person or the enterprise is determined by: determining the consistent action relationship of the natural people or the enterprises based on a set rule according to the relationship of the natural people to the relatives, the stock control relationship of the natural people to the enterprises, the stock holding information of two natural people or the enterprises to a third enterprise and the high-management personnel information of the enterprises, wherein the stock control relationship of the natural people to the enterprises is obtained according to the stock right relationship of the enterprises.

Optionally, the stock holder of the enterprise is determined by: if the enterprise is a listed company, acquiring the stock holder of the enterprise through the latest annual report of the enterprise; and if the enterprise is a non-listed company, taking the stockholder combination with the stock control proportion larger than the set limit value as the stock control stockholder of the enterprise.

Optionally, the actual controller of the enterprise is determined by: if the enterprise is a listed company, acquiring an actual controller of the enterprise through a latest enterprise annual report; if the enterprise is a non-listed company, respectively determining the actual controllers of the enterprise according to the enterprise type of the enterprise, wherein the steps comprise: when the type of the enterprise is individual, taking a legal representative of the enterprise as an actual controller of the enterprise; when the enterprise type is non-individual and a unique stock holder exists, taking the unique stock holder as an actual controller of the enterprise; when the enterprise type is non-individual and has no unique stock holder, determining the level of the investment enterprises or natural persons of the enterprise according to the investment relationship among the enterprises, then sequentially accumulating the stock holding proportion of the investment enterprises or natural persons for each layer of the investment enterprises or natural persons according to the relationship of the natural persons and the consistent action relationship of the natural persons or enterprises in the order from small to large of the level, and taking the combination of the investment enterprises or natural persons with the stock holding proportion exceeding the set proportion for the first time as the actual controller of the enterprise.

Optionally, after mining the enterprise group relationship according to the key index, the method further includes: and constructing an enterprise group mining model so as to mine enterprise group relations by using the enterprise group mining model.

Optionally, the constructing of the enterprise group mining model includes: acquiring existing group data; merging the mined enterprise group data with the existing group data, and constructing a directed graph by using a distributed graph processing framework; and taking all enterprises in each connected block of the directed graph as an enterprise group, and selecting core enterprises from each connected block as the core enterprises of the enterprise group to construct an enterprise group mining model.

According to another aspect of the embodiment of the invention, an excavating device for enterprise group relations is provided.

An enterprise group relationship mining device, comprising: the data acquisition module is used for acquiring enterprise related data; a data processing module, configured to process the enterprise-related data to obtain a key indicator, where the key indicator includes: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises; and the relationship mining module is used for mining the enterprise group relationship according to the key indexes.

According to another aspect of the embodiment of the invention, an electronic mining device for enterprise group relations is provided.

An electronic device for mining corporate group relationships, comprising: one or more processors; the storage device is used for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors realize the mining method of enterprise group relations provided by the embodiment of the invention.

According to yet another aspect of embodiments of the present invention, a computer-readable medium is provided.

A computer readable medium, on which a computer program is stored, when being executed by a processor, implements the mining method for enterprise group relations provided by the embodiments of the present invention.

One embodiment of the above invention has the following advantages or benefits: obtaining enterprise related data; processing the enterprise related data to obtain key indexes, wherein the key indexes comprise: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises; the enterprise group relationship is mined according to the key indexes, so that the enterprise group relationship is mined based on the consistent action relationship of natural people or enterprises, stock holders of the enterprises and actual control people of the enterprises, the mined enterprise group relationship data is full, and the group number and the enterprise coverage range are greatly improved; in addition, the method is used for mining enterprise group relations, manpower is not needed, time and labor are saved, and timeliness of mining data is improved conveniently.

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 illustrating the main steps of a mining method for enterprise group relationships according to an embodiment of the present invention;

FIG. 2 is a schematic diagram illustrating an implementation principle of the group relation mining according to the embodiment of the present invention;

FIG. 3 is a schematic diagram of an implementation of a natural person fusion algorithm according to an embodiment of the invention;

FIG. 4 is a schematic diagram of an implementation of a natural person fusion algorithm according to another embodiment of the present invention;

FIG. 5 is a schematic diagram of an implementation of a natural person fusion algorithm according to yet another embodiment of the present invention;

FIG. 6 is a schematic diagram illustrating an implementation of a natural human relationship determination algorithm according to an embodiment of the present invention;

FIG. 7 is a schematic diagram of the determination of the actual controller of the enterprise in accordance with one embodiment of the present invention;

FIG. 8 is a schematic diagram of the determination of the actual controller of an enterprise according to another embodiment of the present invention;

FIG. 9 is a model diagram of clique relationship mining, according to an embodiment of the invention;

FIG. 10 is a schematic diagram of the main modules of an enterprise corporate linkage mining apparatus according to an embodiment of the present invention;

FIG. 11 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;

fig. 12 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.

With the development of technologies such as big data, artificial intelligence and knowledge maps, in order to better further assist the transformation development of business and exert the value of the knowledge maps and the big data in the fields of public wind control and marketing of banks, technical means are used for automatically constructing multi-level association relationship maps among enterprises through data such as investment relationship, high-management personnel relationship and the like, deeply mining and verifying various association relationships among the enterprises and providing a basis for subsequent marketing application and risk control.

Therefore, the invention discloses a map mining model based on a knowledge map, which is used for mining key indexes such as natural human relatives, enterprise consistent action relations, enterprise stock control shareholders, enterprise actual controllers and the like based on the core principle of control rights according to investment, stock right, high-level management personnel and corporate data and finally mining the group relations of the enterprises.

Based on the mined group relationships, marketing and risk scenarios can be applied. Such as: members of non-credit clients in the group can be used as credit targets to carry out marketing push; and when some members in the group have risk events, the wind control supervision and early warning on other members in the group can be enhanced, and the risk source spread can be prevented.

A knowledge graph is essentially a semantic network, a graph-based data structure consisting of nodes and edges. In the knowledge-graph, each node represents an "entity" existing in the real world, and each edge is a "relationship" between entities.

The enterprise-level knowledge graph is actually the graph of an industry field falling on the ground of an enterprise, and the graph falling on the ground of a commercial bank is mainly based on big data and artificial intelligence related technologies, so that modeling is performed on the existing entities (customers, accounts, equipment, cards and the like) and relations (account transfer, stock holding, guarantee and the like) of the commercial bank and the attributes of the entity relations, and finally the enterprise-level professional field graph of the whole set of commercial bank is formed.

There is no "group" in the "official act", which is only the proposition of the company with limited responsibility and the company with limited shares. Some companies carry out diversified operation strategies, corresponding subsidiary companies are established in a plurality of fields, and thus, the subsidiary companies and the parent companies form an enterprise group due to the relationship of 'blooding margin'.

The application of the enterprise group in the commercial bank is as follows: for such an enterprise group, a business bank does not individually give credit to a certain subsidiary company, but generally gives credit to the same group, and the group risk is avoided, which is called a credit granting group. Meanwhile, similar product recommendations are made to businesses in these groups during marketing.

The enterprise group mining method is based on enterprise-level knowledge map data constructed by commercial banks through big data and artificial intelligence technology, and a hidden enterprise group is mined from the enterprise-level knowledge map data, so that the defects of insufficient coverage and timeliness check caused by manually determining enterprise association relation in the prior art are overcome.

The data source of the invention is mainly based on enterprise knowledge maps constructed by intelligent banks, the technical platform is based on a big data platform, a distributed computing framework uses Spark Graph12 for computing, the mining process mainly comprises the steps of mining natural human relationship, fusing relationship, mining enterprise consistent action relationship, enterprise stock control stakeholders, actual control persons and other key indexes according to investment, stock right, high management personnel and legal personnel data and based on the core principle of control right, and finally mining enterprise affiliated group relationship. Specifically, the enterprise group mining method of the present invention, when researching, may include the following steps:

1. determining a mining target:

based on the existing enterprise knowledge graph, the method carries out implicit knowledge mining and completes model research, verification and deployment of knowledge graph models of enterprise consistency actors, enterprise actual controllers, enterprise stock holders, enterprise groups and the like. Implicit knowledge refers to knowledge that cannot be directly seen through a graph, such as multi-level stock control. Implicit knowledge can be mined through expert rules or algorithms;

2. and (3) mining and modeling data processing of group relations:

the current data and environment are deeply researched and analyzed to know whether the data meet the project requirements or not, and the data quality is evaluated. Evaluating the data quality from the viewpoints of repeated data, missing values, abnormal values, data ranges and the like;

according to the determined design parameters (namely, parameters which need to be input for program implementation, data tables or data fields which need to be mined and the like), data extraction is carried out aiming at model development, and data in a data warehouse are processed into a required format;

and cleaning the data, wherein the cleaning is mainly based on logic judgment, business experience and related business rules, and simultaneously, analyzing and processing some data which do not meet the standard. For example, for the case that a certain enterprise has more than 100% shares in stock, the rule is made to modify or discard;

3. designing and developing a group relationship mining model:

designing a group relation mining algorithm, holding up a model design conference, discussing a design scheme of a project, discussing with a service expert and a verification team and determining details of model development. Subdivision scheme determination, to reduce the impact of outliers on corporate relationship mining, it is common to process outlier data (e.g., data like a business having multiple bank client ids) prior to modeling. The final mining result is not a simple statistical decision, and the business cognition and the data understanding influence the form of the final model to a great extent;

4. and (5) result verification:

and sampling verification is performed by a professional, and the existing mining result, business experience cognition and other conditions are compared to ensure the effect of the model.

Fig. 1 is a schematic diagram illustrating main steps of a mining method for enterprise group relationships according to an embodiment of the present invention. As shown in fig. 1, the mining method for corporate linkage according to the embodiment of the present invention mainly includes steps S101 to S103 as follows.

Step S101: acquiring enterprise related data;

step S102: processing the enterprise related data to obtain key indexes, wherein the key indexes comprise: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises;

step S103: and mining the enterprise group relationship according to the key indexes.

According to the technical scheme of the invention, the enterprise related data can be acquired from the enterprise knowledge map established in advance. The enterprise-related data includes, for example, the name of the enterprise, the equity relationship of the enterprise, the status of the enterprise, information about highly managed persons of the enterprise, identification codes of the enterprise, information about natural human entities included in the enterprise, and the like. After the enterprise-related data is obtained, the enterprise-related data can be processed to extract key indexes for enterprise group relationship mining.

According to the technical scheme of the invention, before the enterprise-related data is processed, the enterprise-related data can be preprocessed, which specifically comprises the following steps:

uploading enterprise related data to a distributed system framework for processing, wherein the enterprise related data comprises an enterprise name, a stock right relationship of an enterprise and an enterprise state;

removing the weight of the enterprise according to the enterprise name and the equity relationship of the enterprise;

and filtering the enterprise according to the enterprise state and the enterprise naming rule so as to preprocess the enterprise related data.

Specifically, in the embodiment of the present invention, the data is preprocessed as follows:

1. uploading original text data to a specified path of a distributed system framework (HDFS);

2. aiming at the condition that one enterprise name in the enterprise information data corresponds to a plurality of enterprise identifiers uuid (universal Unique Identifier) or client identifiers id (historical data is not cleared), enterprise disambiguation needs to be carried out on the data. Then, filtering out repeated uuid/client id of the enterprise name; if data of one enterprise name corresponding to a plurality of enterprise uuids/client ids still exist, taking the data of which the enterprise identification code is not empty;

3. filtering enterprises with cancelled and cancelled states;

4. enterprises which do not belong to any enterprise group, such as 'State funding Commission', 'State Council', and the like, appear in the enterprise names are filtered according to the enterprise naming rules.

After the enterprise related data is preprocessed, the preprocessed data can be further processed to obtain set key indexes.

According to the embodiment of the present invention, when processing the enterprise-related data to obtain the key index, the following steps may be specifically performed:

carrying out first processing on the enterprise related data to obtain the fusible relationship of natural people and the relatives of the natural people;

and performing second processing on the enterprise related data to obtain a key index based on the fusible relationship of the natural person and the relative relationship of the natural person.

Fig. 2 is a schematic diagram of an implementation principle of the group relation mining according to the embodiment of the present invention. As shown in fig. 2, when performing the group relationship mining, by processing the enterprise related data, it is required to mine the natural person fusion relationship of the enterprise, the relationship of the natural persons, the consistent action relationship of the enterprise or the natural persons, the stock holder of the enterprise and the actual controller of the enterprise, wherein the stock holder of the enterprise can be calculated according to the investment proportion and the consistent action relationship of the enterprise; and the stock control relationship among the enterprises can be mined according to the investment proportion, so that the method is used for mining the actual controller of the enterprises. And finally, performing group relation mining according to the consistent action relation of the enterprises or natural persons, the actual control persons of the enterprises and the stock holders of the enterprises, wherein the importance of the enterprises in the group can be calculated according to the investment proportion during the group relation mining, and further the core enterprises of the group are determined.

The following describes a specific implementation process when enterprise-related data processing is performed in the embodiment of the present invention with reference to the drawings.

When determining key indexes such as stockholder relationship of an enterprise, actual controller of the enterprise, and consistent action relationship of the enterprise or natural people, natural people fusion is first required so that relevant data of the same natural people can be acquired more accurately, for example: when the unique identification code does not exist, how to judge that the natural people with the same name are fusible is carried out, and then the fusible natural people are treated as the same natural people. According to the technical scheme of the invention, the fusible relationship of the natural person can be determined by the following modes:

if two enterprises have a plurality of high-rate managers or natural person shareholders with the same name, every two high-rate managers or natural person shareholders with the same name are in a interfusible relationship;

if two enterprises have an investment relationship and have the same-name high-rate manager or natural person shareholder, the same-name high-rate manager or natural person shareholder is in a confldable relationship;

if the names of two enterprises are similar and the two enterprises have the same name of high-speed manager or natural person shareholder, the same name of high-speed manager or natural person shareholder is in a confldable relationship.

Fig. 3 is a schematic diagram of an implementation principle of a natural person fusion algorithm according to an embodiment of the present invention. As shown in fig. 3, it is shown that if enterprise a and enterprise B have two or more high-pipe or natural human stockholders with the same name, each two high-pipe or natural human stockholders with the same name are in a interfusible relationship with each other.

Fig. 4 is a schematic diagram of an implementation principle of a natural human fusion algorithm according to another embodiment of the present invention. As shown in fig. 4, it is shown that if enterprise a and enterprise B have an investment relationship and a and B have a high pipe or natural human shareholder of the same name, the two high pipe or natural human shareholder of the same name are in a fusible relationship.

Fig. 5 is a schematic diagram of an implementation principle of a natural human fusion algorithm according to another embodiment of the present invention. As shown in fig. 5, it is shown that if the names of the business a and the business B are similar (e.g., similarity is determined according to the number or ratio of repeated words in the names, etc.), and a and B have a same-name high pipe or natural human shareholder, the two same-name high pipe or natural human shareholder are in a confldable relationship.

In an embodiment of the present invention, a natural person fusion algorithm is also presented, with the output fields of the algorithm being shown in table 1 below, for example.

TABLE 1

Attribute name	Attribute definition
		attr	Nature of natureHuman defined fusible relationship attribute description
src_uuid	Starting point (natural human entity uuid)
		dst_uuid	Terminal point (natural human entity uuid)
data_dt	Date of business data

After the fusible natural persons are determined, the relatives of the natural persons may also be determined, "relatives" referring to the relatives between the natural person entities in the enterprise graph. According to the technical scheme of the invention, the relationship of the natural people can be determined by the following modes:

obtaining an enterprise range according to the invested relation of the enterprise stockholders or enterprises;

taking high-management personnel or stockholders included in each enterprise within the enterprise range as selected natural people;

and judging the relationship of the natural person according to the name of the natural person.

In one embodiment of the present invention, the determination process of the relationship is, for example, as follows:

tracing back 3 layers according to the external investment relation of enterprise shareholders or enterprises to define the range of the enterprises;

selecting a natural person range according to the high-ranking or natural person shareholder relationship of the enterprise in a defined enterprise range;

for English names, finding out natural human entities with the same surnames, and adding relatives;

for Chinese names, all natural person entity entities with same names and different names with the length of more than or equal to 3 words are found, and the relationship of relatives is added.

Fig. 6 is a schematic diagram illustrating an implementation principle of a natural human relationship determination algorithm according to an embodiment of the present invention. As shown in fig. 6, in this embodiment, jeb bush and george bush are relatives with each other; the WangYiwen, WangErwen, WangSanwen, WangSiwen and WangWuwen are relatives.

In an embodiment of the present invention, a relationship determination algorithm is also proposed, and the output fields of the algorithm are shown in table 2 below, for example.

TABLE 2

Attribute English name	Attribute service definition
		attr	Description of relationship attributes
src_uuid	Starting point (natural human entity uuid)
		dst_uuid	Terminal point (natural human entity uuid)
data_dt	Date of business data

After the fusion relationship of the natural people and the relatives of the natural people are determined, natural people entities or enterprise entities which are in consistent action relationship with each other can be mined. In an embodiment of the present invention, the coherency action relationship is defined as: a relationship where a set of consistent entities (two or more natural persons or businesses) may take a consistent action with respect to a business entity or anything. Before mining the consistent action relationship, the construction of investment and high-management relationship is acquired according to statistical enterprise related data, the mining of the fusion relationship of natural people and the relatives of the natural people is completed, and then the relationship is packaged in a GraphX form for data calling.

Consistent action relationships of natural persons or businesses may be determined according to embodiments of the present invention by:

and determining the consistent action relationship of the natural people or the enterprises based on a set rule according to the relationship of the natural people to the relatives, the stock control relationship of the natural people to the enterprises, the stock holding information of the two natural people or the enterprises to the third enterprise and the high-management personnel information of the enterprises, wherein the stock control relationship of the natural people to the enterprises is obtained according to the stock right relationship of the enterprises.

In an embodiment of the present invention, mining the consensus action relationship may be performed according to the rules shown in Table 3 below.

TABLE 3

In table 3, the stock control ratio is used as a configurable parameter, the default value is greater than 50%, and other values can be set according to the requirement. Wherein board of high refers to a high manager of the enterprise.

In an embodiment of the present invention, the output fields of an algorithm for determining a consensus action relationship of a natural person or business are shown in table 4 below, for example.

TABLE 4

Attribute English name	Attribute service definition
		attr	Consistent action relationship attribute description
target	Consistent action relationship targets
		rule	Rule of action of agreement
src_uuid	Starting point (natural human entity uuid)
		dst_uuid	Terminal point (natural human entity uuid)
data_dt	Date of business data

Wherein the value of target is the MD5 value of the enterprise name of the consistent action object, or is all (when the consistent action object is anything); rule refers to the Rule number in table 3, and takes the value "Rule 3" as generated according to Rule 3.

According to one embodiment of the invention, the holding stockholders of the enterprise are determined by:

if the enterprise is a listed company, acquiring the stock holder of the enterprise through the latest annual report of the enterprise;

and if the enterprise is a non-listed company, taking the stockholder combination with the stock control proportion larger than the set limit value as the stock control stockholder of the enterprise.

Before carrying out stock control and stockholder mining on an enterprise, investment and construction of a high-management relationship are required to be obtained according to statistical enterprise related data, mining of a natural person fusion relationship, a natural person relative relationship and a consistent action relationship is completed, and then the relationship is packaged in a GraphX form for data calling.

According to the official act, the term "stock control" means that the capital of the stock control accounts for more than fifty percent of the total capital of the company with limited liability or the stock control accounts for more than fifty percent of the total stock of the company with limited stocks; although the percentage of the capital or held shares is less than fifty percent, the votes on the capital or held shares are sufficient to have a significant impact on the resolution of the stockholder's meeting or great stock meeting. In the embodiment of the invention, the stock holder of the determined enterprise may have one stock holder (which may be a natural person or an enterprise entity), may have multiple stock holders with a consistent action relationship, and may not have the stock holder.

In one embodiment of the invention, an algorithm is provided for determining stockholders for a business, in which the stock holding proportion defaults to greater than or equal to 50%. The rule for determining the stock holder of the enterprise realized by the algorithm is as follows:

1) for a listed company, manually labeling the latest annual newspaper of the enterprise, and searching a keyword 'stock control shareholder' to obtain the stock control shareholder of the enterprise;

2) for non-public companies, when using the algorithm, the determined fusible natural human entity is fused, and then the following steps are carried out:

(1) if only one shareholder exists, the only shareholder is used as a stock control shareholder;

(2) if there are more than one stockholders, the stockholder with the stock holding ratio larger than the stock control ratio is used as the stock control stockholder. If no stockholder larger than the stock control ratio exists, merging the stockholders according to the consistent action relationship and the relative relationship (adding the stock holding ratios), and if a stockholder combination larger than the stock control ratio exists, combining the stockholder combination into the stock control stockholder;

(3) otherwise, there is no stockholder.

In an embodiment of the present invention, the output fields of the determination algorithm for the holding stockholders of the enterprise are shown in table 5 below, for example.

TABLE 5

Attribute name	Attribute service definition
		src_uuid	Starting point (natural person/enterprise entity uuid)
dst_uuid	Terminal (Enterprise entity uuid)
		attr	Shareholder relationship attribute description
ratio	Controlling the ratio of strands
		data_dt	Date of business data

And finally, excavating actual controllers of the enterprise. After the mining of the fusion relationship, the relative relationship, the consistent action relationship and the stock holder of the natural people is completed, the construction of investment and high-management relationship is obtained according to the statistical enterprise related data, and then the relationship is packaged in a GraphX form for data calling.

An actual controller is a person who, although not necessarily a stakeholder of a company, can actually govern the behavior of the company through investment relations, agreements, or other arrangements. In short, the actual control person is a natural person or an enterprise of the actual control company. Because information such as control protocols/actual governance inside an enterprise is difficult to obtain, the invention infers an actual control relationship from the perspective of stock control. The actual controller performs a deeper excavation in the case where the stock holder of the enterprise cannot be found or the stock holder is a plurality of stock holders acting in unison. According to the technical scheme of the invention, the actual controller of the enterprise may be one natural person or enterprise, or may be a plurality of natural persons or enterprises, or may not have the actual controller.

According to one embodiment of the invention, the actual controller of the enterprise is determined by:

if the enterprise is a listed company, acquiring an actual controller of the enterprise through the latest annual report of the enterprise;

if the enterprise is a non-listed company, respectively determining the actual controller of the enterprise according to the enterprise type of the enterprise, including:

when the enterprise type is individual, taking a legal representative of the enterprise as an actual controller of the enterprise;

when the enterprise type is non-individual and the only stock control shareholder exists, the only stock control shareholder is used as an actual controller of the enterprise;

when the enterprise type is non-individual and has no unique stock holder, the level of the investment enterprises or the natural persons of the enterprise is determined according to the investment relation among the enterprises, then the accumulation of the stock holding proportion of the investment enterprises or the natural persons is carried out on each layer of the investment enterprises or the natural persons in turn according to the relationship of the natural persons and the consistent action relation of the natural persons or the enterprises in the order from small to large, and the combination of the investment enterprises or the natural persons with the stock holding proportion exceeding the set proportion for the first time is used as the actual controller of the enterprise.

In one embodiment of the invention, an algorithm is provided for determining the actual controller of the enterprise, which implements the determination rule for the actual controller of the enterprise as follows:

1) for a listed company, the actual controller may be exposed in the yearbook, which is accomplished by way of manual notes;

2) for non-public companies, when using an algorithm to determine the actual controller of a business, a natural human entity that is determined to be fusible is fused, and then the following operations are performed:

(1) if the enterprise type is individual, selecting a legal representative of the enterprise as an actual controller of the enterprise;

(2) if the enterprise type is non-individual and only one stockholder exists, the actual controller is the stockholder;

(3) if the enterprise type is non-individual and the enterprise has a plurality of stock control stockholders or does not have stock control stockholders, setting a tracing layer number to be 3, starting from the 1 st layer, tracing to the 3 rd layer, developing according to the investment relationship, and every tracing layer is 1:

firstly, merging and adding shares of all natural people or enterprise entities according to the relationship of relatives and consistent action;

then, as long as an entity or entity combination with stock exceeding 50% is counted, the current search can be exited, and the entity/entity combination is taken as the actual controller. When the set tracing layer number 3 is reached, if the entity/entity combination with stock holding more than 50% does not exist, no actual controller exists.

FIG. 7 is a schematic diagram of the determination of the actual controller of the enterprise in accordance with one embodiment of the present invention. As shown in FIG. 7, in this embodiment, Enterprise A is a non-individual type and there is no unique stockholder. Because the enterprise B and the enterprise C are enterprises in a consistent action relationship with each other and have a common unique shareholders Wang II, the actual controller of the enterprise A is Wang II.

Fig. 8 is a schematic diagram of the determination principle of an enterprise real controller according to another embodiment of the present invention. As shown in FIG. 8, in this embodiment, Enterprise A is a non-individual type and there is no unique stockholder. Because enterprise E and enterprise F are mutually the same actors, and enterprise E and enterprise F are the only stockholders of enterprise B and enterprise C, respectively, enterprise E and enterprise F are the actual controllers of enterprise a together.

In an embodiment of the present invention, the output fields of the algorithm for determination of the actual controller of the enterprise are shown, for example, in table 6 below.

TABLE 6

In one embodiment of the present invention, for a rule for stock holding, where:

rule 1: the individual enterprise actually controls the artificial legal person, and if no legal person exists, no actual control person exists (the corresponding depth is 0);

rule 2: the method does not use the combination of the relationship of relatives and the relationship of consistent action, and the stock holding of the limited company or natural people is more than half (excavation is carried out within 5 layers, depth is less than or equal to 5);

rule 3: using the relationship combination, the stock holding of the limited company or the natural person is more than half (digging within 5 layers, depth is less than or equal to 5);

rule 4: using concordant action relationship consolidation, limited or natural people hold more than half (within 5 tiers to dig, depth is less than or equal to 5).

The key indexes can be obtained by processing the enterprise-related data according to the introduction. And then, mining the enterprise group relationship according to the key indexes. According to the embodiment of the present invention, when mining the enterprise group relationship according to the key indexes, the mining method specifically includes:

determining enterprises brought into the group according to the consistent action relationship of natural people or enterprises, the stock holding ratio of stock holders of the enterprises and the actual controller of the enterprises respectively;

calculating the importance score of each enterprise according to the relationship among the enterprises in the group;

and determining the core enterprises of the group according to the importance score of each enterprise so as to establish enterprise group relationship.

An enterprise group is an enterprise group with a consistent interest relationship formed around a central enterprise, and other enterprises are often the layouts of the central enterprise in various fields, and the central enterprise is the core enterprise of the enterprise group.

In an embodiment of the present invention, an algorithm for determining an enterprise group relationship is provided, and the rule for determining the enterprise group relationship implemented by the algorithm is as follows:

1. and (3) control association:

(1) directly controlling: the external direct stock holding ratio is more than 50%, and the investment objects (enterprises) are brought into the group;

(2) is directly controlled: one layer of shareholders with the share holding ratio more than 50 percent are brought into the group;

(3) first large stockholder: one layer of shareholders with a stock holding ratio of more than 30% and less than 50%, but the shareholders (enterprises/natural people) with the highest stock holding ratio are brought into the group;

(4) two stockholders each accounting for 50%, wherein a party (either a director of his own or a superior or subordinate related enterprise) who is a director or a general manager is brought into the group;

(5) indirect control: the external holding proportion is more than 50%, the number of investment layers is more than or equal to 2, and investment objects (enterprises) are brought into the group;

(6) is indirectly controlled: two or more layers of shareholders, wherein the total holding ratio is more than 50 percent, and the shareholders (enterprises/natural people) are brought into the group;

(7) and (3) parent-offspring joint control: the direct or indirect stock holding ratio of a plurality of natural persons who are relatives is more than 50 percent, and then the natural persons are brought into the group;

2. and (3) personnel association:

bringing enterprises with more than 2 common core high-management personnel into a group;

3. controlled by a third party:

and the system is directly or indirectly controlled by a common third party, and can be automatically brought into a group after the control association is expanded. And each layer of expansion is 1, new nodes are brought into the group, and further expansion is carried out according to the newly brought nodes. When the expansion is carried out, the expansion is stopped if government agencies such as national resource committee, the national government and the like are met.

In an embodiment of the present invention, the output fields of the mining algorithm for enterprise group relationships are shown in table 7 below, for example.

TABLE 7

The importance of the core enterprise of the enterprise group is the highest, and the page level pagerank calculated in the distributed graph processing framework is the highest. In the specific determination, the investment relationship is regarded as a directed graph, and iterative computation is performed based on two assumptions, and the method is as follows:

(1) companies that invest in many companies are important, with significant impact within the clique (initial assumptions of iteration);

(2) companies that invest in important companies are important (iterative recursive assumption, passing company weights).

According to the embodiment of the invention, after the enterprise group relationship is mined according to the key indexes, an enterprise group mining model can be further constructed to mine the enterprise group relationship by using the enterprise group mining model.

FIG. 9 is a model diagram of clique relationship mining, in accordance with an embodiment of the present invention. As shown in FIG. 9, starting from Enterprise A, enterprises B-H and natural people "Zhang" are included in the same group because:

(1) b is the common high pipe related party of the enterprise A, and 'Zhang' is the corresponding high pipe;

(2) c is the direct control object of A;

(3) d is the largest shareholder of A;

(4) e is the indirect control object of A;

(5) f is an indirect control shareholder of A;

(6) g and A are both directly or indirectly controlled by F;

(7) h and A are both directly or indirectly controlled by D;

(8) two 'li' family names, natural people, are in a relationship of relatives, and control the enterprise a together.

Since the business side already has a group identified according to the accumulated business experience, but the group size is small, in order to combine the excavated enterprise group relationship with the existing group data for fusion, a group with a larger range is formed, which is the key of the enterprise group excavation model.

In the embodiment of the present invention, the process of constructing the enterprise group mining model specifically includes:

acquiring existing group data;

merging the mined enterprise group data with the existing group data, and constructing a directed graph by using a distributed graph processing framework;

all enterprises in each connected block of the directed graph are used as an enterprise group, and core enterprises are selected from each connected block to be used as core enterprises of the enterprise group, so that an enterprise group mining model is constructed.

The algorithm for constructing the enterprise group mining model in one embodiment of the invention mainly comprises the following steps:

1. reading the existing group data from a database (such as Hive), and performing data preprocessing: the method comprises the following steps that data of an enterprise belonging to a plurality of groups exist, and the group with the largest scale is taken as the group to which the enterprise belongs;

2. acquiring newly added group side data according to the previously mined enterprise group data, and performing union operation on the newly added group side data and the existing group data to acquire total side data so as to construct a directed graph G;

3. calculating a connected branch of the directed graph G, and calculating an enterprise with the largest pagerank value in each connected block as a core enterprise;

4. all enterprises in a connected block form a new enterprise group, and the group fuses existing group data and newly added group data.

In an embodiment of the present invention, a mining algorithm for an enterprise clique mining model is constructed, with output fields such as those shown in table 8 below.

TABLE 8

And finally, after enterprise group relation mining is carried out, verification can be carried out on the existing enterprise relation knowledge graph data. The fact proves that when the enterprise group relationship mining method is used for enterprise group relationship mining, the whole group data volume and the number of clustered enterprises are doubled on the original data number.

For example: for an existing group A, 11 identified enterprises are under the group, a group B taking the enterprise a as a core enterprise is excavated through the enterprise group excavation model, the group B comprises 13 enterprises, 9 enterprises are the same as the enterprises in the group A, and a new enterprise group C obtained through the enterprise group excavation model comprises 15 enterprises, wherein 13 enterprises in the original identified group A and the rest 2 enterprises are newly integrated into the group. And verification proves that the newly added 2 enterprises really have actual relationship with the enterprises in the group A.

According to another aspect of the invention, an excavating device for enterprise group relations is also provided.

Fig. 10 is a schematic diagram of main modules of an enterprise group relationship mining device according to an embodiment of the present invention. As shown in fig. 10, the mining apparatus 1000 for enterprise group relationships according to the embodiment of the present invention mainly includes a data obtaining module 1001, a data processing module 1002, and a relationship mining module 1003.

A data acquisition module 1001 configured to acquire enterprise-related data;

a data processing module 1002, configured to process the enterprise-related data to obtain key indexes, where the key indexes include: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises;

and the relationship mining module 1003 is configured to mine the enterprise group relationship according to the key index.

According to an embodiment of the present invention, the relationship mining module 1003 may further be configured to:

determining enterprises brought into the group according to the consistent action relationship of the natural people or the enterprises, the stock holding ratio of stock holders of the enterprises and the actual controllers of the enterprises respectively;

and determining the core enterprises of the clique according to the importance score of each enterprise so as to establish enterprise clique relationship.

According to another embodiment of the present invention, the mining apparatus 1000 for enterprise group relations may further include a data preprocessing module (not shown in the figure) for:

uploading enterprise-related data to a distributed system framework for processing before processing the enterprise-related data, wherein the enterprise-related data comprises an enterprise name, a share relationship of an enterprise and an enterprise state;

removing the weight of the enterprise according to the enterprise name and the share right relation of the enterprise;

and filtering the enterprise according to the enterprise state and the enterprise naming rule so as to preprocess enterprise related data.

According to yet another embodiment of the present invention, the data processing module 1002 may be further configured to:

performing first processing on the enterprise related data to obtain a fusible relationship of natural people and a relationship of relatives of the natural people;

and performing second processing on the enterprise related data to obtain the key index based on the fusible relationship of the natural person and the relative relationship of the natural person.

According to yet another embodiment of the present invention, the fusible relationship of the natural person is determined by:

if two enterprises have an investment relationship and the two enterprises have the same-name high-rate manager or natural human shareholder, the same-name high-rate manager or natural human shareholder is in a confldable relationship;

if the names of the two enterprises are similar and the two enterprises have the same-name high-speed manager or natural person shareholder, the same-name high-speed manager or natural person shareholder is in a fusible relationship.

According to a further embodiment of the invention, the natural person's relativity is determined by:

taking a high-management person or a stock holder included in each enterprise in the enterprise range as a selected natural person;

According to yet another embodiment of the present invention, the consensus action relationship of the natural person or business is determined by:

determining the consistent action relationship of the natural people or the enterprises based on a set rule according to the relationship of the natural people to the relatives, the stock control relationship of the natural people to the enterprises, the stock holding information of two natural people or the enterprises to a third enterprise and the high-management personnel information of the enterprises, wherein the stock control relationship of the natural people to the enterprises is obtained according to the stock right relationship of the enterprises.

According to yet another embodiment of the present invention, the holding stockholders of the enterprise are determined by:

According to yet another embodiment of the invention, the actual controller of the enterprise is determined by:

if the enterprise is a listed company, acquiring an actual controller of the enterprise through a latest enterprise annual report;

if the enterprise is a non-listed company, respectively determining the actual controllers of the enterprise according to the enterprise type of the enterprise, wherein the steps comprise:

when the type of the enterprise is individual, taking a legal representative of the enterprise as an actual controller of the enterprise;

when the enterprise type is non-individual and a unique stock holder exists, taking the unique stock holder as an actual controller of the enterprise;

when the enterprise type is non-individual and has no unique stock holder, determining the level of the investment enterprises or natural persons of the enterprise according to the investment relationship among the enterprises, then sequentially accumulating the stock holding proportion of the investment enterprises or natural persons for each layer of the investment enterprises or natural persons according to the relationship of the natural persons and the consistent action relationship of the natural persons or enterprises in the order from small to large of the level, and taking the combination of the investment enterprises or natural persons with the stock holding proportion exceeding the set proportion for the first time as the actual controller of the enterprise.

According to another embodiment of the present invention, the mining apparatus 1000 for enterprise corporate linkage may further include a model building module (not shown in the figure) for:

after mining the corporate group relationships according to the key indexes, an corporate group mining model is constructed to mine corporate group relationships using the corporate group mining model.

According to yet another embodiment of the invention, the model building module (not shown in the figures) may be further configured to:

acquiring existing group data;

and taking all enterprises in each connected block of the directed graph as an enterprise group, and selecting core enterprises from each connected block as the core enterprises of the enterprise group to construct an enterprise group mining model.

According to the technical scheme of the embodiment of the invention, enterprise related data is obtained; processing the enterprise related data to obtain key indexes, wherein the key indexes comprise: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises; the enterprise group relationship is mined according to the key indexes, so that the enterprise group relationship is mined based on the consistent action relationship of natural people or enterprises, stock holders of the enterprises and actual control people of the enterprises, the mined enterprise group relationship data is full, and the group number and the enterprise coverage range are greatly improved; in addition, the method is used for mining enterprise group relations, manpower is not needed, time and labor are saved, and timeliness of mining data is improved conveniently.

Fig. 11 shows an exemplary system architecture 1100 of an enterprise corporate relationship mining method or an enterprise corporate relationship mining apparatus to which an embodiment of the present invention may be applied.

As shown in fig. 11, the system architecture 1100 may include

terminal devices

1101, 1102, 1103, a network 1104, and a server 1105. The network 1104 is a medium to provide communication links between the

terminal devices

1101, 1102, 1103 and the server 1105. Network 1104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

A user may use

terminal devices

1101, 1102, 1103 to interact with a server 1105 over a network 1104 to receive or send messages or the like. Various messaging client applications, such as shopping applications, web browser applications, search applications, instant messaging tools, mailbox clients, social platform software, etc. (examples only) may be installed on the

terminal devices

1101, 1102, 1103.

The

terminal devices

1101, 1102, 1103 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 1105 may be a server that provides various services, such as a backend management server (for example only) that provides support for shopping-like websites browsed by users using the

terminal devices

1101, 1102, 1103. 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 mining method for corporate linkage provided by the embodiment of the present invention is generally executed by the server 1105, and accordingly, the mining device for corporate linkage is generally provided in the server 1105.

It should be understood that the number of terminal devices, networks, and servers in fig. 11 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Referring now to FIG. 12, shown is a block diagram of a computer system 1200 suitable for use with a terminal device or server implementing an embodiment of the present invention. The terminal device or the server shown in fig. 12 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. 12, the computer system 1200 includes a Central Processing Unit (CPU)1201, which can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)1202 or a program loaded from a storage section 1208 into a Random Access Memory (RAM) 1203. In the RAM 1203, various programs and data necessary for the operation of the system 1200 are also stored. The CPU 1201, ROM 1202, and RAM 1203 are connected to each other by a bus 1204. An input/output (I/O) interface 1205 is also connected to bus 1204.

The following components are connected to the I/O interface 1205: an input section 1206 including a keyboard, a mouse, and the like; an output portion 1207 including a display device such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 1208 including a hard disk and the like; and a communication section 1209 including a network interface card such as a LAN card, a modem, or the like. The communication section 1209 performs communication processing via a network such as the internet. A driver 1210 is also connected to the I/O interface 1205 as needed. A removable medium 1211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is mounted on the drive 1210 as necessary, so that a computer program read out therefrom is mounted into the storage section 1208 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 section 1209, and/or installed from the removable medium 1211. The computer program performs the above-described functions defined in the system of the present invention when executed by the Central Processing Unit (CPU) 1201.

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 units or modules described in the embodiments of the present invention may be implemented by software, or may be implemented by hardware. The described units or modules may also be provided in a processor, and may be described as: a processor includes a data acquisition module, a data processing module, and a relationship mining module. Where the names of such units or modules do not in some cases constitute a limitation on the units or modules themselves, for example, the data acquisition module may also be described as a "module for acquiring enterprise-related data".

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: acquiring enterprise related data; processing the enterprise-related data to obtain key indicators, wherein the key indicators include: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises; and mining the enterprise group relationship according to the key indexes.

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 mining method of enterprise group relationship is characterized by comprising the following steps:

acquiring enterprise related data;

processing the enterprise-related data to obtain key indicators, wherein the key indicators include: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises;

and mining the enterprise group relationship according to the key indexes.

2. The method of claim 1, wherein mining corporate group relationships according to the key indicators comprises:

3. The method of claim 1, further comprising, prior to processing the enterprise-related data:

uploading enterprise-related data to a distributed system framework for processing, wherein the enterprise-related data comprises an enterprise name, a stock right relationship of an enterprise and an enterprise state;

4. The method of claim 1, wherein processing the enterprise-related data to derive key metrics comprises:

5. The method of claim 4, wherein the fusible relationship of the natural person is determined by:

6. The method of claim 4, wherein the natural person's relativity is determined by:

7. The method of claim 4, wherein the consensus action relationship of the natural person or business is determined by:

8. The method of claim 4, wherein the holding stockholders of the enterprise are determined by:

9. The method of claim 4, wherein the actual controller of the business is determined by:

10. The method of claim 1, further comprising, after mining corporate group relationships based on the key metrics:

and constructing an enterprise group mining model so as to mine enterprise group relations by using the enterprise group mining model.

11. The method of claim 10, wherein constructing an enterprise corporate mining model comprises:

acquiring existing group data;

12. An apparatus for mining enterprise group relationships, comprising:

the data acquisition module is used for acquiring enterprise related data;

a data processing module, configured to process the enterprise-related data to obtain a key indicator, where the key indicator includes: the consistent action relationship of natural people or enterprises, stock holders of enterprises and actual controllers of enterprises;

and the relationship mining module is used for mining the enterprise group relationship according to the key indexes.

13. An electronic device for mining enterprise group 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 1-11.

14. 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 1-11.