CN116503163A

CN116503163A - Service risk identification method, device, computer equipment and storage medium

Info

Publication number: CN116503163A
Application number: CN202310736985.3A
Authority: CN
Inventors: 邱俊凌; 赵芳
Original assignee: CCB Finetech Co Ltd
Current assignee: CCB Finetech Co Ltd
Priority date: 2023-06-21
Filing date: 2023-06-21
Publication date: 2023-07-28

Abstract

The disclosure relates to a business risk identification method, a business risk identification device, computer equipment and a storage medium. Relates to the technical field of big data processing, which comprises the following steps: acquiring service application information of a target object; determining association information associated with the target object by using the service application information, and establishing a target graph data structure of the target object when the service application is performed based on the service application information and the association information, wherein the target graph data structure comprises: a target node corresponding to the target object and a credit source node corresponding to the service type information in the service application information; calculating a risk conduction coefficient between the credit giving source node and the target node according to the associated attribute information between adjacent nodes in the target graph data structure and preset weight corresponding to the attribute information; and determining that the risk exists when the target object is subjected to the acceptance service in response to the risk conduction coefficient being larger than a preset risk threshold. By adopting the method, various data of enterprises can be synthesized, and the business risk can be accurately estimated.

Description

Service risk identification method, device, computer equipment and storage medium

Technical Field

The disclosure relates to the technical field of big data processing, in particular to a business risk identification method, a business risk identification device, computer equipment and a storage medium.

Background

With the development of social economy, more and more enterprises are required to borrow banks or financial institutions due to their own development requirements. The bank or the financial institution evaluates the repayment capability of the enterprise, and when the repayment capability of the enterprise reaches the standard, the bank or the financial institution can trust the enterprise.

However, the credit limit and the auditing of the client are gradually increased due to the influence of various factors, so that the client is easy to excessively trust, or the credit difficulty is additionally increased. In addition, the trust risk is evaluated only aiming at the index of a single enterprise, so that the trust risk evaluation is inaccurate.

Disclosure of Invention

Based on the above, it is necessary to provide a business risk identification method, device, computer equipment and storage medium capable of integrating various data of enterprises and accurately evaluating the credit risk.

In a first aspect, the present disclosure provides a business risk identification method. The method comprises the following steps:

Acquiring service application information of a target object;

determining association information associated with the target object by utilizing the service application information, and establishing a target graph data structure of the target object when the service application is carried out based on the service application information and the association information, wherein the target graph data structure at least comprises: a target node corresponding to the target object and a credit source node corresponding to the service type information in the service application information;

calculating a risk conduction coefficient between a credit giving source node and the target node according to the associated attribute information between adjacent nodes in the target graph data structure and preset weight corresponding to the attribute information;

and determining that the risk exists when the target object is subjected to the acceptance service when the risk conduction coefficient is larger than a preset risk threshold, wherein the service is determined according to service application information.

In one embodiment, the calculating the risk conduction coefficient between the credit giving source node and the target node according to the associated attribute information between the adjacent nodes in the target graph data structure and the preset weight corresponding to the attribute information includes:

Determining associated attribute information between each neighboring node in the target graph data structure;

according to the attribute information and preset weight corresponding to the attribute information, calculating to obtain a risk conduction score between each two adjacent nodes;

and calculating to obtain the risk conduction coefficient between the credit giving source node and the target node according to the node between the credit giving source node and the target node and the risk conduction value between each adjacent node.

In one embodiment, the calculating, according to the node between the trusted source node and the target node and the risk conduction value between each adjacent node, the risk conduction coefficient between the trusted source node and the target node includes:

acquiring a path node between the information service source node and the target node;

determining risk conduction scores between the path nodes according to the risk conduction scores between each two adjacent nodes;

according to the risk conduction scores between the credit source node and the adjacent credit source node, the risk conduction scores between the path nodes and the risk conduction scores between the adjacent target nodes and the target nodes, calculating to obtain risk conduction coefficients between the credit source node and the target nodes;

The adjacent credit source node is a node adjacent to the credit source node in the path node; the adjacent target node is a node adjacent to the target node in the path node.

In one embodiment, the calculating, according to the risk conduction score between the credit source node and the adjacent credit source node, the risk conduction score between the path nodes, and the risk conduction score between the adjacent target node and the target node, obtains a risk conduction coefficient between the credit source node and the target node, includes:

determining path nodes in each conductive path in response to a plurality of conductive paths existing between the trusted source node and the target node;

for each conduction path, calculating to obtain a risk conduction coefficient between the credit giving source node and the target node according to the risk conduction score between the credit giving source node and the adjacent credit giving source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node;

and determining the risk conduction coefficient with the largest value in the risk conduction coefficients between the credit source node and the target node in each conduction path, and determining the target risk conduction coefficient between the credit source node and the target node according to the risk conduction coefficient with the largest value.

and multiplying the risk conduction score between the credit source node and the adjacent credit source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node, and calculating to obtain the risk conduction coefficient between the credit source node and the target node.

In one embodiment, the determining, using the service application information, association information associated with the target object includes:

determining association information associated with the object information by utilizing the object information of the target object in the service application information;

performing data processing on the object information and the associated information, wherein the data processing comprises: and screening according to the information value, performing null value processing and data binning processing.

In one embodiment, the establishing the target graph data structure when the target object applies for the service based on the service application information and the association information includes:

combining the object information after data processing with the associated information according to an association relation to form a first graph data structure of a target object, wherein in the first graph data structure, the object information and the associated information are nodes, and the association relation between the object information and the associated information is an edge;

and adding the service type information into the first graph data structure according to the association relation between the service type information, the object information and the association information to form a target graph data structure when the target object applies for service.

In one embodiment, the nodes in the first graph data structure include: and the node corresponding to the target object and/or the node of the enterprise corresponding to the target object.

In a second aspect, the present disclosure further provides a business risk identification device. The device comprises:

the information acquisition module is used for acquiring service application information of the target object;

The map data structure creation module is configured to determine association information associated with the target object by using the service application information, and create a target map data structure of the target object when performing service application based on the service application information and the association information, where the target map data structure at least includes: a target node corresponding to the target object and a credit source node corresponding to the service type information in the service application information;

the risk conduction coefficient calculation module is used for calculating a risk conduction coefficient between a credit giving source node and the target node according to the associated attribute information between adjacent nodes in the target graph data structure and preset weight corresponding to the attribute information;

and the risk determination module is used for determining that the target object is trusted to have risk in response to the risk conduction coefficient being larger than a preset risk threshold.

In one embodiment, the risk conduction coefficient calculation module includes:

an attribute information determining module, configured to determine associated attribute information between each neighboring node in the target graph data structure;

the risk conduction value calculation module is used for calculating and obtaining a risk conduction value between each two adjacent nodes according to the attribute information and preset weight corresponding to the attribute information;

And the conduction coefficient calculation module is used for calculating and obtaining the risk conduction coefficient between the credit giving source node and the target node according to the nodes between the credit giving source node and the target node and the risk conduction value between each adjacent node.

In one embodiment, the conductivity coefficient calculating module includes:

the path node acquisition module is used for acquiring the path node between the credit source node and the target node;

the path node score calculation module is used for determining the risk conduction scores among the path nodes according to the risk conduction scores among each adjacent node;

the conduction coefficient calculation sub-module is used for calculating and obtaining a risk conduction coefficient between the credit giving source node and the target node according to the risk conduction score between the credit giving source node and the adjacent credit giving source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node;

In one embodiment, the conductivity coefficient calculation sub-module is further configured to determine a path node in each conductive path in response to the presence of multiple conductive paths for the trusted source node and the target node; for each conduction path, calculating to obtain a risk conduction coefficient between the credit giving source node and the target node according to the risk conduction score between the credit giving source node and the adjacent credit giving source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node; and determining the risk conduction coefficient with the largest value in the risk conduction coefficients between the credit source node and the target node in each conduction path, and determining the target risk conduction coefficient between the credit source node and the target node according to the risk conduction coefficient with the largest value.

In one embodiment, the conduction coefficient calculation submodule is further configured to multiply the risk conduction score between the credit source node and the adjacent credit source node, the risk conduction score between the path nodes, and the risk conduction score between the adjacent target node and the target node, and calculate to obtain the risk conduction coefficient between the credit source node and the target node.

In one embodiment, the graph data structure creation module includes:

the associated information determining module is used for determining associated information associated with the object information by utilizing the object information of the target object in the service application information;

the data processing module is used for performing data processing on the object information and the associated information, and the data processing comprises the following steps: and screening according to the information value, performing null value processing and data binning processing.

In one embodiment, the graph data structure creation module further includes:

the first graph data structure creation module is used for combining the object information subjected to data processing and the associated information according to an association relation to form a first graph data structure of a target object, wherein in the first graph data structure, the object information and the associated information are nodes, and the association relation between the object information and the associated information is an edge;

and the information adding module is used for adding the service type information into the first graph data structure according to the association relation between the service type information, the object information and the association information to form a target graph data structure when the target object applies for the service.

In a third aspect, the present disclosure also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the steps of any of the method embodiments described above when the processor executes the computer program.

In a fourth aspect, the present disclosure also provides a computer-readable storage medium. The computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of any of the method embodiments described above.

In a fifth aspect, the present disclosure also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the steps of any of the method embodiments described above.

In the above embodiments, when the target object needs to accept the service, the service application information of the target object may be obtained. And determining the association information associated with the target object by utilizing the service application information, and establishing a target graph data structure of the target object when the service application is carried out based on the service application information and the association information. Various associated information corresponding to the target object can be dynamically determined, various associated information corresponding to the target object can be deeply mined, various data are synthesized, and a target graph data structure with the associated information is generated. And calculating the risk conduction coefficient between the information service source node and the target node according to the associated attribute information between adjacent nodes and the preset weight corresponding to the attribute information, and accurately determining the strength of the risk conduction force between the target node and the information service source node under various associations. And determining that the risk exists when the target object is subjected to the acceptance service in response to the risk conduction coefficient being larger than a preset risk threshold, and accurately evaluating the credit giving risk through the risk conduction score. In addition, unlike the traditional technology, the scheme does not need to depend on expert experience excessively, and a neural network model is not needed when risk conduction data are calculated. The method can effectively reduce the resources consumed by using a large amount of data exercise for improving the model under the neural network model, improve the efficiency and reduce the calculation resources consumed in evaluating the trust risk.

Drawings

In order to more clearly illustrate the embodiments of the present disclosure or the prior art, the drawings that are required in the detailed description or the prior art will be briefly described, it will be apparent that the drawings in the following description are some embodiments of the present disclosure, and other drawings may be obtained according to the drawings without inventive effort for a person of ordinary skill in the art.

FIG. 1 is a schematic diagram of an application environment of a business risk identification method in one embodiment;

FIG. 2 is a flow chart of a business risk identification method in one embodiment;

FIG. 3 is a schematic diagram of a target graph data structure in one embodiment;

FIG. 4 is a flow chart of step S206 in one embodiment;

FIG. 5 is a flow chart of step S306 in one embodiment;

FIG. 6 is a flow chart of step S406 in one embodiment;

FIG. 7 is a schematic diagram of a target graph data structure in which multiple conductive paths exist, in one embodiment;

FIG. 8 is a flow chart of a portion of step S204 in one embodiment;

FIG. 9 is a flow chart of another part of the step S204 in one embodiment;

FIG. 10 is a schematic diagram of a graph data structure formed by business type information, object information, and association information in one embodiment;

FIG. 11 is a flowchart of a business risk identification method according to another embodiment;

FIG. 12 is a block diagram schematically illustrating a construction of a business risk recognition apparatus in one embodiment;

FIG. 13 is a schematic diagram of the internal structure of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present disclosure more apparent, the present disclosure will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present disclosure.

It should be noted that the terms "first," "second," and the like in the description and claims herein and in the foregoing figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, apparatus, article, or device that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or device.

In this document, the term "and/or" is merely one association relationship describing the associated object, meaning that three relationships may exist. For example, a and/or B may represent: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.

As described in the background, the conventional manner of trust risk assessment at present generally includes the following methods: 1. and evaluating the credit risk by utilizing an early warning rule formulated by an expert. However, the early warning rules formulated by the expert are formulated mainly by expert analysis of scenes and model establishment, and have high requirements on expert business knowledge and experience, and when business attributes are changed, the model update period is long and excessively depends on expert experience, and certain subjective judgment factors exist, so that the flexibility of combing complex association relations of the business scenes is lacking.

2. And evaluating the credit risk by utilizing a data mining technology. The data mining technology comprises a neural network, artificial intelligence, a Bayesian network and the like, the mode relates to calculation extension of probability variables and deep data mining, a large amount of data is needed for training and learning, the complexity of a model is high, under the condition that a business scene is more and more complex, the calculation and training geometric multiple of a data model are improved, the requirement on server resources is high, the load of a model running batch on the server resources is excessive, and the accuracy of a result evaluation result cannot be guaranteed.

Therefore, to solve the above-mentioned problem, the embodiment of the present disclosure provides a business risk identification method, which can be applied to an application environment as shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. When a customer needs to make a loan, business application information corresponding to the loan may be input into the terminal 102. The terminal 102 may determine association information associated with the target object using the service application information. The terminal 102 may establish a target graph data structure when the target object applies for a service based on the service application information and the association information, where the target graph data structure at least includes: and the target node corresponding to the target object and the credit source node corresponding to the service type information in the service application information. The terminal 102 may calculate the risk conduction coefficient between the credit giving source node and the target node according to the associated attribute information between the adjacent nodes in the target graph data structure and the preset weight corresponding to the attribute information. In response to the risk conduction coefficient being greater than a preset risk threshold, the terminal 102 determines that there is a risk in trust of the target object and performs early warning. And transmits the early warning information to the service 104, and the server 104 can perform steps such as rechecking or rectifying. The terminal 102 may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, etc. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.

In one embodiment, as shown in fig. 2, a business risk identification method is provided, and the method is applied to the terminal 102 in fig. 1 for illustration, and includes the following steps:

s202, acquiring service application information of a target object.

The service application information may be some information input when the target object needs to transact the service. Different services have different corresponding service application information. For example, if the business is a loan business, the business application information may include: the type of loan, the amount of the loan, the purpose of the loan, the time of the loan, mortgage of the loan, etc., and the personal identity information of the target object. The target object may generally be a user conducting a business in some embodiments of the present disclosure. No absolute limitation is placed on the service application information in some embodiments of the present disclosure.

Specifically, when the user needs to transact the service, service application information corresponding to the transaction of the service may be input into the terminal. The terminal acquires service application information of the user (target object).

S204, determining the association information associated with the target object by using the service application information, and establishing a target graph data structure of the target object when the service application is performed based on the service application information and the association information, wherein the target graph data structure at least comprises: and the target node corresponding to the target object and the credit source node corresponding to the service type information in the service application information.

The association information is generally information having a certain relationship with the service application information. For example, the service application information includes work unit information, and the associated information may be information associated with each person in the work unit. For another example, the service application information includes: the information of the working position can be various information corresponding to the working position, the information of the leading corresponding to the working position and the information of the corresponding subordinate staff. The associated information may be a person, an enterprise, some product, or the like, as is typically the case. The related information may also be information related to the business application information, and if the business application information is loan information, the related information may be loan basic information, mortgage information, loan risk information, loan related person information, loan overdue details, and the like. According to the different information types in the service application information, the corresponding associated information is also different, and the specific type of the associated information is not absolutely limited in some embodiments of the present disclosure. The target graph data structure may be a NEO4J database structure, and the graph data structure may be a collection of nodes and edges from a data perspective, and the edges are constructed from node to node. The business type information may generally be the type of business that the target object is required to apply for, e.g., different business operations such as loans, mortgages, repayment, etc.

In particular, various identity information of the target object in the service application information may be utilized to match various associated information associated with the target object in the database. After the associated information is obtained, the node can be established with each information in the service application information and the associated information, and the node is also established according to the service type information. And then combining according to the established association relation between each node to establish edges. And combining the nodes and the edges to obtain the target graph data structure. The corresponding node of the target object itself in the target graph data structure may be the target node. The node corresponding to the service type information may be a credit source node. The target graph data structure may be as shown in fig. 3, where a may be a target object, A1, A2, A3 may be nodes corresponding to association information associated with the target object, and B may be a source node. An association relationship can exist between A and A1, an association relationship can also exist between A and A2, an association relationship can exist between A2 and A3, and an association relationship can exist between A3 and B, so that a connection can be established between a target node corresponding to a target object and a credit source node.

S206, calculating a risk conduction coefficient between the credit giving source node and the target node according to the associated attribute information between the adjacent nodes in the target graph data structure and the preset weight corresponding to the attribute information.

Wherein the attribute information may be information associated in the neighboring nodes in general, for example, association information (upper and lower levels) between job positions may be attribute information. If the node a and the node A1 are adjacent nodes, the association information between the node a and the node A1 is that the target object corresponding to the node a borrows the target object of the node A1, and the lending information between the adjacent nodes may also be attribute information. It will be appreciated that if the information associated with adjacent nodes is different, the corresponding attribute information is also different, and in some embodiments of the present disclosure, no absolute limitation is imposed on the attribute information. Each attribute information has a corresponding weight, and the weight can be set in a preset mode or can be obtained by processing adjacent nodes through a data mining neural network model. The risk conduction coefficient can be used for measuring the risk level possibly existing when the target object receives the service corresponding to the service application information. The higher the risk conduction coefficient, the more likely there is a risk, such as a risk of default, in accepting the business for the target object. The score for the risk conductivity coefficient is shown as 0 to 1 points.

Specifically, according to the attribute information between each adjacent node and the preset weight corresponding to the attribute information in the target graph data structure, the risk score between each adjacent node is calculated, and then the risk conduction coefficient between the information service source node and the target node is obtained according to the risk score between each adjacent node.

In some exemplary embodiments, the illustration continues with the target graph data structure in FIG. 3. The risk score between A and A1 can be calculated according to the attribute information between A and A1 and the weight corresponding to the attribute information, and the risk score between A and A2, the risk score between A2 and A3 and the risk score between A3 and B can be calculated by analogy. And obtaining the risk conduction coefficient between the target node and the credit source node according to the calculated risk score.

And S208, determining that risks exist and early warning is carried out when the target object is subjected to service in response to the risk conduction coefficient being larger than a preset risk threshold, wherein the service is determined according to service application information.

Specifically, a service corresponding to the service application information may be determined. When the risk conduction coefficient is larger than a preset risk threshold, the risk that the target object possibly has risk when the target object accepts the service can be determined, and early warning can be performed on the target object accepting the service.

In the business risk identification method, when the target object needs to accept business, the business application information of the target object can be acquired. And determining the association information associated with the target object by utilizing the service application information, and establishing a target graph data structure of the target object when the service application is carried out based on the service application information and the association information. Various associated information corresponding to the target object can be dynamically determined, various associated information corresponding to the target object can be deeply mined, various data are synthesized, and a target graph data structure with the associated information is generated. And calculating the risk conduction coefficient between the information service source node and the target node according to the associated attribute information between adjacent nodes and the preset weight corresponding to the attribute information, and accurately determining the strength of the risk conduction force between the target node and the information service source node under various associations. And when the risk conduction coefficient is larger than a preset risk threshold, determining that the risk exists when the target object is subjected to the acceptance service, and carrying out early warning, wherein the trust risk can be accurately estimated through the risk conduction value. In addition, unlike the traditional technology, the scheme does not need to depend on expert experience excessively, and a neural network model is not needed when risk conduction data are calculated. The method can effectively reduce the resources consumed by using a large amount of data exercise for improving the model under the neural network model, improve the efficiency and reduce the calculation resources consumed in evaluating the trust risk.

In one embodiment, as shown in fig. 4, the calculating a risk conduction coefficient between an information-providing source node and the target node according to the associated attribute information between adjacent nodes in the target graph data structure and a preset weight corresponding to the attribute information includes:

s302, determining associated attribute information between each adjacent node in the target graph data structure.

And S304, calculating to obtain a risk conduction value between each two adjacent nodes according to the attribute information and the preset weight corresponding to the attribute information.

Specifically, adjacent nodes in the target graph data structure may be determined first. Attribute information of the association between the neighboring nodes is then determined from each neighboring node. And further, calculating to obtain the risk conduction value between the adjacent nodes according to the attribute information between each adjacent node and the weight corresponding to the attribute information.

In some exemplary embodiments, the attribute information existing before the neighboring node may be X, Y and Z, where the weight corresponding to X may be a, the weight corresponding to Y may be b, and the weight corresponding to Z may be C, and the risk conduction score f=ax+by+cz.

S306, calculating to obtain the risk conduction coefficient between the credit source node and the target node according to the node between the credit source node and the target node and the risk conduction value between each adjacent node.

Specifically, the risk conduction score of the adjacent node between the credit source node and the target node can be determined according to the node between the credit source node and the target node and the risk conduction score between the adjacent nodes, and then the risk conduction coefficient is calculated according to the risk conduction score of the adjacent node between the credit source node and the target node.

In some exemplary embodiments, continuing with the illustration of FIG. 3, after the risk conductance scores between A and A1, A and A2, A2 and A3, A3 and B are calculated, the risk conductance coefficients may be calculated using only the risk conductance scores between A and A2, A2 and A3, A3 and B, since the risk conductance scores between A and A1 are not in the path of A and B. For example, the risk conduction coefficients may be obtained by adding or multiplying the risk conduction scores between a and A2, A2 and A3, and A3 and B.

In this embodiment, according to attribute information and preset weights corresponding to the attribute information, a risk conduction score between each adjacent node may be calculated, and further, according to the risk conduction score between each adjacent node and a node between a credit source node and a target node, a risk conduction coefficient between the credit source node and the target node may be calculated, and a correlation between adjacent nodes may be used to obtain a risk conduction score. And then, the risk conduction scores of all adjacent nodes between the credit giving source node and the target node are integrated to obtain a risk conduction coefficient, and the accuracy of risk assessment is ensured.

In one embodiment, as shown in fig. 5, the calculating, according to the node between the trusted source node and the target node and the risk conduction value between each adjacent node, the risk conduction coefficient between the trusted source node and the target node includes:

s402, obtaining a path node between the credit source node and the target node.

S404, determining the risk conduction value between the path nodes according to the risk conduction value between each two adjacent nodes.

S406, calculating to obtain a risk conduction coefficient between the credit source node and the target node according to the risk conduction score between the credit source node and the adjacent credit source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node;

the adjacent credit source node is a node adjacent to the credit source node in the path node; the adjacent target node is a node adjacent to the target node in the path node. Continuing with the description by way of example of the target graph data structure in FIG. 3, the path nodes may be the A2 and A3 nodes.

Specifically, since there may be multiple nodes in the target graph data structure, the location of each node is different. To calculate the risk conductance between the trusted source node and the target node, a path node between the trusted source node and the target node may be determined. And then, as the risk conduction score of each adjacent node is calculated in the steps, the risk conduction score corresponding to each adjacent path node can be obtained, the risk conduction score corresponding to the node adjacent to the information source node in the information source node and the path node, the risk conduction score corresponding to the node adjacent to the target node in the target node and the path node, and the risk conduction coefficient between the information source node and the target node can be calculated.

In this embodiment, the risk conduction score between the path nodes may be calculated and determined by using the path nodes between the credit source node and the target node, so that the risk conduction coefficient between the credit source node and the target node may be obtained by calculating according to the risk conduction score between the credit source node and the adjacent credit source node, the risk conduction score between the path nodes, and the risk conduction score between the adjacent target node and the target node, so that all the association relations between the credit source node and the target node may be integrated, further, the accuracy of the risk conduction coefficient may be ensured, and the credit risk may be accurately estimated.

In one embodiment, as shown in fig. 6, the calculating, according to the risk conduction score between the trusted source node and the adjacent trusted source node, the risk conduction score between the path nodes, and the risk conduction score between the adjacent target node and the target node, obtains a risk conduction coefficient between the trusted source node and the target node, includes:

s502, judging whether a plurality of conduction paths exist between the credit source node and the target node.

S504, in response to a plurality of conduction paths exist between the credit source node and the target node, determining path nodes in each conduction path.

S506, calculating, for each conduction path, a risk conduction coefficient between the credit giving source node and the target node according to the risk conduction score between the credit giving source node and the adjacent credit giving source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node.

S508, determining the risk conduction coefficient with the largest value in the risk conduction coefficients between the credit source node and the target node in each conduction path, and determining the target risk conduction coefficient between the credit source node and the target node according to the risk conduction coefficient with the largest value.

The conductive path may be a path connecting the trusted source node and the target node. At least one path node may be included in the conductive path.

Specifically, as shown in fig. 7, a may be a target node and B may be a source node. The conductive path between A and B may include A-A1-A2-B, A-A3-B, A-A4-A5-B, A-A6-A7-A8-A9-B. It can be determined that there are multiple conductive paths between a and B. For each conductive path, the risk conductivity coefficient between the trusted source node and the target node in each conductive path may be calculated in the manner as described above in steps S404 to S406. After the risk conduction coefficients between the credit source node and the target node corresponding to each conduction path are calculated, the largest risk conduction coefficient between the credit source node and the target node in all the conduction paths can be selected as the final risk conduction coefficient between the credit source node and the target node.

In this embodiment, when there are multiple conduction paths in the credit source node and the destination node, the risk conduction coefficient between the credit source node and the destination node corresponding to each conduction path is calculated. Therefore, the risk conduction coefficient between the largest credit giving source node and the target node in the conduction path can be selected as the final risk conduction coefficient, the maximum risk conduction coefficient is ensured to be used for measuring the risk intensity of the credit giving source node and the target node, and the accuracy of evaluating the credit giving risk can be improved.

Specifically, the risk conduction value between the credit source node and the adjacent credit source node may be multiplied by the risk conduction value between the adjacent nodes connected in series with the path node, and then multiplied by the risk conduction value between the adjacent target node and the target node, so as to obtain the risk conduction coefficient between the credit source node and the target node.

In this embodiment, by multiplying the risk conduction values, the final risk conduction coefficient is not affected by the number of path nodes in the conduction path, so that the risk conduction coefficient is ensured to be within a certain range, and the risk conduction coefficient can be measured according to a preset risk threshold.

In one embodiment, as shown in fig. 8, the determining, using the service application information, association information associated with the target object includes:

s602, determining association information associated with the object information by utilizing the object information of the target object in the service application information;

s604, performing data processing on the object information and the associated information, wherein the data processing comprises: and screening according to the information value, performing null value processing and data binning processing.

The object information may be identity information input by the target object when transacting business, such as a mobile phone number, a credit card number, a bank card number, liability information, and the like. The binning process may be referred to as a bin process in some embodiments of the present disclosure.

In some exemplary embodiments, the object information may include: customer number, customer name, credential type, credential number, gender, date of birth, nationality, household, marital status, contact details, spouse name, spouse identification number, workplace name, etc.

Specifically, the association information associated with the object information may be determined by matching from the database according to the object information in the service application information input by the target object. For example, if the object information is a mobile phone number, the associated information may be identity information bound to the mobile phone number, or a main number corresponding to the mobile phone number, etc., it will be understood that the foregoing is only used for illustration. Since some invalid information may exist in the associated information and the object information, in order to eliminate interference of the invalid information, data processing may be performed on the object information and the associated information. For example, information values (IV values) of the object information and the associated information are calculated, and the object information and the associated information whose information values are greater than a preset information threshold value are retained. Null values in the object information and the associated information may also be removed. The object information and the associated information may also be subjected to a data division bin process.

In one embodiment, as shown in fig. 9, the establishing a target graph data structure when the target object applies for a service based on the service application information and the association information includes:

s702, combining the object information subjected to data processing and the associated information according to an association relationship to form a first graph data structure of a target object, wherein in the first graph data structure, the object information and the associated information are nodes, and the association relationship between the object information and the associated information is an edge;

s704, according to the association relation between the service type information, the object information and the association information, the service type information is added into the first graph data structure, and a target graph data structure when the target object applies for service is formed.

Wherein the nodes in the first graph data structure comprise: and the node corresponding to the target object and/or the node of the enterprise corresponding to the target object.

Specifically, after the target object inputs the service application information, the target object can be queried according to the service application information to determine whether the target object is a business owner. For example, the method can use the industrial and commercial information to inquire, three codes to integrate, confirm whether enterprises and small micro enterprises exist under the name, if not, then the target object can be determined not to be a business owner, a graph data structure which mainly uses the target object can be established, the object information and the association information can be set as nodes, then the object information and the association information are combined according to the association relation, and the connection is established among the object information with the association relation, the object information and the association information to form edges. The first graph data structure is formed from nodes and edges.

If so, the target object can be confirmed as the enterprise owner. If the identity is the identity of the enterprise owner, an enterprise-based graph data structure can be established. The information of the enterprise corresponding to the target object can be obtained, then a transaction chain, a subsidiary company and the like between enterprises are determined according to the object information, the enterprise information and the association information of the target object, the information is used as a node, and then the information is combined according to the association relation to form an edge. Finally, a first graph data structure is formed. Thus, the nodes in the first graph data structure are different in node type according to different situations.

After the first graph data structure is formed, the target object needs to transact business, so that the business application information can also comprise business type information. The service type information can be used as a node, and the association relationship is used as an edge according to the association relationship between the service type information and the object information and the association information, so that the service type information is added into the first graph data structure to form a target graph data structure when the target object applies for the service.

In some exemplary embodiments, for example, the node formed by the service type information may be a C node, and the nodes formed by the object information and the association information may be X1, X2, and X3. The C node may have an association relationship with X1, or the C node may have an association relationship with X2, and X2 and X3 may have an association relationship, so that a graph data structure as shown in fig. 10 may be formed finally.

In one embodiment, the present disclosure further provides another business risk identification method, as shown in fig. 11, including:

s802, acquiring service application information of a target object.

S804, determining the associated information associated with the object information by utilizing the object information of the target object in the service application information.

S806, performing data processing on the object information and the association information, where the data processing includes: and screening according to the information value, performing null value processing and data binning processing.

S808, combining the object information subjected to data processing and the associated information according to an association relationship to form a first graph data structure of a target object, wherein in the first graph data structure, the object information and the associated information are nodes, and the association relationship between the object information and the associated information is an edge.

S810, adding the service type information into the first graph data structure according to the association relation between the service type information, the object information and the association information to form a target graph data structure when the target object applies for service.

And S812, determining associated attribute information between each adjacent node in the target graph data structure.

S814, calculating to obtain a risk conduction value between each two adjacent nodes according to the attribute information and the preset weight corresponding to the attribute information.

S816, obtaining the path node between the credit source node and the target node.

And S820, determining the risk conduction value between the path nodes according to the risk conduction value between each two adjacent nodes.

S822, calculating to obtain a risk conduction coefficient between the credit giving source node and the target node according to the risk conduction score between the credit giving source node and the adjacent credit giving source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node.

S824, in response to a plurality of conductive paths existing between the trusted source node and the target node, determining path nodes in each conductive path.

S826, calculating a risk conduction coefficient between the credit giving source node and the target node according to the risk conduction score between the credit giving source node and the adjacent credit giving source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node for each conduction path.

S828, determining the risk conduction coefficient with the largest value in the risk conduction coefficients between the credit source node and the target node in each conduction path, and determining the target risk conduction coefficient between the credit source node and the target node according to the risk conduction coefficient with the largest value.

Reference may be made to the foregoing embodiments for specific implementation and limitation in this embodiment, and the detailed description is not repeated here.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the disclosure also provides a service risk identification device for implementing the service risk identification method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of the service risk identification device or devices provided below may refer to the limitation of the service risk identification method hereinabove, and will not be described herein.

In one embodiment, as shown in fig. 12, there is provided a business risk identification device 900, including: an information acquisition module 902, a graph data structure creation module 904, a risk conductivity coefficient calculation module 906, and a risk determination module 908, wherein:

an information obtaining module 902, configured to obtain service application information of a target object;

the map data structure creation module 904 is configured to determine association information associated with the target object by using the service application information, and create a target map data structure of the target object when performing a service application based on the service application information and the association information, where the target map data structure at least includes: a target node corresponding to the target object and a credit source node corresponding to the service type information in the service application information;

A risk conduction coefficient calculation module 906, configured to calculate a risk conduction coefficient between a credit giving source node and the target node according to associated attribute information between adjacent nodes in the target graph data structure and a preset weight corresponding to the attribute information;

the risk determining module 908 is configured to determine that there is a risk in trust of the target object and perform early warning in response to the risk conduction coefficient being greater than a preset risk threshold.

In one embodiment of the apparatus, the risk conductivity coefficient calculating module 906 includes:

In one embodiment of the apparatus, the conductivity calculation module comprises:

In one embodiment of the apparatus, the conductivity coefficient calculation sub-module is further configured to determine a path node in each conductive path in response to the trusted source node and the target node having multiple conductive paths; for each conduction path, calculating to obtain a risk conduction coefficient between the credit giving source node and the target node according to the risk conduction score between the credit giving source node and the adjacent credit giving source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node; and determining the risk conduction coefficient with the largest value in the risk conduction coefficients between the credit source node and the target node in each conduction path, and determining the target risk conduction coefficient between the credit source node and the target node according to the risk conduction coefficient with the largest value.

In an embodiment of the apparatus, the conductivity coefficient calculating submodule is further configured to multiply a risk conductivity value between the credit source node and an adjacent credit source node, a risk conductivity value between the path nodes, and a risk conductivity value between the adjacent target node and the target node, so as to calculate a risk conductivity coefficient between the credit source node and the target node.

In one embodiment of the apparatus, the graph data structure creation module 904 includes:

In one embodiment of the apparatus, the graph data structure creation module 904 further includes:

In an embodiment of the apparatus, the nodes in the first graph data structure comprise: and the node corresponding to the target object and/or the node of the enterprise corresponding to the target object.

The various modules in the business risk identification device described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 13. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing service application information and a target graph data structure. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a business risk identification method.

It will be appreciated by those skilled in the art that the structure shown in fig. 13 is merely a block diagram of a portion of the structure associated with the disclosed aspects and is not limiting of the computer device to which the disclosed aspects apply, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of any of the method embodiments described above when the computer program is executed.

In one embodiment, a computer readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, implements the steps of any of the method embodiments described above.

In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the method embodiments described above.

It should be noted that, the service application information, the object information, the associated information, and the like related to the present application are all information and data authorized by the user or fully authorized by each party, and the collection, the use, and the processing of the related data need to comply with the related laws and regulations and standards of the related country and region.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided by the present disclosure may include at least one of non-volatile and volatile memory, among others. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as Static Random access memory (Static Random access memory AccessMemory, SRAM) or dynamic Random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided by the present disclosure may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors involved in the embodiments provided by the present disclosure may be general-purpose processors, central processing units, graphics processors, digital signal processors, programmable logic, quantum computing-based data processing logic, etc., without limitation thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples have expressed only a few embodiments of the present disclosure, which are described in more detail and detail, but are not to be construed as limiting the scope of the present disclosure. It should be noted that variations and modifications can be made by those skilled in the art without departing from the spirit of the disclosure, which are within the scope of the disclosure. Accordingly, the scope of the present disclosure should be determined from the following claims.

Claims

1. A business risk identification method, the method comprising:

acquiring service application information of a target object;

determining that a risk exists when the target object is subjected to a business acceptance process in response to the risk conduction coefficient being greater than a preset risk threshold, wherein the business is determined according to business application information;

the calculating a risk conduction coefficient between a credit giving source node and the target node according to the associated attribute information between adjacent nodes in the target graph data structure and the preset weight corresponding to the attribute information comprises the following steps:

2. The method according to claim 1, wherein the calculating a risk conduction coefficient between the trusted source node and the target node according to the node between the trusted source node and the target node and the risk conduction value between each adjacent node includes:

3. The method according to claim 2, wherein the calculating a risk conduction coefficient between the trusted source node and the target node according to the risk conduction score between the trusted source node and the adjacent trusted source node, the risk conduction score between the path nodes, and the risk conduction score between the adjacent target node and the target node includes:

4. A method according to claim 2 or 3, wherein said calculating a risk conduction coefficient between a trusted source node and a target node based on a risk conduction score between the trusted source node and an adjacent trusted source node, a risk conduction score between the path nodes, and a risk conduction score between the adjacent target node and the target node comprises:

5. The method of claim 1, wherein the determining association information associated with the target object using the service application information comprises:

6. The method according to claim 5, wherein the creating the target graph data structure when the target object applies for the service based on the service application information and the association information includes:

7. The method of claim 6, wherein the nodes in the first graph data structure comprise: and the node corresponding to the target object and/or the node of the enterprise corresponding to the target object.

8. A business risk identification device, the device comprising:

the risk determination module is used for determining that the target object is trusted to have risk in response to the risk conduction coefficient being larger than a preset risk threshold;

The risk conduction coefficient calculation module includes:

9. The apparatus of claim 8, wherein the conductivity coefficient calculating module comprises:

10. The apparatus of claim 9 wherein the conductivity coefficient calculation sub-module is further configured to determine path nodes in each of the conductive paths in response to the presence of multiple conductive paths for the trusted source node and the destination node; for each conduction path, calculating to obtain a risk conduction coefficient between the credit giving source node and the target node according to the risk conduction score between the credit giving source node and the adjacent credit giving source node, the risk conduction score between the path nodes and the risk conduction score between the adjacent target node and the target node; and determining the risk conduction coefficient with the largest value in the risk conduction coefficients between the credit source node and the target node in each conduction path, and determining the target risk conduction coefficient between the credit source node and the target node according to the risk conduction coefficient with the largest value.

11. The apparatus according to claim 9 or 10, wherein the conductivity coefficient calculation submodule is further configured to multiply a risk conductivity score between the trusted source node and an adjacent trusted source node, a risk conductivity score between the path nodes, and a risk conductivity score between the adjacent target node and the target node, to calculate a risk conductivity coefficient between the trusted source node and the target node.

12. The apparatus of claim 8, wherein the graph data structure creation module comprises:

13. The apparatus of claim 12, the graph data structure creation module further comprising:

14. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.

15. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.

16. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.