CN114066619A

CN114066619A - Guarantee ring risk determination method and device, electronic equipment and storage medium

Info

Publication number: CN114066619A
Application number: CN202111415561.4A
Authority: CN
Inventors: 陈琼
Original assignee: CCB Finetech Co Ltd
Current assignee: CCB Finetech Co Ltd
Priority date: 2021-11-25
Filing date: 2021-11-25
Publication date: 2022-02-18

Abstract

The application discloses a guarantee ring risk determination method and device, electronic equipment and a storage medium. The guarantee ring risk determination method comprises the following steps: acquiring multidimensional service data corresponding to each guarantee object; constructing a guarantee ring knowledge graph based on the multi-dimensional service data; identifying the guarantee ring in the guarantee ring knowledge graph to obtain at least one guarantee ring; calculating a first risk score corresponding to any target guarantee object in each guarantee circle; for each of the wagers, determining a target risk score for the wagers based on the first risk score for each target wagering object in the wagers. By adopting the guarantee ring risk determining method provided by the application, the effect of accurately determining the risk of the guarantee ring can be realized.

Description

Guarantee ring risk determination method and device, electronic equipment and storage medium

Technical Field

The application relates to the field of intelligent finance, in particular to a guarantee ring risk determination method and device, electronic equipment and a storage medium.

Background

The risk of the guarantee circle is one of the hot spot risks which are much concerned by the banking industry, and particularly during the period of economic relaxation and descending, the risk exposure degree of the guarantee circle is increased, so that the development of related enterprises in the guarantee circle is influenced, and even regional financial risks can be caused in serious conditions.

At present, no better and accurate assessment method for the security guarantee risk exists.

Disclosure of Invention

The embodiment of the application aims to provide a guarantee ring risk determination method, a guarantee ring risk determination device, electronic equipment and a storage medium, so that the effect of accurately determining the risk of a guarantee ring is achieved.

The technical scheme of the application is as follows:

in a first aspect, a warranty ring risk determination method is provided, which includes:

acquiring multidimensional service data corresponding to each guarantee object;

constructing a guarantee ring knowledge graph based on the multi-dimensional service data;

identifying the guarantee ring in the guarantee ring knowledge graph to obtain at least one guarantee ring;

calculating a first risk score corresponding to any target guarantee object in each guarantee circle;

for each of the wagers, determining a target risk score for the wagers based on the first risk score for each target wagering object in the wagers.

In a second aspect, there is provided a warranty risk determination apparatus, comprising:

the acquisition module is used for acquiring multidimensional service data corresponding to each guarantee object;

the construction module is used for constructing a guarantee ring knowledge graph based on the multidimensional service data;

the first determination module is used for identifying the guarantee ring in the guarantee ring knowledge graph to obtain at least one guarantee ring;

a second determination module, configured to calculate, for any target security object in each of the security circles, a first risk score corresponding to the target security object;

a third determination module to determine, for each of the wagers, a target risk score for the wagers based on the first risk score for each target wagering object in the wagers.

In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program or instructions stored on the memory and executable on the processor, where the program or instructions, when executed by the processor, implement the steps of any one of the guaranty risk determinations described in the embodiments of the present application.

In a fourth aspect, the present application provides a readable storage medium, on which a program or instructions are stored, where the program or instructions, when executed by a processor, implement the steps of the guaranty risk determination method according to any one of the embodiments of the present application.

The technical scheme provided by the embodiment of the application at least has the following beneficial effects:

according to the guarantee ring risk determining method provided by the embodiment of the application, a guarantee ring knowledge graph is constructed by acquiring multi-dimensional service data corresponding to each guarantee object, guarantee rings in the guarantee ring knowledge graph are identified to obtain at least one guarantee ring, and a first risk score corresponding to a target guarantee object is calculated for any target guarantee object in each guarantee ring; the method comprises the steps of determining a target risk score of a security circle based on the first risk score of each target security object in the security circle, so that multi-dimensional business data of the security circle are obtained, multi-dimensional data are included, the risk identification precision of the security circle is improved, and meanwhile, based on the propagation characteristics of a single target security object in a security circle knowledge graph, the first risk score of the single target security object can be quickly and accurately based on, the target analysis score of the security circle is determined, and the security risk of the security circle is accurately measured.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the application and are not to be construed as limiting the application.

FIG. 1 is a first schematic diagram of a security risk determination method according to an exemplary embodiment of the present application;

FIG. 2 is one of the schematic partial views of a warranty circle knowledge-graph according to an embodiment of the first aspect of the present application;

FIG. 3 is a second schematic diagram of a portion of a warranty circle knowledge-graph according to an embodiment of the first aspect of the present application;

FIG. 4 is a schematic illustration of a security circle according to an embodiment of the first aspect of the present application;

FIG. 5 is a second schematic diagram of a warranty ring risk determination method according to an exemplary embodiment of the present application;

FIG. 6 is a schematic diagram illustrating a security risk determination apparatus according to an exemplary embodiment of the present application;

fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the technical solutions of the present application better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are intended to be illustrative only and are not intended to be limiting. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present application by illustrating examples thereof.

It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples consistent with certain aspects of the present application, as detailed in the appended claims.

The data acquisition, storage, use, processing and the like in the embodiment of the application all conform to relevant regulations of national laws and regulations.

As described in the background section, in the prior art, when determining the security circle risk, the maximum subgraph in the security knowledge graph is obtained mainly by using the maximum clique algorithm, and the maximum subgraph is the security circle closed loop of the complete graph, so when calculating the security circle risk, the associated security risk in the incomplete graph may be missed, resulting in risk omission, and causing the risk of the finally calculated security circle to be inaccurate.

In order to solve the above problems, an embodiment of the present application provides a guarantee ring risk determining method, which includes obtaining multidimensional service data corresponding to each guarantee object, constructing a guarantee ring knowledge graph based on the multidimensional service data, identifying guarantee rings in the guarantee ring knowledge graph to obtain at least one guarantee ring, and calculating a first risk score corresponding to a target guarantee object for any target guarantee object in each guarantee ring; the method comprises the steps of determining a target risk score of a security circle based on the first risk score of each target security object in the security circle, so that multi-dimensional business data of the security circle are obtained, multi-dimensional data are included, the risk identification precision of the security circle is improved, and meanwhile, based on the propagation characteristics of a single target security object in a security circle knowledge graph, the first risk score of the single target security object can be quickly and accurately based on, the target analysis score of the security circle is determined, and the security risk of the security circle is accurately measured.

The guarantee ring risk determination method provided by the embodiment of the present application is described in detail below with reference to the accompanying drawings through specific embodiments and application scenarios thereof.

Fig. 1 is a schematic flowchart of a security risk determination method according to an embodiment of the present disclosure, where an execution subject of the security risk determination method may be a server. The above-described execution body does not constitute a limitation of the present application.

As shown in fig. 1, the guarantee ring risk determination method provided by the embodiment of the present application may include steps 110 to 150.

And step 110, acquiring multidimensional service data corresponding to each guarantee object.

And 120, constructing a guarantee ring knowledge graph based on the multi-dimensional service data.

And step 130, identifying the guarantee ring in the guarantee ring knowledge graph to obtain at least one guarantee ring.

Step 140, for any target security object in each security circle, calculating a first risk score corresponding to the target security object.

Step 150, for each warranty circle, determining a target risk score for the warranty circle based on the first risk score for each target warranty object in the warranty circle.

In the embodiment of the application, a guarantee ring knowledge graph is constructed by acquiring multi-dimensional service data corresponding to each guarantee object, a guarantee ring in the guarantee ring knowledge graph is identified to obtain at least one guarantee ring, and a first risk score corresponding to a target guarantee object is calculated for any target guarantee object in each guarantee ring; the method comprises the steps of determining a target risk score of a security circle based on the first risk score of each target security object in the security circle, so that multi-dimensional business data of the security circle are obtained, multi-dimensional data are included, the risk identification precision of the security circle is improved, and meanwhile, based on the propagation characteristics of a single target security object in a security circle knowledge graph, the first risk score of the single target security object can be quickly and accurately based on, the target analysis score of the security circle is determined, and the security risk of the security circle is accurately measured.

The warranty risk determination method provided by the embodiment of the application is described in detail below.

First, step 110 is introduced to obtain multidimensional service data corresponding to each security object.

The security object may be an object that secures another object. For example, it may be an individual, a business, etc.

In one example, if an object A wants to loan on a credit agency, then an object B is required to guarantee the object A, so that if the object A cannot repay the loan, the credit agency can discuss the loan with the object B, and the object B is the guarantee object.

In some embodiments of the present application, the business data may be data related to a guarantee, such as a name of the guarantee object, an amount of the guarantee object, a guarantee by whom the guarantee object is made, and the like.

In some embodiments of the present application, the business data may be structured data and unstructured data of the acquired security object, and the business data may include, but is not limited to: business data, financial data, credit data, tax data, official documents, and the like.

In some embodiments of the present application, the structured data may be understood as an excel table, in which some information of the guarantee object is stored, for example, name, age, telephone number, guarantee made by whom and amount of guarantee made, etc. of the guarantee object, and the user can obtain corresponding data as long as making a query in the table.

In some embodiments of the present application, the unstructured data may be understood as a word document, for example, a referee document and the like, where the document has a title, an original party, an advanced party, a referee amount, and the like, and when a user needs to know some information in the referee document, the user needs to refer to the referee document, unlike an excel table, and only by querying in the table, the corresponding data can be directly obtained.

In the prior art, when business data corresponding to a guarantee object is obtained, the data is time sliced according to time (for example, according to a month), and the business data at the time is obtained, but in an actual process, most clients have a business data missing state, so that sample data is few, potential unknown risks cannot be identified, and the determined risk of a guarantee circle is not accurate enough.

However, in the embodiment of the present application, by directly obtaining the multidimensional service data of the guarantee object, the service data is not cut in time, and there is no problem of missing part of the service data corresponding to time, so that the accuracy of the risk of the guarantee circle is ensured. Meanwhile, the multidimensional service data of the guarantee object are obtained in the embodiment of the application, so that a data pool is enriched, and the guarantee ring risk identification precision is improved.

In some embodiments of the present application, after the unstructured data is obtained, the unstructured data may be analyzed by using a natural language processing algorithm, so as to convert the unstructured data into structured data. Therefore, the query of the unstructured data can be facilitated, and the user experience is improved.

In some embodiments of the present application, in order to further improve the accuracy of determining the security risk, after step 110, the security risk determining method related to the above may further include:

and cleaning the multidimensional service data to obtain the target multidimensional service data.

The target multidimensional service data may be data obtained by cleaning the multidimensional service data.

In some embodiments of the present application, cleansing multidimensional business data may include, but is not limited to, at least one of:

supplementing missing fields in the multidimensional service data;

deleting the blank space in the multidimensional service data;

unifying abnormal data orders of magnitude in the multidimensional service data;

and deleting repeated data in the multidimensional service data.

In the embodiment of the application, the target multi-dimensional business data can be obtained by cleaning the multi-dimensional business data, so that the business data used for calculating the security risk are ensured to be in accordance with the standard, and the accuracy of determining the security risk is further improved.

Next, a step 120 of constructing a guarantee ring knowledge graph based on the multidimensional business data is introduced.

The security circle knowledge map may be a knowledge map having security objects and security relationships between the security objects.

The security circle knowledge map is a directed map, for example, two security objects a and B having a security relationship, and if the corresponding pointing direction is from a to B, the following relationship for performing security between the two security objects is from a to B.

In one example, as shown in fig. 1, fig. 1 is only one guarantee circle in the guarantee circle knowledge graph, each node N1-N5 represents each guarantee object, and the connecting line between N1-N5 represents the guarantee relationship between N1-N5.

With reference to the above example, the guarantee amount may be written between the connection lines of the guarantee objects in fig. 1, attribute information of each guarantee object may be noted on each node (guarantee object), for example, the attribute information may be a name, business information such as registered capital, financial data such as business income and business cost, legal information such as reported times and opening times, credit information such as administrative penalty number and administrative penalty category, and the like.

In some embodiments of the present application, a directed guaranty circle knowledge graph is constructed from multidimensional business data of each guaranty object, and the graph can be used for subsequent analysis of other information.

In some embodiments of the present application, in order to further improve the accuracy of determining the security risk, after step 120, the security risk determining method related to above may further include:

and clustering the guarantee objects with the same investment object in all the guarantee objects in the guarantee circle knowledge graph to form a target guarantee circle knowledge graph.

The investment target may be a target for making an investment in the secured target. For example, in the case where the secured object is a business, the investment object may be a stockholder or the like.

In some embodiments of the present application, the target underwriting knowledge graph may be a underwriting knowledge graph formed by clustering underwriting objects having the same investment object among the various underwriting objects in the underwriting knowledge graph.

In some embodiments of the present application, the security circle knowledge graph may be identified by using a community discovery algorithm, specifically, the security circle knowledge graph may be divided into one community, where a plurality of security circles may be located in the community, and then each security object in the community may be clustered by using a K-means clustering algorithm, specifically, security objects having a common investment object (for example, shareholders) in the community may be clustered, and a security object having a common investment object (for example, shareholders) may be an object having a consistent action relationship.

It can be understood that the guarantee ring knowledge graph is a guarantee object of one area and its guarantee relationship, for example, it may be a southern area in Tianjin city, and the guarantee ring knowledge graph is identified by using a community discovery algorithm, so as to obtain the guarantee object of each cell and its guarantee relationship, and then the guarantee objects with the same investment object in each cell are clustered by using a K-means clustering algorithm, and the guarantee objects with the same investment object are regarded as objects with consistent action relationship.

In some embodiments of the present application, after determining the objects having consistent action relationships, the objects having consistent action relationships may be labeled, for example, if N1 and N2 in fig. 2 are the objects having consistent action relationships, then the consistent action relationship line (line 32) other than the guaranteed relationship line (line 31) between N1 and N2 may be drawn in fig. 2, resulting in the relationship diagram shown in fig. 3.

Correspondingly, after determining the target warranty circle knowledge-graph, step 130 may specifically include: and identifying the guarantee ring in the target guarantee ring knowledge graph to obtain at least one guarantee ring.

Then, step 130 is introduced to identify the warranty rings in the warranty ring knowledge-graph to obtain at least one warranty ring.

In some embodiments of the present application, when calculating the security assurance risk, since the security assurance relationship is a one-to-many security relationship and multiple pairs of crossed security relationships form a interlinked security assurance, in order to eliminate the situation of closed-loop propagation encountered by risk propagation in the security assurance knowledge graph, a depth-first search algorithm may be used to traverse the graph generated by the security assurance relationship, and search all the security relationships until all the security relationships are not closed and not crossed, so that the risk propagation analysis may be facilitated by using the depth-first search algorithm.

Specifically, identifying the guarantee ring in the guarantee ring knowledge graph by using a depth-first search algorithm to obtain at least one guarantee ring, which may include:

setting an initial value of an accessed value of each guarantee object in a guarantee ring knowledge graph and an initial value of a guarantee search value among the guarantee objects with guarantee relations;

repeating the following steps until the guarantee object in the guarantee knowledge map has no guarantee search relationship, and obtaining at least one guarantee circle:

setting a first guarantee object as an initial search object;

acquiring a guarantee search relationship corresponding to the first guarantee object, and updating a guarantee search value of a second guarantee object corresponding to the guarantee search relationship;

the second secured object is updated to the first secured object.

Wherein, the accessed value may be a value for characterizing that the secured object is accessed, and the initial value of the accessed value may be set to 0. For example, if the secured object is accessed once, the initial value is incremented by 1.

The guarantee search value may be a value for representing that a guarantee relationship between the respective guarantee objects having the guarantee relationship is searched, and an initial value of the guarantee search value may be set to 0. For example, the security object a and the security object B have a security relationship, and when the security relationship between the security object a and the security object B is searched once, the initial value is incremented by 1.

In some embodiments of the present application, the first vouching object may be any one of the vouching objects in a vouching knowledge-graph.

The initial search object may begin searching starting with the object.

The vouching search relationship may be used to characterize that the first vouching object and the second vouching object have a vouching relationship.

The second security object may be an object having a security relationship with the first security object, and for example, as shown in fig. 2, if N1 is the first security object, N2 is the second security object.

How the depth first search algorithm determines the guarantee circle is described below as a specific example:

as shown in fig. 2, which is any guarantee circle in the guarantee circle knowledge graph, each node in fig. 2 is attached with an accessed attribute (i.e., isvisit attribute, i.e., whether accessed or not), and the initial value of the accessed value is set to 0. Let each edge (i.e., the connecting line between two nodes) in fig. 2 be accompanied by a guaranteed search attribute (i.e., an issearch attribute, i.e., whether or not it is searched), and let the initial value of the guaranteed search value be 0.

If N1.isvisit is 0 and there is an edge having a guarantee relationship (N1, N2), that is, N1 and N2 have a guarantee relationship, N1 (i.e., the first guarantee object) is set as an initial search object, and a guarantee search relationship corresponding to N1 (i.e., R1) is selected from N1, then the value of R1 (guarantee search value) is added by 1, and after the guarantee search relationship corresponding to N1 is searched, N2 (the second guarantee object) can be searched, then N2 is accessed, and the accessed value of N2 is added by 1, that is, N1.isvisit is 1. Then N2 is updated as the first guarantee object, the search is started from N2, the following accesses are continued until no edge of the Ni node can go, and the traversal is finished to obtain at least one guarantee circle, as shown in fig. 4, which is an unclosed and uncrossed guarantee circle.

It should be noted that, in the process of traversing the nodes, if there are multiple paths that can be taken, the guarantee ring that passes through more nodes is selected as the final guarantee ring. For example, each node in fig. 2 is traversed, and the following path is followed: N1-R1-N2-R3-N3(-R2) (because the obtained guarantee circle is not closed, so that R2 is not taken), the path passes through 3 nodes of N1, N2 and N3, if the user walks according to the guarantee circle formed in the above-mentioned fig. 4, the path passes through 5 guarantee circles of N1, N2, N3, N4 and N5, and when the guarantee circle is selected, the algorithm automatically selects the guarantee circle passing through the most nodes as the final guarantee circle.

In another embodiment, if different paths are traversed and the number of nodes traversed is the same during the traversal of the nodes, one path can be selected as the final guarantee circle. For example, when the nodes in fig. 2 are traversed and the nodes are walked to form the security circles in fig. 4, the nodes pass through 5 security circles of N1, N2, N3, N4, and N5. If the walking is carried out according to the following path: N1-R1-N2-R3-N3-R6-N5-R5-N4(-N2), and one path can be selected as the final guarantee circle by passing through 5 guarantee circles of N1, N2, N3, N4 and N5.

In the embodiment of the application, at least one guarantee ring is obtained by traversing guarantee objects in the guarantee ring knowledge graph, so that the problem of inaccurate guarantee ring risk determination caused by the fact that multiple pairs of guarantee relationships form a chain of guarantee rings can be avoided, and the accuracy of guarantee ring risk determination is further improved.

Then, step 140 is introduced, where for any target security object in each security circle, a first risk score corresponding to the target security object is calculated.

Here, the target security object may be any one of the security circles for each security circle. For example, if FIG. 4 is a guarantee ring, then N1, N2, N3, N4, and N5 in FIG. 4 may all be target guarantee objects.

The first risk score may be a risk score of the target secured object.

In some embodiments of the present application, step 140 may specifically include:

calculating a self risk value of the target guarantee object based on the negative business data of the target guarantee object;

determining the influence value of the target guarantee object on other guarantee objects in the guarantee circle according to the self risk value of the target guarantee object and the risk adjustment coefficient corresponding to the target guarantee object;

and determining a first risk score corresponding to the target guarantee object based on the self risk value of the target guarantee object and the influence value of the target guarantee object on other guarantee objects in the guarantee circle.

The negative business data may be some business data with a negative target guarantee object, and may include, but is not limited to: the target security object enters a blacklist, the target security object enters an abnormal operation record, the target security object is subjected to administrative punishment, the target security object has bond default, the target security object has tax law violation, the target security object is subjected to industry and commerce reimbursement and the like.

The self risk value may be the risk value of the target vouching object itself.

In some embodiments of the present application, the target collateral object's own risk value may be derived based on negative business data of the target collateral object. The negative business data may be obtained based on a corresponding functional relationship between the negative business data and the risk value of the negative business data, or may be obtained according to a preset algorithm, which is not limited herein.

In some embodiments of the present application, the risk adjustment factor may be a factor of a risk impact value of the target wager object on the wager circle.

In some embodiments of the present application, the risk adjustment coefficient may be determined based on a distance coefficient between the target security object and other security objects in the security circle, a security amount scaling coefficient of the target security object and other security objects, and a capability coefficient of the target security object itself.

In some embodiments of the present application, the distance coefficient may be a parameter for characterizing a distance between a target security object and other security objects in the security circle, the more distant security objects are less affected by the target security object, and the influence value of the target security object on other security objects in the security circle decreases exponentially with increasing distance, and the distance coefficient may be calculated by the following formula (1):

distance coefficient alpha_distanceA distance value (1)

Wherein alpha is_distanceThe parameter may be a preset distance parameter, and the parameter may be set according to the user requirement, which is not limited herein.

In the general case, α_distanceSet to 0.618.

In some embodiments of the present application, the wager amount ratio coefficient of the target wager object to the other wager object may be a coefficient for characterizing the ratio of the wager amounts of the target wager object to the other wager object. The larger the proportion coefficient of the wager amount is, the greater the influence of the target wager object on other wager objects. The secured amount scaling factor may be determined by the following equation (2):

the capital (2) for the object to be secured/the amount of the secured amount is given as the proportional factor of the amount of the secured amount

In some embodiments of the present application, the target vouching object self-ability coefficient may be a coefficient for characterizing the target vouching object self-ability.

In some embodiments of the present application, the self-capability coefficient of the target security object may be obtained based on a pre-trained machine model, and specifically, the self-capability coefficient of the target security object may be directly obtained by inputting the multidimensional service data of the target security object into the pre-trained machine model, so that the self-capability coefficient of the target security object may be obtained quickly and accurately, and the efficiency of determining the security risk is improved.

In some embodiments of the present application, the machine model may be derived based on a machine learning algorithm — a random gradient descent regression algorithm.

In some embodiments of the present application, an influence value of the target security object on other security objects in the security circle may be determined according to the risk value of the target security object itself and the risk adjustment coefficient corresponding to the target security object. Specifically, the influence value of the target security object on other security objects in the security circle may be determined based on the risk value of the target security object and the risk adjustment coefficient corresponding to the target security object according to the following formula (3):

R＝r_ki·c_adjust (3)

wherein R is the influence value of the target guarantee object on other guarantee objects in the guarantee circle; r is_kiIs the self risk value of the target guarantee object, i is the target guarantee object in the guarantee circle, i is 1,2,3, … …, n; c. C_adjustA risk adjustment factor corresponding to the target secured object.

In some embodiments of the present application, after determining the impact value of the target security object on other security objects in the security circle, a first risk score corresponding to the target security object may be determined based on the own risk value of the target security object and the impact value of the target security object on other security objects in the security circle by the following formula (4):

M＝r+R (4)

wherein M is a first risk score; r is the self risk value of the target guarantee object; r is the influence value of the target security object on other security objects in the security circle.

In some embodiments of the present application, the influence of the target guarantee object on the associated guarantee object can be accurately estimated by comprehensively considering various factors such as a distance factor, a guarantee amount factor, and a guarantee object self factor.

In some embodiments of the present application, considering the cumulative effect of the influences, that is, the influences of the events on the same node through different paths will be accumulated, so to further improve the accuracy of the security risk determination, the influence value of a certain target security object on other security objects in the security circle is the sum of the influence values of other source security objects on the target object, and specifically, the influence value of the target object on other security objects in the security circle can be determined by the following formula (5):

wherein R1 is an influence value of the obtained target security object on other security objects in the security circle in consideration of the cumulative effect of the influence; k is a source guarantee object for the target guarantee object, k is 1,2,3, … …, n; i is a target guarantee object.

In the embodiment of the application, the self risk value of the target guarantee object is calculated based on the negative business data of the target guarantee object; determining the influence value of the target guarantee object on other guarantee objects in the guarantee circle according to the self risk value of the target guarantee object and the risk adjustment coefficient corresponding to the target guarantee object; and determining a first risk score corresponding to the target guarantee object based on the self risk value of the target guarantee object and the influence value of the target guarantee object on other guarantee objects in the guarantee circle, so that the accurate first risk score of the target guarantee object can be obtained, and the accurate risk of the guarantee circle is further obtained.

Finally, step 150 is introduced, for each warranty circle, determining a target risk score for the warranty circle based on the first risk score for each target warranty object in the warranty circle.

Wherein the target risk score may be a score for evaluating the risk of the warranty ring.

In some embodiments of the present application, step 150 may specifically include:

accumulating the first risk score of each target guarantee object in the guarantee circle aiming at each guarantee circle to obtain a second risk score corresponding to the guarantee circle;

and determining a target risk score of the warranty ring according to the second risk score.

The second risk score may be a score obtained by adding the first risk scores of the target security subjects in the security circle.

In some embodiments of the present application, the second risk score may be derived by equation (6) as follows:

wherein R2 is a second risk score; l is the number of target guarantee objects in the guarantee circle, and l is 1,2,3, … …, n; rl is the first risk score corresponding to each target security object in the security circle.

In some embodiments of the present application, Rl is R in the above equation (4) in the case where the target security object has only one source security object, and Rl is R1 in the above equation (5) in the case where the target security object has a plurality of source security objects.

In some embodiments of the present application, after the second risk score is determined, the second risk score may be taken as the target risk score.

In the embodiment of the application, for each guarantee circle, the first risk scores of each target guarantee object in the guarantee circle are accumulated to obtain the second risk score corresponding to the guarantee circle, and the target risk score of the guarantee circle can be accurately determined according to the second risk score, so that the accurate guarantee circle risk can be obtained.

In some embodiments of the present application, since in some security circles, the security objects in the security circle have a consistent action relationship in addition to the security relationship, so as to further improve the accuracy of the security circle risk determination, before step 150, the security circle risk determination method as described above may further include:

for each guarantee circle, a scale attribute value of the guarantee circle and a guarantee amount attribute value of the guarantee circle are set.

Wherein, the scale attribute value can be an attribute value for representing the scale of the guarantee ring, the attribute value can be determined based on the guarantee relationship in the guarantee ring and the consistent action relationship, and the target risk score of the guarantee ring can be increased in response when the number of the consistent actors and guarantee relatives involved is larger.

The wager amount attribute value may be an attribute value that characterizes a total wager amount for the wager circle, the greater the total wager amount, the more responsive the target risk score for the wager circle will increase.

Correspondingly, the step 150 may specifically include:

and determining a target risk score of the guarantee circle according to the second risk score, the scale attribute value of the guarantee circle and the guarantee amount attribute value of the guarantee circle.

In the embodiment of the application, the scale attribute value of the guarantee ring and the guarantee amount attribute value of the guarantee ring are set for each guarantee ring, so that the scale attribute value of the guarantee ring and the guarantee amount attribute value of the guarantee ring can be comprehensively considered, and the accurate target risk score of the guarantee ring can be obtained.

In some embodiments of the present application, in order to make the user more intuitively aware of the risk type of each security circle, after step 150, the security circle risk determination method as described above may further include:

for each guaranty circle, determining a risk type for the guaranty circle based on the target risk score for the guaranty circle.

In some embodiments of the present application, the risk type of each warranty ring may be the risk type to which each warranty ring belongs, and may be, for example, three types, normal, suspicious, and loss.

In some embodiments of the present application, determining the risk type of each guaranty may be performed by:

(1) and grading the guarantee ring by adjusting the skewness coefficient and the bee degree coefficient according to the beta distribution algorithm.

In some embodiments of the present application, how to grade the security circle by adjusting the skewness coefficient and the waviness coefficient according to the beta distribution algorithm belongs to the prior art, and is not described herein again.

(2) And comparing the target risk score of each guarantee ring with a preset score threshold, and determining the risk type of each guarantee ring based on the comparison result.

In one example, a security circle is preset to be normal when the target risk score of the security circle is less than or equal to 0.5, is greater than 0.5, is less than 0.7, is suspicious, and is lost when the target risk score of the security circle is greater than or equal to 0.7, and then the security circle is normal if the target risk score of the security circle is 0.4.

In the embodiment of the application, the risk type of the guarantee ring is determined based on the target risk score of the guarantee ring aiming at each guarantee ring, so that a user can know the risk type of each guarantee ring visually, and the user experience is improved.

In some embodiments of the present application, in the process of calculating the target risk score of the security circle, new data may be generated, and thus the new data needs to be calculated. In the prior art, when newly added data is faced, the newly added data must be added again, and then the data including the newly added data is calculated once again, so that huge time and resources are consumed.

In order to solve the above problem, in some embodiments of the present application, after step 150, the above-mentioned guarantee risk determination method may further include:

and under the condition that the newly added information is determined, calculating a newly added risk score corresponding to the newly added information.

Wherein, the new information may be newly added information.

The new risk score may be a new risk score corresponding to the new information, that is, a risk score obtained by calculating only the new information.

Correspondingly, step 150 may specifically include: and determining the target risk score of the guarantee ring based on the newly increased risk score corresponding to the newly increased information.

In the embodiment of the application, under the condition that the newly added information is determined, only the newly added risk score corresponding to the newly added information is calculated; and determining the target risk score of the guarantee ring based on the newly-increased risk score corresponding to the newly-increased information, and thus, only calculating the newly-increased risk score corresponding to the newly-increased information to further obtain the target risk score of the guarantee ring without carrying out full calculation, thereby improving the calculation efficiency of the target risk score.

In some embodiments of the present application, the new information may be a new guarantee object, or may also be new negative service data, and different methods are executed for different pieces of new information to determine a target risk score of a guarantee ring, which specifically may be:

(1) new information as new guarantee object

In the case that it is determined that there is new information, calculating a new risk score corresponding to the new information may specifically be:

under the condition that a new guarantee object is determined, calculating new influence values of the new guarantee object on other guarantee objects in the guarantee circle;

the determining of the target risk score of the security circle based on the newly added risk score corresponding to the newly added information includes:

and scoring the newly increased influence value and the initial risk of the guarantee ring to obtain a target risk score of the guarantee ring.

The new influence value may be an influence value of the new object on other security objects in the security circle, and the specific calculation mode may refer to formula (4) or (5) to calculate the new influence value.

The initial risk score may be a target risk score of the security circle obtained before adding the additional information (i.e., a target risk score obtained by equation (6) above).

In some embodiments of the present application, after obtaining a new influence value of the new security object on other security objects in the security circle, the new influence value may be accumulated with the initial risk score of the security circle to obtain a target risk score of the security circle.

(2) New information is added negative service data

under the condition that the newly added negative business data exist, calculating a newly added self risk value of a newly added guarantee object;

calculating new influence values of the new guarantee objects on other guarantee objects in the guarantee ring based on the new self risk values;

and accumulating the newly added influence value and the initial risk score of the guarantee ring to obtain the target risk score of the guarantee ring.

If negative business data of a certain security guarantee object in a certain security guarantee circle is newly added, the newly added self-risk value can be a new self-risk value generated by the newly added negative business data.

In some embodiments of the present application, for a certain security object in a certain security circle, if negative service data of the security object is newly added, a newly added self-risk value generated by the newly added negative service data needs to be calculated, based on the newly added self-risk value, only a newly added influence value generated by the newly added self-risk value may be calculated, and then the newly added influence value and an initial risk score of the security circle are accumulated to obtain a target risk score of the security circle.

In some embodiments of the present application, in order to implement the guarantee ring risk determination method, Python may be used for programming, and the guarantee ring risk determination method is implemented based on a written program, and the specific programming is implemented based on the following procedures:

step 510, data is acquired.

In some embodiments of the present application, step 510 is the same as step 110 described above, and is not described herein again.

Step 520, constructing a guarantee ring knowledge graph and obtaining a guarantee ring of knowledge.

In some embodiments of the present application, step 520 is the same as step 120-step 130 described above, and is not described herein again.

Step 530, the risk assessment calculation of the guarantee subject is carried out (namely, the target risk score of the guarantee circle is calculated).

In some embodiments of the present application, step 530 is the same as step 140 to step 150, and is not described herein again.

And 540, determining the risk score of the guarantee ring based on the initial risk score, the scale attribute value and the guarantee amount attribute value, and grading the guarantee ring according to the score.

In some embodiments of the present application, step 540 corresponds to determining the security coverage target risk score based on the initial risk score, the scale attribute value, and the security amount attribute value, and classifies the security coverage according to the security coverage target risk score, which is not described herein again.

In the guarantee ring risk determination method provided in the embodiment of the present application, the execution subject may be a guarantee ring risk determination device, or a control module in the guarantee ring risk determination device for executing the guarantee ring risk determination method.

Based on the same inventive concept as the guarantee ring risk determination method, the application also provides a guarantee ring risk determination device. The security risk determination device provided by the embodiment of the present application is described in detail below with reference to fig. 6.

Fig. 6 is a schematic diagram illustrating a configuration of a warranty risk determination apparatus according to an exemplary embodiment.

As shown in fig. 6, the warranty risk determination apparatus 600 may include:

an obtaining module 610, configured to obtain multidimensional service data corresponding to each guarantee object;

a construction module 620, configured to construct a guarantee ring knowledge graph based on the multidimensional service data;

a first determining module 630, configured to identify a guarantee ring in the guarantee ring knowledge graph to obtain at least one guarantee ring;

a second determining module 640, configured to calculate, for any target security object in each of the security circles, a first risk score corresponding to the target security object;

a third determining module 650 for determining, for each of said wagers, a target risk score for said wagers based on said first risk score for each target wagering object in said wagers.

In the embodiment of the application, the acquisition module is used for acquiring multi-dimensional service data corresponding to each guarantee object, the construction module is used for constructing a guarantee ring knowledge graph based on the multi-dimensional service data, the first determination module is used for identifying guarantee rings in the guarantee ring knowledge graph to obtain at least one guarantee ring, and the second determination module is used for calculating a first risk score corresponding to a target guarantee object aiming at any target guarantee object in each guarantee ring; the third determining module determines the target risk score of the security circle based on the first risk score of each target security object in the security circle, so that the multidimensional service data of the security object is obtained, the multidimensional data are included, the risk identification precision of the security circle is favorably improved, and meanwhile, the first risk score of the single target security object can be quickly and accurately based on the propagation characteristic of the single target security object in the security circle knowledge map, the target analysis score of the security circle is determined, and the security risk of the security circle is accurately measured.

In some embodiments of the present application, to further improve the accuracy of the warranty risk determination, the second determining module 640 may be specifically configured to:

calculating a self risk value of each target guarantee object in each guarantee circle based on negative business data of the target guarantee object;

determining, for any one target security object in each of the security circles, an influence value of the target security object on other security objects in the security circle according to the self risk value of the target security object and a risk adjustment coefficient corresponding to the target security object; wherein the risk adjustment factor is determined based on a distance factor between the target security object and another security object in the security circle, a security amount ratio factor between the target security object and the another security object, and a capability factor of the target security object;

and for any target guarantee object in each guarantee circle, determining a first risk score corresponding to the target guarantee object based on the self risk value of the target guarantee object and the influence value of the target guarantee object on other guarantee objects in the guarantee circle.

In some embodiments of the present application, to further improve the accuracy of the warranty risk determination, the third determination module 650 may include:

a first determining unit configured to accumulate the first risk score of each target security object in the security circle for each of the security circles to obtain a second risk score corresponding to the security circle;

and the second determining unit is used for determining the target risk score of the guarantee ring according to the second risk score.

In some embodiments of the present application, in order to further improve the accuracy of the warranty risk determination, the warranty risk determination apparatus mentioned above may further include:

a setting module, configured to set, for each of the guarantee circles, a scale attribute value of the guarantee circle and a guarantee amount attribute value of the guarantee circle;

correspondingly, the second determining unit may be specifically configured to:

In some embodiments of the present application, in order to make the user more intuitively aware of the risk type of each security circle, the security circle risk determining apparatus mentioned above may further include:

a fourth determination module to determine, for each of the guaranties, a risk type for the guaranty based on the target risk score for the guaranty.

the clustering module is used for clustering guarantee objects with the same investment object in all guarantee objects in the guarantee ring knowledge graph to form a target guarantee ring knowledge graph;

correspondingly, the first determining module 630 may specifically be configured to:

and identifying the guarantee ring in the target guarantee ring knowledge graph to obtain at least one guarantee ring.

In some embodiments of the present application, to further improve the accuracy of the warranty risk determination, the first determining module 630 may be specifically configured to:

setting an initial value of an accessed value of each guarantee object in the guarantee circle knowledge graph and an initial value of a guarantee search value among the guarantee objects with guarantee relations;

repeating the following steps until the guarantee object in the guarantee knowledge graph has no guarantee search relationship, and obtaining at least one guarantee circle:

setting a first guarantee object as an initial search object;

acquiring a guarantee search relationship corresponding to the first guarantee object, and updating a guarantee search value of a second guarantee object corresponding to the guarantee search relationship; wherein the vouching search relationship characterizes that the first vouching object and the second vouching object have a vouching relationship;

and updating the second guarantee object to be the first guarantee object.

the calculation module is used for calculating a newly added risk score corresponding to newly added information under the condition that the newly added information is determined;

and the third determining module is specifically configured to determine a target risk score of the guarantee ring based on the newly-added risk score corresponding to the newly-added information.

In some embodiments of the present application, the new addition information includes: newly adding a guarantee object;

under the condition that the newly added information is determined, calculating a newly added risk score corresponding to the newly added information includes:

under the condition that a new guarantee object is determined, calculating a new influence value of the new guarantee object on other guarantee objects in the guarantee circle;

determining the target risk score of the guarantee ring based on the newly-added risk score corresponding to the newly-added information, wherein the determining the target risk score of the guarantee ring comprises the following steps:

and accumulating the newly added influence value and the initial risk score of the guarantee ring to obtain a target risk score of the guarantee ring.

In some embodiments of the present application, the new addition information includes: newly adding negative service data;

under the condition that the newly added negative service data exist, calculating a newly added self risk value of the newly added guarantee object;

The guarantee ring risk determination device provided in the embodiment of the present application may be configured to execute the guarantee ring risk determination method provided in each of the above method embodiments, and the implementation principle and the technical effect are similar, and for the sake of brevity, no further description is given here.

Based on the same inventive concept, the embodiment of the application also provides the electronic equipment.

Fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 7, the electronic device may include a processor 701 and a memory 702 storing computer programs or instructions.

Specifically, the processor 701 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured as one or more Integrated circuits implementing an embodiment of the present invention.

Memory 702 may include a mass storage for data or instructions. By way of example, and not limitation, memory 702 may include a Hard Disk Drive (HDD), a floppy Disk Drive, flash memory, an optical Disk, a magneto-optical Disk, tape, or a Universal Serial Bus (USB) Drive or a combination of two or more of these. Memory 702 may include removable or non-removable (or fixed) media, where appropriate. The memory 702 may be internal or external to the integrated gateway disaster recovery device, where appropriate. In a particular embodiment, the memory 702 is non-volatile solid-state memory. In a particular embodiment, the memory 702 includes Read Only Memory (ROM). Where appropriate, the ROM may be mask-programmed ROM, Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), electrically rewritable ROM (EAROM), or flash memory or a combination of two or more of these.

The processor 701 may implement any of the warranty risk determination methods described in the embodiments above by reading and executing computer program instructions stored in the memory 702.

In one example, the electronic device may also include a communication interface 703 and a bus 710. As shown in fig. 6, the processor 701, the memory 702, and the communication interface 703 are connected by a bus 710 to complete mutual communication.

The communication interface 703 is mainly used for implementing communication between modules, devices, units, and/or devices in the embodiment of the present invention.

Bus 710 includes hardware, software, or both to couple the components of the electronic device to each other. By way of example, and not limitation, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hypertransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus or a combination of two or more of these. Bus 710 may include one or more buses, where appropriate. Although specific buses have been described and shown in the embodiments of the invention, any suitable buses or interconnects are contemplated by the invention.

The electronic device may execute the guarantee ring risk determination method in the embodiment of the present invention, so as to implement the guarantee ring risk determination method described in fig. 1 or fig. 5.

In addition, in combination with the guarantee risk determination method in the foregoing embodiment, the embodiment of the present invention may be implemented by providing a readable storage medium. The readable storage medium having stored thereon program instructions; the program instructions, when executed by a processor, implement any of the warranty risk determination methods of the above embodiments.

It is to be understood that the invention is not limited to the specific arrangements and instrumentality described above and shown in the drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present invention are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications and additions or change the order between the steps after comprehending the spirit of the present invention.

The functional blocks shown in the above-described structural block diagrams may be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the invention are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.

It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.

As described above, only the specific embodiments of the present invention are provided, and it can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. It should be understood that the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present invention, and these modifications or substitutions should be covered within the scope of the present invention.

Claims

1. A warranty risk determination method, characterized in that said method comprises:

acquiring multidimensional service data corresponding to each guarantee object;

2. The method of claim 1, wherein said calculating a first risk score for said target secured object comprises:

determining the influence value of the target guarantee object on other guarantee objects in the guarantee ring according to the self risk value of the target guarantee object and the risk adjustment coefficient corresponding to the target guarantee object; wherein the risk adjustment factor is determined based on a distance factor between the target security object and another security object in the security circle, a security amount ratio factor between the target security object and the another security object, and a capability factor of the target security object;

3. The method of claim 1, wherein said determining, for each said wager circle, a target risk score for said wager circle based on said first risk score for each target wager object in said wager circle comprises:

accumulating the first risk score of each target guarantee object in the guarantee circle to obtain a second risk score corresponding to the guarantee circle;

and determining a target risk score of the guarantee ring according to the second risk score.

4. The method of claim 3, wherein prior to said calculating, for any target wagering object in each of said wagers, a corresponding first risk score for said target wagering object, said method further comprises:

setting a scale attribute value of the guarantee circle and a guarantee amount attribute value of the guarantee circle for each guarantee circle;

determining a target risk score for the warranty circle according to the second risk score, comprising:

5. The method of any one of claims 1-4, wherein after said determining a target risk score for said warranty, said method further comprises:

for each of the guaranties, determining a risk type for the guaranty based on the target risk score for the guaranty.

6. The method of any of claims 1-4, wherein after said building a warranty knowledge-graph based on said business data, said method further comprises:

clustering the guarantee objects with the same investment object in all the guarantee objects in the guarantee circle knowledge graph to form a target guarantee circle knowledge graph;

the identifying the guarantee ring in the guarantee ring knowledge graph to obtain at least one guarantee ring comprises the following steps:

7. The method of any of claims 1-4, wherein said identifying a guaranty circle in said guaranty circle knowledge-graph to obtain at least one guaranty circle comprises:

setting a first guarantee object as an initial search object;

and updating the second guarantee object to be the first guarantee object.

8. The method of any one of claims 1-4, wherein after said determining a target risk score for said warranty, said method further comprises:

under the condition that the newly added information is determined, calculating a newly added risk score corresponding to the newly added information;

the determining a target risk score for the warranty circle comprises:

and determining the target risk score of the guarantee ring based on the newly increased risk score corresponding to the newly increased information.

9. The method of claim 8, wherein the addition information comprises: newly adding a guarantee object;

10. The method of claim 8, wherein the addition information comprises: newly adding negative service data;

11. A warranty claim risk determination apparatus, characterized in that said apparatus comprises:

12. An electronic device comprising a processor, a memory, and a program or instructions stored on the memory and executable on the processor, the program or instructions when executed by the processor implementing the steps of the warranty risk determination method of any of claims 1-10.

13. A readable storage medium, storing thereon a program or instructions which, when executed by a processor, carry out the steps of the warranty risk determination method according to any of claims 1-10.

14. A computer program product, wherein instructions in the computer program product, when executed by a processor of an electronic device, enable the electronic device to perform the lap risk determination method of any one of claims 1-10.