An enterprise relationship graph construction method and system

Abstract

The present invention provides a method and system for constructing an enterprise relationship graph, which relates to the field of big data processing. The method includes: determining the basic dimension data and business data of the enterprise to be processed, where the basic dimensions include the registered address of the enterprise, the legal representative of the enterprise, the directors of the enterprise, the supervisors of the enterprise, the names of shareholders and other registered senior executives, and the business data includes the names of investment companies received by the enterprise and the corresponding number of transactions, transaction amounts, and transaction times, as well as the names of other market entities that have contractual relationships with the enterprise and the corresponding number of transactions, transaction amounts, and transaction times; matching the basic dimension data and the business data to obtain enterprise association relationships; storing the enterprise association relationships in a graph database to obtain an enterprise relationship graph. On the basis of associating industrial and commercial information, the present invention quantifies the market transaction behaviors between enterprises, excavates deeper association relationships between enterprises, and provides a more complete enterprise relationship graph from multiple dimensions.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and particularly relates to a method and system for constructing an enterprise relationship graph. Background Art

[0002] Currently, common enterprise relationship graphs are all based on the financing and investment in industrial and commercial registrations to form associated relationships, and it is impossible to deeply construct a more extensive enterprise relationship graph based on the business operations of the legal persons and senior executives of the company, resulting in a relatively single relationship of the queried enterprises and being unable to comprehensively and deeply understand the overall situation of an enterprise.

[0003] In the prior art, in order to more comprehensively and deeply understand the general situation of an enterprise, there are also solutions to propose an enterprise relationship graph based on the query of the legal person of the company. However, since the legal person identity information is not disclosed in the current industrial and commercial registrations, the relationship graph constructed between companies through the legal person is inaccurate and has a large error. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for constructing an enterprise relationship graph, which solves the problem that the enterprise relationship graph obtained only by using simple industrial and commercial relationships currently cannot deeply understand the overall situation of the enterprise. The solution, on the basis of industrial and commercial associations, quantifies the market transaction behaviors between enterprises and infers and excavates deeper relationships between enterprises; provides a more rich and multi-dimensional relationship graph for evaluating the business status and overall situation of an enterprise.

[0005] To achieve the above object, the present invention proposes the following technical solution: A method for constructing an enterprise relationship graph, comprising:

[0006] Determine the basic dimension data of the enterprise to be processed, where the basic dimensions include the registered address of the enterprise, the legal representative of the enterprise, the directors of the enterprise, the supervisors of the enterprise, the names of the shareholders and other registered senior executives;

[0007] Determine the business data of the enterprise to be processed, where the business data is the upstream and downstream market entities and transaction behaviors that have had industrial and commercial behaviors with the enterprise during its operation, including the names of the investment companies received by the enterprise and the corresponding number of transactions, transaction amounts, and transaction times, and the names of other market entities that have a contractual relationship with the enterprise and the corresponding number of transactions, transaction amounts, and transaction times;

[0008] Match the basic dimension data and the business data, and obtain the enterprise association relationship according to the association degree between the enterprise to be processed and the associated market entities;

[0009] Store the enterprise association relationship in a graph database to obtain an enterprise relationship graph.

[0010] Further, the process of matching the basic dimension data and the business data includes: obtaining all associated market entities related to the enterprise to be processed, and determining the enterprise association type between the associated market entity and the enterprise to be processed;

[0011] The enterprise association type includes enterprise shareholding relationship, enterprise market behavior relationship and enterprise close relationship; among them, the enterprise shareholding relationship is an enterprise association relationship obtained by matching based on the shareholding information in the industrial and commercial registration information according to the basic dimension data; the enterprise market behavior relationship is an enterprise association relationship obtained by matching based on the market entities in the business data and the publicly available enterprise business data; the enterprise close relationship is the relationship between each market entity associated by the senior executives of the enterprise in the basic dimension data obtained during the enterprise operation behavior and transaction behavior.

[0012] Further, before matching the basic dimension data and the business data, data preprocessing is also included;

[0013] The data preprocessing is to clean and normalize the basic dimension data and the business data, including standardizing the registered address of the enterprise into three levels of province-city-district, unifying the transaction amount in ten thousands of units, and unifying the format of the transaction time as YYYY.

[0014] Further, the specific process of obtaining the enterprise association relationship is as follows:

[0015] Construct an enterprise relationship network for the enterprise to be processed, and determine the components of the direct association weight value and the indirect association weight value;

[0016] Calculate the direct association weight value, indirect association weight value and final weight value of each associated market entity matched according to the basic dimension data and the business data with this enterprise respectively;

[0017] Judge the range of each weight threshold corresponding to each preset association relationship to which the final weight value belongs, and determine the enterprise association relationship.

[0018] Further, define the component of the direct association weight value as X1, the component of the indirect association weight value as X2, and X1 + X2 = 1. Then the calculation formulas for the direct association weight value, indirect association weight value and final weight value of the enterprise and any of its associated market entities are respectively:

[0019] Direct association weight value = Sum((Number of transactions + Transaction amount) * (1 - (Current year - Transaction year) / N));

[0020] Indirect association weight value = Direct association weight value of the associated market entity * (N - Z) / N + nN;

[0021] Final weight value = direct association weight value * X1 + indirect association weight value * X2;

[0022] Where N is the number of years counted in the enterprise relationship graph, Z is the association level of the enterprise and its associated market entities determined by the shortest path method, and n is the number of senior executives who concurrently hold positions in the enterprise and its associated market entities.

[0023] Furthermore, the method further includes:

[0024] Obtain the enterprise name and enterprise association relationship ID according to the enterprise relationship graph;

[0025] Store the enterprise name and enterprise association relationship ID in the ES index, and provide a quick enterprise name query window;

[0026] Receive a search command using the quick enterprise name query window, query the enterprise association relationship ID corresponding to the search command, and obtain the corresponding enterprise relationship graph for the query.

[0027] Another technical solution of the present invention lies in providing an enterprise relationship graph construction system, which system includes:

[0028] A first determination module, used to determine the basic dimension data of the enterprise to be processed, and the basic dimensions include the registered address of the enterprise, the legal representative of the enterprise, the directors of the enterprise, the supervisors of the enterprise, the names of the shareholders and other registered senior executives;

[0029] A second determination module, used to determine the business data of the enterprise to be processed, and the business data are the upstream and downstream market entities and transaction behaviors that have had industrial and commercial behaviors with the enterprise during the operation process of the enterprise, including the names of the investment companies received by the enterprise and the corresponding number of transactions, transaction amounts, and transaction times, and the names of other market entities that have a contractual relationship with the enterprise and the corresponding number of transactions, transaction amounts, and transaction times;

[0030] A matching module, used to match the basic dimension data and the business data, and obtain enterprise association relationships according to the association degree between the associated market entities and the enterprise to be processed;

[0031] A first storage module, used to store the enterprise association relationships in the graph database to obtain an enterprise relationship graph.

[0032] Furthermore, the process of the matching module obtaining enterprise association relationships is executed by the following execution units, including:

[0033] A construction unit, used to construct an enterprise relationship network for the enterprise to be processed, and determine the components of the direct association weight value and the indirect association weight value;

[0034] A calculation unit for calculating respectively the direct association weight value, the indirect association weight value and the final weight value between each associated market entity matched according to the said basic dimension data and the said business data and the enterprise;

[0035] A determination unit for determining the enterprise association relationship by determining the range of each weight threshold corresponding to each preset association relationship to which the final weight value belongs.

[0036] Furthermore, the formulas for the calculation unit to calculate the direct association weight value, the indirect association weight value and the final weight value between the enterprise to be processed and any one of its associated market entities are respectively:

[0037] Define the component of the direct association weight value as X1, the component of the indirect association weight value as X2, and X1 + X2 = 1, then:

[0038] Direct association weight value = Sum((transaction times + transaction amount) * (1 - (current year - transaction year) / N));

[0039] Indirect association weight value = direct association weight value of the associated market entity * (N - Z) / N + nN;

[0040] Final weight value = direct association weight value * X1 + indirect association weight value * X2;

[0041] Wherein, N is the number of years counted in the enterprise relationship graph, Z is the association level between the enterprise and its associated market entity determined by the shortest path method, and n is the number of senior executives who are concurrently employed in the enterprise and its associated market entity.

[0042] The present invention also provides an enterprise relationship graph construction device, which includes a processor and a memory. A computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the above-mentioned enterprise relationship graph construction method.

[0043] It can be seen from the above technical solutions that the technical solutions of the present invention have obtained the following beneficial effects:

[0044] The method and system for constructing an enterprise relationship graph disclosed by the present invention, the method includes: determining the basic dimension data and business data of the enterprise to be processed, the basic dimensions include the registered address of the enterprise, the legal representative of the enterprise, the directors of the enterprise, the supervisors of the enterprise, the shareholders and the names of other registered senior executives, and the business data includes the names of the investment companies accepted by the enterprise and the corresponding number of transactions, transaction amounts and transaction times, and the names of other market entities having a contractual relationship with the enterprise and the corresponding number of transactions, transaction amounts and transaction times; matching the basic dimension data and the business data, and obtaining the enterprise association relationship according to the association degree between the associated market entity and the enterprise to be processed; storing the enterprise association relationship into a graph database to obtain an enterprise relationship graph. The present invention quantifies the market transaction behaviors between enterprises on the basis of associated industrial and commercial information, excavates deeper association relationships between enterprises, and provides a more complete enterprise relationship graph from multiple dimensions.

[0045] When the method and system of the present invention are implemented, the determination of the enterprise association relationship is obtained by calculating the final weight value which is the sum of the direct association weight value and the indirect association weight value between the enterprise and its associated market entity. The association tightness between the associated market entity and the enterprise is measured by the weight value, and thus can be clearly shown in the enterprise relationship graph; when the user uses the enterprise relationship graph to understand the enterprise, the user can fully understand the overall situation and business status of the enterprise through the rich and multi-dimensional relationship graph, providing more accurate and intuitive data support for measuring the development of the enterprise.

[0046] It should be understood that all combinations of the foregoing concepts and additional concepts described in greater detail below can be regarded as part of the inventive subject matter of the present disclosure as long as such concepts do not contradict each other.

[0047] The foregoing and other aspects, embodiments and features of the teachings of the present invention can be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the present invention such as the features and / or beneficial effects of the exemplary embodiments will be apparent from the following description, or will be learned through the practice of the specific embodiments according to the teachings of the present invention. Brief Description of the Drawings

[0048] The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in each figure may be represented by the same reference numeral. For clarity, not every component is labeled in each figure. Now, embodiments of various aspects of the present invention will be described by way of example and with reference to the drawings, wherein:

[0049] Figure 1 is a structural block diagram for constructing an enterprise relationship graph of the present invention;

[0050] Figure 2 is a flowchart of the method for constructing an enterprise relationship graph of the present invention. Detailed implementation manners

[0051] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only a part rather than all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein shall have the ordinary meanings as understood by those of ordinary skill in the art to which the present invention pertains.

[0052] The "first", "second" and similar terms used in the description and claims of this patent application of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, unless the context clearly indicates otherwise, the singular forms such as "a", "an" or "the" do not limit the quantity, but indicate the existence of at least one. The terms such as "comprising" or "including" mean that the elements or objects appearing before "comprising" or "including" cover the features, wholes, steps, operations, elements and / or components listed after "comprising" or "including", and do not exclude the existence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations. The terms such as "upper", "lower", "left" and "right" are only used to represent the relative position relationship. When the absolute position of the object to be described changes, the relative position relationship may also change accordingly.

[0053] Based on the fact that the enterprise relationship graph formed by financing and investment using basic industrial and commercial registration in the prior art is relatively simple, and it is impossible to obtain the business behaviors of the senior management personnel of the enterprise, so it is impossible to deeply understand the overall situation of the company; while for the enterprise relationship graph obtained by using the corporate legal person in the prior art, due to the inability to obtain the real legal person identity information, the graph data has large errors and cannot be used to comprehensively understand the situation of the enterprise. The present invention aims to propose a method and system for constructing an enterprise relationship graph, which quantifies the market transaction behaviors of an enterprise and its associated market entities through multiple dimensions, infers and mines deeper relationships between enterprises, and provides a comprehensive and detailed enterprise relationship graph for the enterprise to understand the overall situation of the enterprise.

[0054] The following further specifically introduces the method and system for constructing an enterprise relationship graph disclosed in the present invention with reference to the embodiments shown in the accompanying drawings.

[0055] Combined with Figure 1As shown in the figure, the process of constructing an enterprise relationship map disclosed by the present invention includes four aspects, namely, four steps of obtaining enterprise basic dimension data, enterprise operation data, enterprise association relationship judgment, and finally generating an enterprise relationship map, and then providing an application method, that is, applying the generated enterprise relationship map to enterprise relationship retrieval to obtain the association relationship between any two or more enterprises.

[0056] Specifically, the method for constructing an enterprise relationship map disclosed by the present invention, as Figure 2 shown, includes the following steps:

[0057] Step S102, determining the basic dimension data of the enterprise to be processed, where the basic dimensions include the registered address of the enterprise, the legal representative of the enterprise, the directors of the enterprise, the supervisors of the enterprise, the names of shareholders and other registered senior executives;

[0058] By obtaining the basic dimension data related to the enterprise and combining the information registered with the industrial and commercial department, it is possible to deeply explore the market entities indirectly associated with the enterprise. In the process of exploring these market entities, on the one hand, it can expand the scope covered by the enterprise relationship map, and on the other hand, it can understand the enterprise's development layout and talent introduction situation, and learn about the enterprise's development plan.

[0059] Step S104, determining the operation data of the enterprise to be processed, where the operation data are the upstream and downstream market entities and transaction behaviors that have had industrial and commercial behaviors with the enterprise during its operation process, including the names of investment companies received by the enterprise and the corresponding number of transactions, transaction amounts, and transaction times, and the names of other market entities that have a contractual relationship with the enterprise and the corresponding number of transactions, transaction amounts, and transaction times;

[0060] The purpose of this step is to collect the transaction data generated by the transaction behaviors in the enterprise operation process. When displayed in the enterprise relationship map, it can comprehensively display the business objects, business scope, and industries involved in the business of the enterprise, and can infer the future development direction and development goals of the enterprise. When implemented, the determination of the enterprise operation data is mainly to sort out the transaction behaviors between the enterprise and each market entity in the statistical year of the enterprise relationship map from the enterprise declaration information, and summarize the transaction time, number of transactions, and transaction amount between enterprises for subsequent calculation of the degree of association.

[0061] As an optional embodiment, after the basic dimension data and operation data of the enterprise to be processed are determined, data preprocessing needs to be performed first. In this solution, the process of data preprocessing is mainly to perform data governance on the determined data. For example, cleaning and normalizing the basic dimension data and the operation data, including standardizing the registered address of the determined enterprise into three levels of province-city-district, unifying the transaction amount in units of ten thousand, and unifying the format of the transaction time as YYYY.

[0062] Step S106: Match the basic dimension data and the business data, and obtain the enterprise association relationship according to the association degree between the enterprise to be processed and the associated market entities.

[0063] Among them, the process of matching the basic dimension data and the business data aims to list all market entities associated with the enterprise to be processed, and such market entities are all recorded as associated market entities. This step realizes the above-mentioned enterprise association relationship judgment process. Based on the basic dimension data information of the enterprise itself and integrating the market behaviors of the enterprise's business data, first, through matching in different dimensions, the associated market entities and the enterprise to be processed are divided into different association types, and then through the association types, the association degree between the associated market entities and the enterprise to be processed is further obtained. Different association degrees correspond to different association relationships.

[0064] For example, the enterprise association types include enterprise shareholding relationship, enterprise market behavior relationship, and enterprise close relationship; among them, the enterprise shareholding relationship is the enterprise association relationship obtained by matching based on the shareholding information in the industrial and commercial registration information according to the basic dimension data; the enterprise market behavior relationship is the enterprise association relationship obtained by matching based on the market entities in the business data and the publicly available enterprise business data; the enterprise close relationship is the relationship between each market entity associated by the senior executives of the enterprise in the enterprise operation behavior and transaction behavior obtained from the basic dimension data of the enterprise.

[0065] Step S108: Store the enterprise association relationship in the graph database to obtain the enterprise relationship graph.

[0066] Through the enterprise relationship graph, the overall profile of the enterprise can be displayed from multiple perspectives of the enterprise's basic dimensions, enterprise association types, and enterprise association relationships, and the enterprise operation track can be understood. The data is not only accurate but also comprehensive.

[0067] As an optional embodiment, the specific process of obtaining the enterprise association relationship is as follows: construct an enterprise relationship network for the enterprise to be processed, and determine the components of the direct association weight value and the indirect association weight value; calculate the direct association weight value, indirect association weight value, and final weight value between each associated market entity matched according to the basic dimension data and the business data and the enterprise respectively; judge the range of each weight threshold corresponding to each preset association relationship to which the final weight value belongs, and determine the enterprise association relationship.

[0068] During implementation, the direct association relationship between two enterprises is characterized by quantifying the transaction behavior between the two enterprises, that is, only when there is a transaction behavior between two enterprises can there be a direct association, otherwise, the two enterprises are recorded as indirectly associated.

[0069] Based on this, an association algorithm model for measuring the association degree between two enterprises in an embodiment is as follows:

[0070] Define the component of the direct association weight value as X1 and the component of the indirect association weight value as X2, where X1 + X2 = 1. Then the calculation formulas for the direct association weight value, indirect association weight value, and final weight value between an enterprise and any of its associated market entities are as follows:

[0071] Direct association weight value = Sum((transaction times + transaction amount) * (1 - (current year - transaction year) / N));

[0072] Indirect association weight value = direct association weight value of the associated market entity * (N - Z) / N + nN;

[0073] Final weight value = direct association weight value * X1 + indirect association weight value * X2;

[0074] Among them, N is the number of years counted in the enterprise relationship graph, Z is the association level between the enterprise and its associated market entity determined by the shortest path method, and n is the number of senior executives who concurrently hold positions in the enterprise and its associated market entity; the sum formula for the direct association weight value is the transaction weight of the enterprise and the associated market entity for each year, and thus the direct association weight value is the total weight within the statistical years.

[0075] For example, when the component of the direct association weight value X1 takes a value of 60 and the component of the indirect association weight value X2 takes a value of 40, and the statistical year of the enterprise relationship graph is 10 years, then for the enterprise and any of its associated market entities, the direct association weight value = Sum(transaction times + transaction amount) * (1 - (current year - transaction year) / 10), the indirect association weight value = direct association weight value * (10 - association level) / 10 + 10n, and the final weight value = direct association weight value * 0.6 + indirect association weight value * 0.4.

[0076] Another example, it can be set that the weight threshold range for a highly close association relationship with the enterprise is not less than 80, the weight threshold range for a close association relationship with the enterprise is [60, 80), the weight threshold range for an association relationship with the enterprise is [40, 60), and the weight threshold range for a weak association relationship with the enterprise is [0, 40). When the calculated final weight value is greater than 80, it is determined that there is a highly close association relationship between the enterprise and the associated entity. If it belongs to the range of [60, 80), it is determined that there is a close relationship between the enterprise and the associated entity. If it belongs to the range of [40, 60), it is determined that there is an association relationship between the enterprise and the associated entity.

[0077] Table 1 provides an example where the data in the enterprise relationship graph is the associated entities of enterprise A that had transaction behaviors between 2012 and 2022. Enterprise B is an associated enterprise of enterprise A. Enterprise B had transaction behavior data with enterprise A in 2018 and from 2020 to 2022. The associated level of enterprise A and enterprise B determined by the shortest path method is 2, and the number of senior executives who served in both enterprise A and enterprise B is 3. According to the following data and the weight division of the above-mentioned associated relationship, the process of calculating the associated relationship between the two enterprises is as follows:

[0078] Table 1 Transaction behavior data of enterprise A and enterprise B from 2018 to 2022

[0079] Year 2022 2021 2020 2018 Number of Transactions 7 9 4 1 Transaction Amount (in ten thousand yuan) 16 18 7 5

[0080] Then:

[0081] Direct association weight value = Sum((transaction times + transaction amount) * (1 - (current year - transaction year) / N)) = ((7 + 16) * (1 - (2022 - 2022) / 10) + (9 + 18) * (1 - (2022 - 2021) / 10) + (4 + 7) * (1 - (2022 - 2020) / 10) + (1 + 5) * (1 - (2022 - 2018) / 10)) = 59.7;

[0082] Indirect association weight value = direct association weight value of the associated market entity * (N - Z) / N + nN = 59.7 * (10 - 2) / 10 + 30 = 77.76;

[0083] Final weight value = direct association weight value * X1 + indirect association weight value * X2 = 69.7 * 0.6 + 77.76 * 0.4 = 66.924

[0084] Since 60 < final weight value < 80, it is determined that there is a close associated relationship between enterprise A and enterprise B, which can be marked as a close relationship in the enterprise relationship graph.

[0085] For a search platform, after obtaining an enterprise relationship graph, a search service can be provided, that is, the above-mentioned enterprise relationship graph application process can be implemented; that is, the enterprise relationship graph construction method further includes: obtaining the enterprise name and enterprise association relationship ID according to the enterprise relationship graph; storing the enterprise name and enterprise association relationship ID in the ES index, and providing a quick enterprise name query window; using the quick enterprise name query window to receive a search command, querying the enterprise association relationship ID corresponding to the search command, and obtaining the corresponding enterprise relationship graph for the query. During implementation, the enterprise association relationship ID corresponding to the enterprise name is unique, and the enterprise relationship graph corresponding to the enterprise association relationship ID is also uniquely corresponding; the user can obtain the corresponding enterprise relationship graph according to the retrieved enterprise association relationship ID. Optionally, it further includes: displaying the enterprise relationship graph on the display screen of the platform for the user to view.

[0086] The present invention first obtains all market entities related to an enterprise from different dimensions, then quantitatively measures the association degree between the market entity and the enterprise based on the transaction behavior occurring between the market entity and the enterprise, and further obtains the association relationship existing between any market entity associated with the enterprise and the enterprise; by mining deeper association relationships between enterprises from multiple dimensions, it makes up for the deficiency of the prior art that can only provide an enterprise relationship graph with a single relationship, enabling query personnel to comprehensively and more deeply understand the overall situation of an enterprise.

[0087] In an embodiment of the present invention, an electronic device is further provided. The electronic device includes a processor and a memory, and a computer program is stored in the memory. When the computer program is loaded and executed by the processor, the above-mentioned enterprise relationship graph construction method is implemented.

[0088] The above program can run in the processor, or it can also be stored in the memory (or referred to as a computer-readable medium). The computer-readable medium includes permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media do not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0089] These computer programs can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process or multiple processes and / or boxes. Different steps can be implemented by different modules. Figure 1 One process or multiple processes and / or boxes Figure 1 The steps for implementing the functions specified in one box or multiple boxes. For different steps, they can be implemented by different modules.

[0090] This embodiment provides such an electronic device or system. This device is called an enterprise relationship graph construction device, and this system is called a text intelligent display system. The system includes: a first determination module for determining the basic dimension data of the enterprise to be processed, where the basic dimensions include the registered address of the enterprise, the legal representative of the enterprise, the directors of the enterprise, the supervisors of the enterprise, the names of shareholders and other registered senior executives; a second determination module for determining the business data of the enterprise to be processed, where the business data is the upstream and downstream market entities and transaction behaviors that have had industrial and commercial behaviors with the enterprise during its operation, including the names of investment companies received by the enterprise and the corresponding number of transactions, transaction amounts, and transaction times, and the names of other market entities that have a contract relationship with the enterprise and the corresponding number of transactions, transaction amounts, and transaction times; a matching module for matching the basic dimension data and the business data, and obtaining enterprise association relationships according to the association degree between the associated market entities and the enterprise to be processed; a first storage module for storing the enterprise association relationships into a graph database to obtain an enterprise relationship graph.

[0091] This system is used to implement the functions of the enterprise relationship graph construction method in the above embodiment. Each module in this system corresponds to each step in the method. Those that have been described in the method will not be elaborated here.

[0092] For example, the process of the matching module obtaining enterprise association relationships is executed by the following execution units, including: a construction unit for constructing an enterprise relationship network for the enterprise to be processed and determining the components of the direct association weight value and the indirect association weight value; a calculation unit for respectively calculating the direct association weight value, the indirect association weight value, and the final weight value between each associated market entity matched according to the basic dimension data and the business data and the enterprise; a judgment and determination unit for judging the range of each weight threshold corresponding to each preset association relationship to which the final weight value belongs and determining the enterprise association relationship.

[0093] For another example, the formulas for the computing unit to calculate the direct association weight value, indirect association weight value, and final weight value between the enterprise to be processed and any of its associated market entities are as follows: Direct association weight value = (transaction times + transaction amount) * (1 - (current year - transaction year) / N); Indirect association weight value = direct association weight value of the associated market entity * (N - Z) / N + nN; Final weight value = direct association weight value * X1 + indirect association weight value * X2; where N is the number of years statistically counted in the enterprise relationship graph, Z is the association level between the enterprise and its associated market entities determined by the shortest path method, and n is the number of senior executives who concurrently hold positions in the enterprise and its associated market entities.

[0094] The enterprise relationship graph constructed by the present invention is not only based on the direct association of simple industrial and commercial relationships, but also on the basis of industrial and commercial associations. Through the quantification of market transaction behaviors between enterprises, deeper relationships between enterprises are inferred and mined. When applied to understanding the overall situation of an enterprise and evaluating its business status, the enterprise relationship graph can provide richer and multi-dimensional analysis results, and also provide more accurate and intuitive data support for measuring the development of the enterprise.

[0095] Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Those with ordinary knowledge in the technical field to which the present invention pertains can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to what is defined by the claims.

Claims

1. A method for constructing an enterprise relationship graph, characterized in that, Including: Determine the basic dimension data of the enterprise to be processed. The basic dimensions include the registered address of the enterprise, the legal representative of the enterprise, the directors of the enterprise, the supervisors of the enterprise, the names of shareholders and other registered senior executives; Determine the business data of the enterprise to be processed. The business data are the upstream and downstream market entities and transaction behaviors that have had industrial and commercial behaviors with the enterprise during its operation, including the names of investment companies received by the enterprise and the corresponding number of transactions, transaction amounts, and transaction times, and the names of other market entities having contractual relationships with the enterprise and the corresponding number of transactions, transaction amounts, and transaction times; Match the basic dimension data and the business data, and obtain the enterprise association relationship according to the association degree between the enterprise to be processed and the associated market entities; Store the enterprise association relationship in the graph database to obtain the enterprise relationship graph; Specifically, the specific process of obtaining the enterprise association relationship is as follows: construct an enterprise relationship network for the enterprise to be processed, and determine the components of the direct association weight value and the indirect association weight value; calculate the direct association weight value, indirect association weight value, and final weight value between each associated market entity matched according to the basic dimension data and the business data and the enterprise respectively; judge the range of each weight threshold corresponding to each preset association relationship to which the final weight value belongs, and determine the enterprise association relationship; Define the component of the direct association weight value as X1, and the component of the indirect association weight value as X2. If X1 + X2 = 1, then the calculation formulas for the direct association weight value, indirect association weight value, and final weight value between the enterprise and any of its associated market entities are respectively: Direct association weight value = Sum((number of transactions + transaction amount) * (1 - (current year - transaction year) / N)); Indirect association weight value = direct association weight value of the associated market entity * (N - Z) / N + nN; Final weight value = direct association weight value * X1 + indirect association weight value * X2; where, N is the number of years statistically in the enterprise relationship graph, Z is the association level between the enterprise and its associated market entities determined by the shortest path method, and n is the number of senior executives who concurrently hold positions in the enterprise and its associated market entities.

2. The method for constructing an enterprise relationship graph according to claim 1, wherein The process of matching the basic dimension data and the business data includes: Obtain all associated market entities associated with the enterprise to be processed, and determine the enterprise association type between the associated market entity and the enterprise to be processed; The enterprise association types include enterprise shareholding relationship, enterprise market behavior relationship, and enterprise close relationship; among them, the enterprise shareholding relationship is the enterprise association relationship obtained by matching based on the shareholding information in the industrial and commercial registration information according to the basic dimension data; the enterprise market behavior relationship is the enterprise association relationship obtained by matching based on the publicly available enterprise business data of the market entities in the business data; the enterprise close relationship is the relationship between each market entity associated by the senior executives of the enterprise in the basic dimension data obtained during the enterprise operation behavior and transaction behavior.

3. The method for constructing an enterprise relationship graph according to claim 1, wherein Before matching the basic dimension data and the business data, data preprocessing is also included; The data preprocessing is to clean and normalize the basic dimension data and the business data, including standardizing the registered address of the enterprise into three levels of province-city-district, unifying the transaction amount in units of ten thousand, and unifying the format of the transaction time as YYYY.

4. The enterprise relationship graph construction method according to claim 1, wherein The method further includes: Obtaining the enterprise name and the enterprise association relationship ID according to the enterprise relationship graph; Storing the enterprise name and the enterprise association relationship ID into the ES index, and providing a quick enterprise name query window; Receiving a search command by using the quick enterprise name query window, querying the enterprise association relationship ID corresponding to the search command, and obtaining the enterprise relationship graph corresponding to the query.

5. An enterprise relationship graph construction system, characterized in that, Including: A first determination module, configured to determine the basic dimension data of the enterprise to be processed, where the basic dimensions include the registered address of the enterprise, the legal representative of the enterprise, the directors of the enterprise, the supervisors of the enterprise, the names of the shareholders and other registered senior executives; A second determination module, configured to determine the business data of the enterprise to be processed, where the business data is the upstream and downstream market entities and transaction behaviors that have had industrial and commercial behaviors during the operation of the enterprise, including the name of the investment company received by the enterprise and the corresponding number of transactions, transaction amount, and transaction time, and the names of other market entities that have a contractual relationship with the enterprise and the corresponding number of transactions, transaction amount, and transaction time; A matching module, configured to match the basic dimension data and the business data, and obtain the enterprise association relationship according to the association degree between the associated market entity and the enterprise to be processed; A first storage module, configured to store the enterprise association relationship into the graph database to obtain the enterprise relationship graph; Among them, the process of the matching module obtaining the enterprise association relationship is executed by the following execution units, including: A construction unit, configured to construct an enterprise relationship network for the enterprise to be processed, and determine the components of the direct association weight value and the indirect association weight value; A calculation unit, configured to calculate the direct association weight value, the indirect association weight value, and the final weight value of each associated market entity matched according to the basic dimension data and the business data with the enterprise respectively; A judgment and determination unit, configured to judge the range of each weight threshold corresponding to each preset association relationship to which the final weight value belongs, and determine the enterprise association relationship; The formulas for the calculation unit to calculate the direct association weight value, the indirect association weight value, and the final weight value of the enterprise to be processed and any of its associated market entities are respectively: Define the component of the direct association weight value as X1, the component of the indirect association weight value as X2, X1 + X2 = 1, then: Direct association weight value = Sum((number of transactions + transaction amount) * (1 - (current year - transaction year) / N)); Indirect association weight value = direct association weight value of the associated market entity * (N - Z) / N + nN; Final weight value = direct association weight value * X1 + indirect association weight value * X2; Among them, N is the number of years counted in the enterprise relationship graph, Z is the association level between the enterprise and its associated market entity determined by the shortest path method, and n is the number of senior executives who concurrently hold positions in the enterprise and its associated market entity.

6. An enterprise relationship graph construction device, characterized in that It includes a processor and a memory, and a computer program is stored in the memory. The computer program is loaded and executed by the processor to implement the enterprise relationship graph construction method according to any one of claims 1 to 4.

Images