Data processing method and device, equipment and storage medium

Abstract

The embodiment of the invention discloses a data processing method which can be applied to the technical field of computers, and the method comprises the steps: determining statistical data of each risk index of a target account under at least one risk statistical dimension; for each risk statistical dimension, determining a risk score of the target account corresponding to the risk statistical dimension according to the relative importance degree between the risk indexes under the risk statistical dimension and statistical data of each risk index of the target account under the risk statistical dimension; determining a comprehensive risk score of the target account according to the relative importance degree between the risk statistical dimensions and the risk score of the target account corresponding to each risk statistical dimension; and determining the risk level of the target account according to the comprehensive risk score. According to the embodiment of the invention, the risk level of the target account can be accurately determined, and the applicability is high.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a data processing method, device, equipment and storage medium. Background Art

[0002] With the rapid development of the Internet, Internet services such as social networking, shopping, and games have penetrated into all aspects of life and work. Usually, people need to log in to the corresponding business platform through a registered account to experience the corresponding Internet services. In this process, some accounts usually obtain benefits through improper means, causing significant losses to the business platform.

[0003] Based on this, how to effectively and accurately assess the risk of an account has become an urgent problem that needs to be solved. Summary of the invention

[0004] The embodiments of the present application provide a data processing method, apparatus, device, and storage medium, which can effectively and accurately determine the risk level of a target account and have high applicability.

[0005] On the one hand, an embodiment of the present application provides a data processing method, the method comprising:

[0006] Determine the statistical data of each risk indicator of the target account under at least one risk statistical dimension;

[0007] For each of the risk statistical dimensions, determine the risk score of the target account corresponding to the risk statistical dimension according to the relative importance of each risk indicator under the risk statistical dimension and the statistical data of each risk indicator of the target account under the risk statistical dimension;

[0008] Determine a comprehensive risk score of the target account according to the relative importance of each of the risk statistical dimensions and the risk score of the target account corresponding to each of the risk statistical dimensions;

[0009] Based on the above comprehensive risk score, determine the risk level of the above target account.

[0010] On the other hand, an embodiment of the present application provides a data processing device, the device comprising:

[0011] A data acquisition module, used to determine the statistical data of each risk indicator of the target account under at least one risk statistical dimension;

[0012] A risk calculation module, for determining, for each of the risk statistical dimensions, a risk score of the target account corresponding to the risk statistical dimension according to the relative importance of each risk indicator under the risk statistical dimension and the statistical data of each risk indicator of the target account under the risk statistical dimension;

[0013] The risk calculation module is used to determine the comprehensive risk score of the target account according to the relative importance of each of the risk statistical dimensions and the risk score of the target account corresponding to each of the risk statistical dimensions;

[0014] The level determination module is used to determine the risk level of the target account according to the comprehensive risk score.

[0015] On the other hand, an embodiment of the present application provides an electronic device, including a processor and a memory, wherein the processor and the memory are connected to each other;

[0016] The memory is used to store computer programs;

[0017] The processor is used to execute the data processing method provided in the embodiment of the present application when calling the computer program.

[0018] On the other hand, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the data processing method provided by the embodiment of the present application.

[0019] On the other hand, an embodiment of the present application provides a computer program product, which includes a computer program. When the above computer program is executed by a processor, it implements the data processing method provided by the embodiment of the present application.

[0020] In the embodiment of the present application, for each risk statistical dimension, the risk score of the target account corresponding to the risk statistical dimension can be determined based on the statistical data of the target account under each risk indicator, taking into account the relative importance of the risk indicators under the risk statistical dimension. Further, the comprehensive risk score of the target account can be obtained based on the risk score of the target account corresponding to each risk statistical dimension, taking into account the relative importance of each risk statistical dimension, so that the risk level of the target account can be accurately and effectively determined based on the comprehensive risk score, which has high applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a schematic diagram of the network structure of the data processing method provided in the embodiment of the present application;

[0023] Figure 2 It is a flowchart of a data processing method provided in an embodiment of the present application;

[0024] Figure 3 It is one of the schematic diagrams of the judgment matrix provided in the embodiment of the present application;

[0025] Figure 4 This is the second schematic diagram of the judgment matrix provided in the embodiment of the present application;

[0026] Figure 5 This is the third schematic diagram of the judgment matrix provided in the embodiment of the present application;

[0027] Figure 6 is a schematic diagram of a scenario for determining a risk score provided in an embodiment of the present application;

[0028] Figure 7 is a schematic diagram of a scenario for determining a comprehensive risk score provided in an embodiment of the present application;

[0029] Figure 8 is a structural schematic diagram of a data processing device provided in an embodiment of the present application;

[0030] Fig. 9 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0032] The data processing method provided in the embodiment of the present application can be applied to the field of computer technology to determine the risk level of an account. For example, the data processing method provided in the embodiment of the present application can be applied to an account management platform (such as a merchant platform, a game management platform, etc.) to determine the risk level of accounts of many users, thereby providing risk warnings in a timely manner.

[0033] See also Figure 1 , Figure 1 Schematic diagram of the network structure of the data processing method provided in the embodiment of the present application. Figure 1 As shown, for any account, such as the target account, the device 20 can obtain the statistical data of each risk indicator of the target account under at least one risk statistical dimension. For any risk statistical dimension, the device 20 can determine the risk score of the target account in the risk statistical dimension based on the relative importance of each risk indicator of the risk statistical dimension and the statistical data of each risk indicator of the target account under the risk statistical dimension.

[0034] Furthermore, device 20 can determine the comprehensive risk score of the target account based on the relative importance of each risk statistical dimension and the risk score of the target account in each risk statistical dimension. Device 20 can then determine the risk level of the target account based on the comprehensive risk score of the target account, thereby facilitating device 20 to process the target account according to the risk level of the target account.

[0035] Among them, device 20 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, as well as big data and artificial intelligence platforms.

[0036] Among them, the terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, a car terminal, an aircraft, a smart home appliance (such as a smart TV) or a wearable device, etc.

[0037] See also Figure 2 , Figure 2 Schematic diagram of the data processing method provided in the embodiment of the present application. Figure 2 As shown, the data processing method provided in the embodiment of the present application may specifically include the following steps:

[0038] Step S21: Determine the statistical data of each risk indicator of the target account under at least one risk statistical dimension.

[0039] In some feasible implementations, each risk statistical dimension is used to describe a behavior type or a data type of the target account. As an example, the above risk statistical dimensions include but are not limited to risk interception dimension, bad debt refund dimension, marketing activity dimension, risk recharge dimension, recharge behavior dimension, active behavior dimension, gift behavior dimension, etc., which can be determined according to actual needs and are not limited here.

[0040] Each risk statistical dimension includes at least one risk indicator, and each risk indicator is used to describe a specific sub-dimension of the corresponding risk statistical dimension.

[0041] As an example, the risk interception dimension is used to describe the situation in which the target account is intercepted within a certain period of time (such as within a week). The risk interception dimension includes but is not limited to the number of times the target account is intercepted by the risk interception strategy, the number of channels for risk interception of the target account, the amount involved in the risk interception of the target account, the number of days the target account is intercepted, the number of devices used when the target account is intercepted, the number of different areas where the target account is located when it is intercepted, the number of risk areas where the target account is located when it is intercepted, the number of trading accounts used when the target account is intercepted, and other risk indicators. The specific risk indicators can be determined based on the actual application scenario requirements and are not limited here.

[0042] As an example, the bad debt refund dimension is used to describe the bad debt situation of the target account in a certain period of time (such as as of the current moment). The bad debt refund dimension includes but is not limited to risk indicators such as the number of days the target account generates bad debts, the number of times the target account generates bad debts, the amount of bad debts generated by the target account, the number of channels through which the target account generates bad debts, and the refund rate of bad debts generated by the target account. The specific risk indicators can be determined based on the actual application scenario requirements and are not restricted here.

[0043] As an example, the marketing activity dimension is used to describe the risk situation of the target account in marketing activities in a certain period of time (such as the most recent year). The marketing activity dimension includes but is not limited to the number of accounts that the target account uses to profit by exploiting loopholes, the amount of money the target account obtains by exploiting loopholes, the number of times the target account uses loopholes to profit, the number of days the target account uses loopholes to profit, and other risk indicators. The specific risk indicators can be determined based on the actual application scenario requirements and are not restricted here.

[0044] As an example, the risk proxy recharge dimension is used to describe the relevant situations of the target account as a third party to recharge other accounts in a certain period of time (such as the last month). The risk proxy recharge dimension includes but is not limited to the number of times the target account recharges other accounts as a third party, the amount of recharges the target account as a third party for other accounts, the number of refunds after the target account recharges other accounts as a third party, the refund rate after the target account recharges other accounts as a third party, the refund amount after the target account recharges other accounts as a third party, and other risk indicators. The specific risk indicators can be determined based on the actual application scenario requirements and are not restricted here.

[0045] For example, the risky recharge dimension can also be used to describe the relevant situations of the target account recharging the target account through a third-party account within a certain period of time (such as the last month). The risky recharge dimension includes but is not limited to the number of times the third-party account recharges the target account, the amount of recharge the target account by the third-party account, the number of refunds after the third-party account recharges the target account, the refund rate after the third-party account recharges the target account, the refund amount after the third-party account recharges the target account, and other risk indicators.

[0046] As an example, the recharge behavior dimension is used to describe the recharge status of the target account in a certain period of time (such as the past week). The recharge behavior dimension includes but is not limited to the amount involved when the target account's account device changes during the recharge process, the proportion of the amount involved when the target account's account device changes during the recharge process to the total recharge amount of the target device, the amount involved when the target account's region changes during the recharge process, the proportion of the amount involved when the target account's region changes during the recharge process to the total recharge amount of the target device, the amount involved when the target account's recharge platform changes during the recharge process, the proportion of the amount involved when the target account's recharge platform changes during the recharge process to the total recharge amount of the target account, the number of times the target device's recharge platform changes during the recharge process, and other risk indicators. The specific risk indicators can be determined based on the actual application scenario requirements and are not limited here.

[0047] As an example, the active behavior dimension is used to describe the account activity of the target account in a certain period of time (such as the past week). The active behavior dimension includes but is not limited to risk indicators such as the account activity duration of the target account, the account level of the target account, the login duration of the target account and the proportion of the active duration. The specific risk indicators can be determined based on the actual application scenario requirements and are not limited here.

[0048] As an example, the gifting behavior dimension is used to describe the gifting of virtual resources by the target account in a certain period of time (such as the last month). The gifting behavior dimension includes but is not limited to risk indicators such as the amount of virtual resources donated by the target account to other accounts (such as point amount, number of discounts, etc.), the number of times the target account has donated virtual resources to other accounts, and the ratio of the amount of virtual resources donated by the target account to all resources of the target account. The specific risk indicators can be determined based on the actual application scenario requirements and are not restricted here.

[0049] In this case, when it is necessary to determine the risk level of the target account, the statistical data of each risk indicator of the target account under at least one risk statistical dimension may be determined.

[0050] Step S22: For each risk statistical dimension, determine the risk score of the target account corresponding to the risk statistical dimension based on the relative importance of each risk indicator under the risk statistical dimension and the statistical data of each risk indicator of the target account under the risk statistical dimension.

[0051] In some feasible implementations, for any two risk indicators (such as a first risk indicator and a second risk indicator) under each risk statistical dimension, the relative importance between the first risk indicator and the second risk indicator may include the importance of the first risk indicator compared to the second risk indicator, and the importance of the second risk indicator compared to the first risk indicator.

[0052] For each risk statistical dimension, the relative importance of each risk indicator under the risk statistical dimension can be represented by a first judgment matrix, and each matrix element in the first judgment matrix corresponding to the risk statistical dimension is used to characterize the relative importance of two risk indicators under the risk statistical dimension.

[0053] As an example, the first judgment matrix corresponding to the risk statistical dimension can be expressed as:

[0054]

[0055] The dimension of the first judgment matrix corresponding to the risk statistical dimension is n×n, where n is the number of risk indicators under the risk statistical dimension. In addition, a matrix element in the first judgment matrix corresponding to the risk statistical dimension represents the importance of a risk indicator under the risk statistical dimension compared to a risk indicator.

[0056] Among them, C ij Indicates the importance of risk indicator i compared to risk indicator j under this risk statistical dimension, C ji Indicates the importance of risk indicator j in this risk statistical dimension compared to risk indicator i, C ij and C ji They are reciprocal of each other. ii It represents the importance of risk indicator i and risk indicator i under the risk statistical dimension, which is a first preset value (such as 1).

[0057] Among them, assuming that the risk indicator i and risk indicator j under this risk statistical dimension are equally important, then C ij is the first preset value.

[0058] Among them, it is assumed that the risk indicator i under this risk statistical dimension is more important than the risk indicator j. If the risk indicator i is slightly more important than the risk indicator j, then C ij is the second preset value; if risk indicator i is significantly more important than risk indicator j, then C ijis the third preset value; if risk indicator i is more important than risk indicator j, then C ij is the fourth preset value; if risk indicator i is more important than risk indicator j, then C ij is the fifth preset value; if the importance of risk indicator i is between equally important and slightly important compared to risk indicator j, then C ij is the average of the first preset value and the second preset value; if the importance of risk indicator i compared to risk indicator j is between slightly important and obviously important, then C ij is the average of the second preset value and the third preset value; if the importance of risk indicator i compared to risk indicator j is between obviously important and strongly important, then C ij is the average of the third preset value and the fourth preset value; if the importance of risk indicator i compared to risk indicator j is between obviously important and extremely important, then C ij is the average value of the fourth preset value and the fifth preset value.

[0059] Among them, the first preset value is smaller than the second preset value, the second preset value is smaller than the third preset value, the third preset value is smaller than the fourth preset value, and the fourth preset value is smaller than the fifth preset value.

[0060] As an example, assuming that the first to fifth preset values ​​are 1, 3, 5, 7, and 9 respectively, if the risk indicator i and the risk indicator j under the risk statistical dimension are equally important, then C ij is 1; if risk indicator i is slightly more important than risk indicator j, then C ij is 3; if risk indicator i is significantly more important than risk indicator j, then C ij is 5; if risk indicator i is more important than risk indicator j, then C ij is 7; if risk indicator i is more important than risk indicator j, then C ij is 9; if the importance of risk indicator i is between equally important and slightly important compared to risk indicator j, then C ij is 2; if the importance of risk indicator i compared to risk indicator j is between slightly important and obviously important, then C ij is 4; if the importance of risk indicator i compared to risk indicator j is between obviously important and strongly important, then C ij is 6; if the importance of risk indicator i compared to risk indicator j is between obviously important and extremely important, then C ij is 8.

[0061] As an example, see Figure 3 , Figure 3 It is one of the schematic diagrams of the judgment matrix provided in the embodiment of the present application.

[0062] If the risk indicators under a risk statistical dimension include risk indicators A to risk indicators F, for example, if risk indicator C is slightly more important than risk indicator A, then Figure 3 In the first judgment matrix shown, the matrix element in the third row and first column is 3, and the matrix element in the first row and third column is 1 / 3; if risk indicator B is between equally important and slightly important compared to risk indicator A, then Figure 3 In the first judgment matrix shown, the matrix element in the second row and first column is 2, and the matrix element in the first row and second column is 1 / 2.

[0063] In some embodiments, for each risk statistical dimension, the relative importance between any two risk indicators under the risk statistical dimension can be determined by a pre-trained prediction model.

[0064] For any two risk indicators of each risk statistical dimension (also taking the first risk indicator and the second risk indicator as an example), the characteristic information of the first risk indicator and the second risk indicator can be input into the pre-trained prediction model, and the prediction model outputs a first prediction value used to characterize the importance of the first risk indicator compared to the second risk indicator and a prediction value of the second risk indicator compared to the first risk indicator. The first prediction value and the second prediction value are reciprocals of each other, and one of the first prediction value and the second prediction value is equal to one of the first to fifth preset values, or equal to the average of any two of the first to fifth preset values.

[0065] Based on this, for each risk statistical dimension, the risk indicators under the risk statistical dimension can be combined in pairs, and multiple groups of first prediction values ​​and second prediction values ​​can be obtained through the pre-trained prediction model, and then the first judgment matrix corresponding to the risk statistical dimension can be constructed.

[0066] Optionally, for each risk statistical dimension, the importance between any two risk indicators under the risk statistical dimension is a predetermined original value, and then the corresponding original values ​​are arranged according to the arrangement method of each risk indicator under the risk statistical dimension to obtain a first judgment matrix. That is, for the first risk indicator and the second risk indicator under the risk statistical dimension, the importance of the first risk indicator compared to the second risk indicator is a predetermined original value, and the importance of the second risk indicator compared to the first risk indicator is a predetermined original value, and the two are reciprocals of each other.

[0067] In other words, for each risk statistical dimension, the first judgment matrix corresponding to the risk statistical dimension is a predetermined matrix.

[0068] Optionally, for each risk statistical dimension, when determining the first judgment matrix corresponding to the risk statistical dimension, an initial judgment matrix determined by multiple judgment objects (such as different scoring experts, different prediction models) for the risk statistical dimension can be obtained, and then the matrix elements in the same position in the initial judgment matrix determined by each judgment object are averaged to finally obtain the first judgment matrix corresponding to the risk statistical dimension.

[0069] Among them, the training of the above prediction model can be achieved based on the machine learning (ML) technology in the field of artificial intelligence (AI).

[0070] Among them, artificial intelligence is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence.

[0071] Machine learning is a special study of how computers simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are spread across all areas of artificial intelligence. In the embodiments of the present application, a machine with the ability to output a predicted value for representing relative importance can be trained based on machine learning.

[0072] In some feasible implementations, for each risk statistical dimension, after determining the first judgment matrix corresponding to the risk statistical dimension, the validity of the first judgment matrix corresponding to the risk statistical dimension can be determined first, that is, whether the first judgment matrix corresponding to the risk statistical dimension passes the consistency verification.

[0073] Specifically, the maximum eigenvalue of the first judgment matrix corresponding to the risk statistical dimension may be determined first.

[0074] The maximum characteristic root can be determined by any method in the prior art.

[0075] Further, according to the maximum characteristic root of the first judgment matrix corresponding to the risk statistical dimension and the first number of risk indicators under the risk statistical dimension, the consistency index of the first judgment matrix corresponding to the risk statistical dimension is determined, which can be determined specifically in the following way:

[0076]

[0077] Among them, CI represents the consistency index of the first judgment matrix corresponding to the risk statistical dimension, λ max It represents the maximum characteristic root of the first judgment matrix corresponding to the risk statistical dimension, and n is the number of risk indicators under the risk statistical dimension.

[0078] Further, according to the first number of risk indicators under the risk statistical dimension, a reference consistency indicator of the first judgment matrix corresponding to the risk statistical dimension is determined.

[0079] Among them, different numbers of risk indicators correspond to different reference consistency indicators, and the reference consistency indicators corresponding to different numbers of risk indicators are preset values.

[0080] As an example, the corresponding relationship between the number of risk indicators and the reference consistency indicators can be shown in the following table:

[0081] n 1 2 3 4 5 6 7 8 9 … RI 0 0 0.58 0.90 1.12 1.24 1.32 1.41 1.45 …

[0082] Among them, RI is the reference consistency index.

[0083] For example, when the number of risk indicators under the risk statistical dimension is 5, the reference consistency index corresponding to the risk statistical dimension is 1.12.

[0084] Alternatively, a certain number of initial judgment matrices can be randomly constructed for the risk statistical dimension, so that each matrix element C in each judgment matrix ij Satisfies the above assignment rules, and C ji =1 / C ij , C ii is the first preset value. Then, the average value of the maximum characteristic root of each initial judgment matrix corresponding to the risk statistical dimension can be determined, and the reference consistency index RI corresponding to the risk statistical dimension can be determined by the following method:

[0085]

[0086] Among them, λ ‘ max is the average value of the maximum characteristic roots of each initial judgment matrix corresponding to the risk statistical dimension, and n is the first number of risk indicators under the risk statistical dimension.

[0087] Furthermore, a first ratio of the consistency index of the first judgment matrix corresponding to the risk statistical dimension to the reference consistency index can be determined. If the first ratio is less than the first threshold, it is determined that the first judgment matrix corresponding to the risk statistical dimension passes the consistency verification, that is, it is determined that the first judgment matrix corresponding to the risk statistical dimension passes the validity verification. If the first ratio is greater than or equal to the first threshold, it is determined that the first judgment matrix corresponding to the risk statistical dimension does not pass the consistency verification, that is, it is determined that the first judgment matrix corresponding to the risk statistical dimension does not pass the validity verification. At this time, the first judgment matrix corresponding to the risk statistical dimension is re-adjusted until it passes the validity verification.

[0088] The first ratio of the consistency index of the first judgment matrix corresponding to the risk statistical dimension to the reference consistency index can be expressed in the following manner:

[0089]

[0090] In some feasible implementations, for each risk statistical dimension, after a first judgment matrix corresponding to the risk statistical dimension passes validity verification, the risk weight of each risk indicator under the risk statistical dimension may be determined according to the first judgment matrix.

[0091] Specifically, for each column in the first judgment matrix corresponding to the risk statistical dimension, the matrix elements of the first judgment matrix can be normalized by column to obtain a first intermediate matrix, and each matrix element in the first intermediate matrix is ​​divided by the sum of all matrix elements in the column to obtain a second intermediate matrix. Further, each matrix element in the second intermediate matrix is ​​summed by row to obtain a third intermediate matrix, and finally each matrix element in the third intermediate matrix is ​​divided by the number of risk indicators under the risk statistical dimension to obtain a first weight vector.

[0092] At this time, each element in the first weight vector is the risk weight of a risk indicator under the risk statistical dimension.

[0093] In other words, the risk weight of risk indicator i under this risk statistical dimension can be determined based on the following method:

[0094]

[0095] Among them, C ij is the matrix element in the i-th row and j-th column in the first matrix, i is the index of the risk indicator under the risk statistical dimension, and n is the number of risk indicators under this risk indicator.

[0096] Optionally, for the first judgment matrix corresponding to the risk statistical dimension, the matrix elements of the first judgment matrix can be multiplied row by row to obtain a first column vector, each element of the first column vector can be raised to the nth power to obtain a second column vector, and then each element of the second column vector can be normalized to obtain a second weight vector.

[0097] At this time, each element in the second weight vector is the risk weight of a risk indicator under the risk statistical dimension.

[0098] In other words, the risk weight of risk indicator i under this risk statistical dimension can be determined based on the following method:

[0099]

[0100] Among them, C ij is the matrix element in the i-th row and j-th column in the first matrix, i is the index of the risk indicator under the risk statistical dimension, and n is the number of risk indicators under this risk indicator.

[0101] Optionally, for the first judgment matrix corresponding to the risk statistical dimension, the maximum eigenvalue root of the first judgment matrix and its corresponding eigenvalue vector can be determined, the eigenvalue vector corresponding to the maximum eigenvalue root can be normalized to obtain a third column vector, and then each element of the third column vector can be normalized to obtain a third weight vector.

[0102] At this time, each element in the third weight vector is the risk weight of a risk indicator under the risk statistical dimension.

[0103] Where Hx = λ max x, H is the first judgment matrix, λ max is the maximum eigenvalue of the first judgment matrix, an x-dimensional non-zero column vector, that is, x is the third column vector.

[0104] In some feasible implementations, for each risk statistical dimension, when determining the risk weight of each risk indicator under the risk statistical dimension based on a first judgment matrix corresponding to the risk statistical dimension, the matrix elements of the first judgment matrix can be adjusted according to the statistical data of each risk indicator of each account in the account set under the risk statistical dimension to obtain a second judgment matrix, and then the risk weight of each risk indicator under the risk statistical dimension can be determined based on the second judgment matrix.

[0105] Specifically, for each risk indicator under the risk statistical dimension, the reference weight of the risk indicator can be determined according to the statistical data of the risk indicator for each account in the account set.

[0106] Among them, each account in the account set can be an account whose risk level is currently to be determined (which may include a target account), or a historical account for which statistical data under various risk indicators have been obtained, etc. The specific details can be determined based on the actual application scenario requirements and are not restricted here.

[0107] For each risk indicator under the risk statistical dimension, a second ratio of the statistical data of each account in the account set on the risk indicator to the target statistical data can be determined, where the target data is the sum of the statistical data of all accounts in the account set on the risk indicator. The statistical data used in calculating the second ratio can be standardized data.

[0108] As an example, the second ratio p of the statistical data of the risk indicator v to the target statistical data of the account u in the account set is uv This can be determined by:

[0109]

[0110] Among them, y uv is the standardized statistical data of account u based on risk indicator v, and r is the number of accounts in the account set.

[0111] Furthermore, for each risk indicator under the risk statistical dimension, the information entropy of the risk indicator can be determined according to the second ratio corresponding to each account in the account set.

[0112] Among them, information entropy is used to measure the uncertainty or average uncertainty of the statistical data of different accounts in risk indicators, that is, to describe the degree of uncertainty of risk indicators. The larger the information entropy of a risk indicator, the more uncertain the statistical data of different accounts in the risk indicator.

[0113] Among them, the information entropy of the risk index v can be determined in the following way:

[0114]

[0115] Wherein, k = 1 / ln(r).

[0116] For each risk indicator under the risk statistical dimension, the information entropy redundancy of the risk indicator can be determined according to the information entropy of the risk indicator.

[0117] Among them, the information entropy redundancy of the risk indicator v can be determined in the following way:

[0118] d v =1-e v

[0119] The reference weight of each risk indicator under the risk statistical dimension can be determined according to the information entropy redundancy of each risk indicator under the risk statistical dimension.

[0120] Among them, the information entropy redundancy of each risk indicator under the risk indicator can be summed to obtain the sum of redundancy, and then for each risk indicator of the risk statistical dimension, the information entropy redundancy of the risk indicator can be divided by the sum of redundancy to obtain the reference weight of the risk indicator.

[0121] As an example, the reference weight of the risk indicator v under the risk statistics dimension may be determined based on the following method:

[0122]

[0123] Where n is the number of risk indicators under this risk statistical dimension.

[0124] Further, for each risk statistical dimension, after determining the reference weights of the risk indicators under the risk statistical dimension, at least one target matrix element may be determined from the first judgment matrix corresponding to the risk statistical dimension.

[0125] Among them, each target matrix element and the reference weights of the two risk indicators corresponding to the target matrix element meet the preset conditions.

[0126] In this case, the corresponding target matrix elements may be adjusted according to the reference weights of the two risk indicators corresponding to each target matrix element to obtain the second judgment matrix.

[0127] Specifically, for each matrix element corresponding to two different risk indicators in the first judgment matrix, first define the risk indicator corresponding to the row where the matrix element is located as the third risk indicator, and the unique risk indicator in the column where the matrix element is located as the fourth risk indicator.

[0128] When the matrix element is greater than 1, if the first ratio of the reference weight of the third risk indicator to the reference weight of the fourth risk indicator is greater than 1, and the second ratio obtained after rounding the first ratio is inconsistent with the matrix element, and the second ratio is less than or equal to the fifth preset value (that is, the upper limit value of all matrix elements in the first judgment matrix), then the matrix element is determined as the target matrix element, that is, it is determined that the matrix element and the two risk indicator reference weights corresponding to the matrix element meet the preset conditions.

[0129] At this time, the target matrix element may be replaced by the second ratio.

[0130] When the matrix element is less than 1, if the first ratio of the reference weight of the third risk indicator to the reference weight of the fourth risk indicator is less than 1, and the third ratio obtained after processing the first ratio by the preset processing method is inconsistent with the matrix element, then the matrix is ​​determined as the target matrix element, that is, it is determined that the matrix element and the two risk indicator reference weights corresponding to the matrix element meet the preset conditions.

[0131] At this time, the target matrix element may be replaced by the third ratio.

[0132] The first ratio is processed by a preset processing method, including:

[0133] If the first ratio is any one of the sixth preset value to the thirteenth preset value, the first ratio is determined as the third ratio;

[0134] If the first ratio is not any one of the sixth to thirteenth preset values, the absolute difference between the first ratio and each of the sixth to thirteenth preset values ​​is determined, and the preset value with the smallest absolute difference is determined as the third ratio.

[0135] Among them, the sixth preset value is the reciprocal of the average of the first preset value and the second preset value; the seventh preset value and the second preset value are the reciprocals of each other; the eighth preset value is the reciprocal of the average of the second preset value and the third preset value; the ninth preset value and the third preset value are the reciprocals of each other; the tenth preset value is the reciprocal of the average of the third preset value and the fourth preset value; the eleventh preset value and the fourth preset value are the reciprocals of each other; the twelfth preset value is the reciprocal of the average of the fourth preset value and the fifth preset value; the thirteenth preset value and the fifth preset value are the reciprocals of each other.

[0136] For example, the first to fifth preset values ​​are 1, 3, 5, 7 and 9 respectively, and the sixth to thirteenth preset values ​​are 1 / 2, 1 / 3, 1 / 4, 1 / 5, 1 / 6, 1 / 7, 1 / 8 and 1 / 9 respectively.

[0137] Alternatively, for each matrix element corresponding to two different risk indicators in the first judgment matrix, first define the risk indicator corresponding to the row where the matrix element is located as the third risk indicator, and the unique risk indicator in the column where the matrix element is located as the fourth risk indicator.

[0138] When the matrix element is greater than 1, if the first ratio of the reference weight of the third risk indicator to the reference weight of the fourth risk indicator is greater than 1, and the second ratio obtained after rounding the first ratio is inconsistent with the matrix element, and the second ratio is less than or equal to the fifth preset value (that is, the upper limit value of all matrix elements in the first judgment matrix), then the matrix element is determined as the target matrix element, that is, it is determined that the matrix element and the two risk indicator reference weights corresponding to the matrix element meet the preset conditions.

[0139] At this time, the target matrix element can be replaced by the second ratio, and the association matrix element corresponding to the target matrix element can be replaced by the inverse of the second ratio. The risk indicator corresponding to the row of the association matrix element corresponding to the target matrix element is the fourth risk indicator, and the risk indicator corresponding to the column is the third risk indicator.

[0140] As an example, see Figure 4 , Figure 4 This is the second schematic diagram of the judgment matrix provided in the embodiment of the present application. Figure 4 In the first judgment matrix shown, when the matrix element in the third row and second column is greater than 1, if the first ratio of the reference weights of risk indicator C and risk indicator B is determined to be 5 / 2, the first ratio is rounded to obtain a second ratio of 3.

[0141] At this time, if the second ratio is inconsistent with the matrix element (4) in the third row and second column, the matrix element in the third row and second column can be replaced by 3.

[0142] In this case, the matrix element (1 / 4) in the second row and third column of the first judgment matrix may be replaced by the inverse of the second ratio 1 / 3.

[0143] Based on this method, the matrix elements in the first judgment matrix can be adjusted according to the reference weights of the risk indicators to obtain the second judgment matrix. Figure 4 After adjusting the matrix elements in the third row and second column and the matrix elements in the second row and third column in the first judgment matrix shown, we can get Figure 5 The second judgment matrix shown.

[0144] In some feasible implementations, for each risk statistical dimension, the risk score of the target account corresponding to the risk statistical dimension can be determined based on the statistical data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension.

[0145] For each risk statistical dimension, a target normalization method can be determined based on the statistical data of each risk indicator of the target account under the risk statistical dimension, thereby performing data normalization processing on the statistical data of each risk indicator of the target account under the risk statistical dimension according to the target normalization method to obtain standardized data.

[0146] Then, based on the standardized data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension, the risk score of the target account corresponding to the risk statistical dimension is determined.

[0147] Specifically, for each risk statistical dimension, when determining the target normalization method based on the statistical data of each risk indicator of the target account under the risk statistical dimension, the data skewness of the target account can be determined based on the statistical data of each risk indicator of the target account under the risk statistical dimension. The data skewness is used to describe the degree of skewness of the data distribution of the target account.

[0148] If the data skewness of the target account is within the first normalization interval, the first normalization method is determined as the target normalization method.

[0149] The data skewness S of the target account can be determined based on the following method:

[0150]

[0151] Among them, X i is the statistical data of risk indicator i of the target account under this risk statistical dimension, μ represents the mean of the statistical data of each risk indicator of the target account under this risk statistical dimension, σ represents the standard deviation of the statistical data of each risk indicator of the target account under this risk statistical dimension, and n is the number of risk indicators under this risk statistical dimension.

[0152] The upper limit value of the first standardized interval is the first skewness value, the lower limit value is the second skewness value, and the first skewness value is greater than the second skewness value.

[0153] The first normalization method may be a Min-Max Scaling normalization method or a Z-score normalization method, which is not limited here.

[0154] For example, if the second skewness value is -0.5 and the first skewness value is 0.5, when the data skewness of the target account is between -0.5 and 0.5, it means that the statistical data of each risk indicator of the target account under this risk statistical dimension are relatively symmetrical. At this time, the first normalization method is determined as the target normalization method.

[0155] If the data skewness of the target account is in the second standardization interval or the third standardization interval, the second standardization method is determined as the target standardization method.

[0156] The upper limit of the second standardized interval is the second skewness value, the lower limit is the third skewness value, and the second skewness value is greater than the third skewness value. The lower limit of the third standardized interval is the first skewness value, the upper limit is the fourth skewness value, and the fourth skewness value is greater than the first skewness value.

[0157] The second normalization method may be a decile normalization (Robust Scaling) method.

[0158] For example, the upper limit of the second standardization interval is -0.5, the lower limit is -1, the lower limit of the third standardization interval is 0.5, and the upper limit is 1. When the data skewness of the target account is between -1 and -0.5, or between 0.5 and 1, it means that the statistical data of each risk indicator of the target account under the risk statistical dimension has a certain skewness. At this time, in order to retain the differences between the statistical data, the second standardization method can be determined as the target standardization method.

[0159] If the data skewness of the target account is outside all standardized intervals, the third standardization method is determined as the target standardization method.

[0160] The third normalization method may be a normalization method based on logarithmic transformation or a normalization method based on square root transformation, etc., which is not limited here.

[0161] For example, the upper limit value of the second standardization interval is -0.5, and the lower limit value is -1, the lower limit value of the third standardization interval is 0.5, and the upper limit value is 1, the lower limit value of the first standardization interval is -0.5, and the upper limit value is 0.5. When the data skewness of the target account is greater than 1 or less than -1, it means that the statistical data distribution of each risk indicator of the target account under this risk statistical dimension is very uneven. In order to make the statistical data close to the Zhengtai distribution, the third standardization method can be determined as the target standardization method.

[0162] As an example, if the target standardization method is the third standardization method, and the third standardization method is a standardization method based on logarithmic transformation, the standardized data G of the risk indicator i of the target account under the risk statistical dimension is i This can be determined by:

[0163]

[0164] Specifically, for a risk statistical dimension, when determining the risk score of the target account corresponding to the risk statistical dimension based on the standardized data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension, for each risk indicator, the risk score of the target account corresponding to the risk indicator can be determined based on the standardized data of the target account under the risk statistical dimension and the risk weight of the risk indicator.

[0165] As an example, for risk indicator i, the risk score of the target account corresponding to the risk indicator is G i w i .

[0166] Furthermore, the risk score of the target account corresponding to the risk statistical dimension may be determined based on the risk score of each risk indicator under the risk statistical dimension corresponding to the target account.

[0167] As an example, for risk statistical dimension a, the risk score of the target account corresponding to the risk statistical dimension is:

[0168]

[0169] Among them, n is the number of risk indicators under risk statistical dimension a, G i w i is the risk score of the target account corresponding to risk indicator i under risk statistical dimension a.

[0170] As an example, see Figure 6 , Figure 6 Schematic diagram of a scenario for determining a risk score provided in an embodiment of the present application. Figure 6 As shown, if a risk statistical dimension includes n risk indicators from risk indicator 1 to risk indicator n, after the statistical data of the target account in each risk indicator is normalized and the risk weight of each risk indicator is determined, the standardized data corresponding to each risk indicator can be multiplied by the risk weight to obtain the risk score of the target account corresponding to the corresponding risk indicator. For example, the standardized data n of risk indicator n is multiplied by the risk weight of risk indicator n to obtain the risk score of the target account corresponding to risk indicator n.

[0171] Furthermore, the risk scores of the target account corresponding to each risk indicator may be added together to finally obtain the risk score of the target account corresponding to the risk statistical dimension.

[0172] Step S23: Determine the comprehensive risk score of the target account according to the relative importance of each risk statistical dimension and the risk score of the target account corresponding to each risk indicator.

[0173] In some feasible implementations, for any two risk statistical dimensions (such as the first risk statistical dimension and the second risk statistical dimension), the relative importance between the first risk statistical dimension and the second risk statistical dimension may include the importance of the first risk statistical dimension compared to the second risk statistical dimension, and the importance of the second risk statistical dimension compared to the first risk statistical dimension.

[0174] The relative importance between the risk statistical dimensions can be represented by a third judgment matrix, and each matrix element in the third judgment matrix is ​​used to represent the relative importance of two risk statistical dimensions.

[0175] As an example, the third judgment matrix can be expressed as:

[0176]

[0177] The dimension of the first judgment matrix corresponding to the risk statistical dimension is m×m, where m is the number of risk statistical dimensions. In addition, a matrix element in the third judgment matrix represents the importance of one risk statistical dimension compared to another risk statistical dimension.

[0178] The construction method of the third judgment matrix may refer to the construction method of the first judgment matrix, which will not be described in detail here.

[0179] In some feasible implementations, after the third judgment matrix is ​​determined, the validity of the third judgment matrix may be determined first, that is, whether the third judgment matrix passes the consistency verification.

[0180] Specifically, the maximum eigenvalue of the third judgment matrix may be determined first.

[0181] The maximum characteristic root can be determined by any method in the prior art.

[0182] Further, a consistency index of the third judgment matrix is ​​determined according to the third maximum characteristic root and the second number of risk statistical dimensions.

[0183] Among them, the specific implementation method of determining whether the third judgment matrix passes the consistency verification according to the consistency index of the third judgment matrix can refer to the related description of determining whether the first judgment matrix passes the consistency verification mentioned above, which will not be repeated here.

[0184] In some feasible implementations, after the third judgment matrix passes the validity verification, the risk weight of each risk statistical dimension may be determined according to the third judgment matrix.

[0185] Specifically, for each column in the third judgment matrix, the matrix elements of the third judgment matrix can be normalized by column to obtain a fourth intermediate matrix, and each matrix element in the fourth intermediate matrix is ​​divided by the sum of all matrix elements in the column to obtain a fifth intermediate matrix. Further, each matrix element in the fifth intermediate matrix is ​​summed by row to obtain a sixth intermediate matrix, and finally each matrix element in the sixth intermediate matrix is ​​divided by the number under the risk statistics dimension to obtain a fourth weight vector.

[0186] At this time, each element in the fourth weight vector is the risk weight of a risk statistical dimension.

[0187] Optionally, for the third judgment matrix, the matrix elements of the third judgment matrix can be multiplied row by row to obtain a fourth column vector, each element of the fourth column vector can be raised to the power of m (the number of risk statistical dimensions) to obtain a fifth column vector, and then each element of the fifth column vector can be normalized to obtain a fifth weight vector.

[0188] At this time, each element in the fifth weight vector is the risk weight of a risk statistical dimension.

[0189] Optionally, for the third judgment matrix, the maximum eigenvalue root of the third judgment matrix and its corresponding eigenvalue vector can be determined, the eigenvalue vector corresponding to the maximum eigenvalue root can be normalized to obtain a fifth column vector, and then each element of the fifth column vector can be normalized to obtain a sixth weight vector.

[0190] At this time, each element in the sixth weight vector is the risk weight of a risk statistical dimension.

[0191] Based on this, the comprehensive risk score of the target account can be determined according to the risk score of the target account corresponding to each risk statistical dimension and the risk weight of each risk statistical dimension.

[0192] As an example, for risk statistical dimension a, the risk score of the target account corresponding to risk statistical dimension a is T a , then the comprehensive risk score of the target account is T total for:

[0193]

[0194] Among them, m is the number of risk statistical dimensions, γ a is the risk score of the target account corresponding to risk statistical dimension a.

[0195] As an example, see Figure 7 , Figure 7 Schematic diagram of a scenario for determining a comprehensive risk score provided in an embodiment of the present application. Figure 7 As shown, if there are m risk statistical dimensions from risk statistical dimension 1 to risk statistical dimension m, after determining the risk score of the target account corresponding to each risk indicator, the risk score corresponding to each risk statistical dimension can be multiplied by the corresponding risk weight to obtain the weighted risk score of the target account corresponding to the corresponding risk statistical dimension.

[0196] For example, the risk score corresponding to risk indicator m is multiplied by the risk weight of risk statistical dimension m to obtain the weighted risk score of the target account corresponding to risk statistical dimension m.

[0197] Furthermore, the weighted risk scores of the target account corresponding to each risk statistical dimension may be added together to ultimately obtain a comprehensive risk score of the target account.

[0198] Step S24: Determine the risk level of the target account based on the comprehensive risk score.

[0199] In some feasible implementations, after determining the comprehensive risk score of the target account, the comprehensive risk score can be compared with multiple score intervals to determine the target score interval in which the comprehensive risk score lies, and then the preset risk level corresponding to the target score interval is determined as the risk level of the target account.

[0200] Among them, each score interval distribution corresponds to a different preset risk level, as shown in the following table. The higher the risk level of the target account, the higher the risk of the target account.

[0201] Risk Level risk Risk Description 0 No risk No obvious malicious behavior or malicious history 1 Minor risk Mild risk profile abnormalities 2 Low risk Abnormal risk profile 3 Medium risk Has obvious malicious characteristics 4 High risk Has obvious malicious characteristics

[0202] The data processing and calculation processes involved in the embodiments of the present application can be implemented based on cloud computing in cloud technology. Among them, cloud technology refers to a hosting technology that unifies a series of resources such as hardware, software, and network in a wide area network or a local area network to realize the calculation, storage, processing, and sharing of data. Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, application technology, etc. based on the cloud computing business model application, which can form a resource pool and be used on demand, flexibly and conveniently.

[0203] Cloud computing is a computing model that is the product of the development and integration of traditional computer and network technologies such as grid computing, distributed computing, parallel computing, utility computing, network storage technologies, virtualization, and load balancing. Cloud computing distributes computing tasks on a resource pool consisting of a large number of computers, allowing various application systems to obtain computing power, storage space, and information services as needed. The network that provides resources is called a "cloud". The resources in the "cloud" are infinitely scalable and can be obtained at any time, used on demand, expanded at any time, and paid for by use.

[0204] In the embodiment of the present application, the statistical data of the target account and other accounts in the account set can be stored in a designated storage space, which includes but is not limited to cloud storage, database (such as MYSQL database), blockchain, and the storage space of the device that executes the data processing method provided in the embodiment of the present application, etc. The specific storage space can be determined based on the actual application scenario requirements and is not limited here.

[0205] Among them, the database can be regarded as an electronic file cabinet in short - a place to store electronic files, which can be a relational database (SQL database) or a non-relational database (NoSQL database), without limitation. In this application, it can be used to store the statistical data of the target account and other accounts in the account set. Blockchain is a new application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, which is a string of data blocks generated by cryptographic methods. In the embodiment of the present application, each data block in the blockchain can store the statistical data of the target account and other accounts in the account set. Cloud storage is a new concept extended and developed from the concept of cloud computing. It refers to the use of cluster applications, grid technology, and distributed storage file systems. The functions of a large number of different types of storage devices (storage devices are also called storage nodes) in the network are combined through application software or application interfaces to work together to store the statistical data of the target account and other accounts in the account set.

[0206] In the embodiment of the present application, for each risk statistical dimension, the risk score of the target account corresponding to the risk statistical dimension can be determined based on the statistical data of the target account under each risk indicator, taking into account the relative importance of the risk indicators under the risk statistical dimension. Further, the comprehensive risk score of the target account can be obtained based on the risk score of the target account corresponding to each risk statistical dimension, taking into account the relative importance of each risk statistical dimension. Based on this, the statistical data of the target account in various risk indicators can be combined, and the comprehensive risk score can be calculated by comprehensively combining different risk statistical dimensions, so that the risk level of the target account can be accurately and effectively determined, and the applicability is high.

[0207] Compared with the traditional expert evaluation method, the data method provided by the embodiment of the present application can effectively avoid the instability caused by the subjective judgment of experts. At the same time, compared with the traditional statistical analysis method, it can effectively cover various risk statistical dimensions and risk indicators, effectively solve the interaction between different risk statistical dimensions and different risk indicators, and can also adapt to different business platform requirements, so as to obtain effective and accurate risk levels.

[0208] See also Figure 8 , Figure 8 : is a schematic diagram of the structure of a data processing device provided in an embodiment of the present application. The data processing device provided in an embodiment of the present application includes:

[0209] A data acquisition module 81 is used to determine the statistical data of each risk indicator of the target account under at least one risk statistical dimension;

[0210] The risk calculation module 82 is used to determine, for each of the risk statistical dimensions, a risk score of the target account corresponding to the risk statistical dimension according to the relative importance of each risk indicator under the risk statistical dimension and the statistical data of each risk indicator of the target account under the risk statistical dimension;

[0211] The risk calculation module 82 is used to determine the comprehensive risk score of the target account according to the relative importance of each of the risk statistical dimensions and the risk score of the target account corresponding to each of the risk statistical dimensions;

[0212] The level determination module 83 is used to determine the risk level of the target account according to the comprehensive risk score.

[0213] In some feasible implementations, for each of the risk statistical dimensions, the risk calculation module 82 is used to:

[0214] According to each risk indicator under the risk statistical dimension, a first judgment matrix is ​​determined, wherein each matrix element in the first judgment matrix is ​​used to represent the relative importance of two risk indicators under the risk statistical dimension;

[0215] In response to the first judgment matrix passing the validity verification, determining the risk weight of each risk indicator under the risk statistical dimension according to the first judgment matrix;

[0216] The risk score of the target account corresponding to the risk statistical dimension is determined based on the statistical data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension.

[0217] In some feasible implementations, the risk calculation module 82 is used to:

[0218] Determine the maximum eigenvalue of the first judgment matrix;

[0219] Determining a consistency index of the first judgment matrix according to the maximum eigenvalue of the first judgment matrix and the first number of risk indicators under the risk statistical dimension;

[0220] Determine a reference consistency index of the first judgment matrix according to the first quantity;

[0221] A first ratio of a consistency index of the first judgment matrix to a reference consistency index is determined, and in response to the first ratio being less than a first threshold, it is determined that the first judgment matrix passes the validity verification.

[0222] In some feasible implementations, the risk calculation module 82 is used to:

[0223] Determine a reference weight of each risk indicator under the risk statistical dimension according to the statistical data of each risk indicator under the risk statistical dimension for each account in the account set;

[0224] According to the reference weight of each risk indicator under the risk statistical dimension, the matrix elements of the first judgment matrix are adjusted to obtain a second judgment matrix;

[0225] According to the second judgment matrix, the risk weight of each risk indicator under the risk statistical dimension is determined.

[0226] In some feasible implementations, for each risk statistical dimension, the risk calculation module 82 is used to:

[0227] For each risk indicator under the risk statistical dimension, determine a second ratio of the statistical data of each account in the account set to the target statistical data for the risk indicator, where the target statistical data is the sum of the statistical data of all accounts in the account set for the risk indicator;

[0228] Determining the information entropy of the risk indicator according to the second ratio corresponding to each account in the account set;

[0229] According to the information entropy of the risk indicator, determining the information entropy redundancy of the risk indicator;

[0230] According to the information entropy redundancy of each risk indicator under the risk statistical dimension, the reference weight of each risk indicator under the risk statistical dimension is determined.

[0231] In some feasible implementations, the risk calculation module 82 is used to:

[0232] Determine at least one target matrix element in the first judgment matrix, wherein each target matrix element and the reference weights of the two risk indicators corresponding to the target matrix element meet a preset condition;

[0233] According to the reference weights of the two risk indicators corresponding to each of the above target matrix elements, the corresponding target matrix elements are adjusted to obtain the second judgment matrix.

[0234] In some feasible implementations, the risk calculation module 82 is used to:

[0235] Determine the target standardization method based on the statistical data of each risk indicator of the target account under the risk statistical dimension;

[0236] According to the above target standardization method, the statistical data of each risk indicator of the above target account under the risk statistical dimension is subjected to data standardization processing to obtain standardized data;

[0237] The risk score of the target account corresponding to the risk statistical dimension is determined based on the standardized data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension.

[0238] In some feasible implementations, the risk calculation module 82 is used to:

[0239] Determine the data skewness corresponding to the target account according to the statistical data of each risk indicator of the target account under the risk statistical dimension, where the data skewness is used to describe the degree of skewness of data distribution of the target account;

[0240] Based on the above data skewness, determine the target normalization method.

[0241] In some feasible implementations, the risk calculation module 82 is used to:

[0242] For each risk indicator under the risk statistical dimension, determine the risk score of the target account corresponding to the risk indicator according to the standardized data of the target account in the risk indicator and the risk weight of the risk indicator;

[0243] According to the risk score of the target account corresponding to each risk indicator under the risk statistical dimension, the risk score of the target account corresponding to the risk statistical dimension is determined.

[0244] In some feasible implementations, the risk calculation module 82 is used to:

[0245] According to each of the above risk statistical dimensions, a third judgment matrix is ​​determined, wherein each matrix element in the above third judgment matrix is ​​used to represent the relative importance of every two of the above risk statistical dimensions;

[0246] In response to the third judgment matrix passing the validity verification, determining the risk weight of each of the risk statistical dimensions according to the third judgment matrix;

[0247] The comprehensive risk score of the target account is determined according to the risk weight of each of the risk statistical dimensions and the risk score of the target account corresponding to each of the risk statistical dimensions.

[0248] In a specific implementation, the above data processing device can execute the above-mentioned Figure 2 For the implementation methods provided in each step, please refer to the implementation methods provided in the above steps for details, which will not be repeated here.

[0249] See also Fig. 9 , Fig. 9Schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Fig. 9 As shown, the electronic device 900 in this embodiment may include: a processor 901, a network interface 904 and a memory 905. In addition, the above-mentioned electronic device 900 may also include: an object interface 903, and at least one communication bus 902. Among them, the communication bus 902 is used to realize the connection and communication between these components. Among them, the object interface 903 may include a display screen (Display), a keyboard (Keyboard), and the object interface 903 may optionally include a standard wired interface and a wireless interface. The network interface 904 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 905 may be a high-speed RAM memory, or it may be a non-volatile memory (NVM), such as at least one disk storage. The memory 905 may optionally also be at least one storage device located away from the aforementioned processor 901. As Fig. 9 As shown, the memory 905 as a computer-readable storage medium may include an operating system, a network communication module, an object interface module, and a device control application program.

[0250] exist Fig. 9 In the electronic device 900 shown, the network interface 904 can provide a network communication function; the object interface 903 is mainly used to provide an input interface for the object; and the processor 901 can be used to call the device control application stored in the memory 905 to achieve:

[0251] Determine the statistical data of each risk indicator of the target account under at least one risk statistical dimension;

[0252] For each of the risk statistical dimensions, determine the risk score of the target account corresponding to the risk statistical dimension according to the relative importance of each risk indicator under the risk statistical dimension and the statistical data of each risk indicator of the target account under the risk statistical dimension;

[0253] Determine a comprehensive risk score of the target account according to the relative importance of each of the risk statistical dimensions and the risk score of the target account corresponding to each of the risk statistical dimensions;

[0254] Based on the above comprehensive risk score, determine the risk level of the above target account.

[0255] In some feasible implementations, for each of the risk statistical dimensions, the processor 901 is configured to:

[0256] According to each risk indicator under the risk statistical dimension, a first judgment matrix is ​​determined, wherein each matrix element in the first judgment matrix is ​​used to represent the relative importance of two risk indicators under the risk statistical dimension;

[0257] In response to the first judgment matrix passing the validity verification, determining the risk weight of each risk indicator under the risk statistical dimension according to the first judgment matrix;

[0258] The risk score of the target account corresponding to the risk statistical dimension is determined based on the statistical data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension.

[0259] In some feasible implementations, the processor 901 is configured to:

[0260] Determine the maximum eigenvalue of the first judgment matrix;

[0261] Determining a consistency index of the first judgment matrix according to the maximum eigenvalue of the first judgment matrix and the first number of risk indicators under the risk statistical dimension;

[0262] Determine a reference consistency index of the first judgment matrix according to the first quantity;

[0263] A first ratio of a consistency index of the first judgment matrix to a reference consistency index is determined, and in response to the first ratio being less than a first threshold, it is determined that the first judgment matrix passes the validity verification.

[0264] In some feasible implementations, the processor 901 is configured to:

[0265] Determine a reference weight of each risk indicator under the risk statistical dimension according to the statistical data of each risk indicator under the risk statistical dimension for each account in the account set;

[0266] According to the reference weight of each risk indicator under the risk statistical dimension, the matrix elements of the first judgment matrix are adjusted to obtain a second judgment matrix;

[0267] According to the second judgment matrix, the risk weight of each risk indicator under the risk statistical dimension is determined.

[0268] In some feasible implementations, the processor 901 is configured to:

[0269] For each risk indicator under the risk statistical dimension, determine a second ratio of the statistical data of each account in the account set to the target statistical data for the risk indicator, where the target statistical data is the sum of the statistical data of all accounts in the account set for the risk indicator;

[0270] Determining the information entropy of the risk indicator according to the second ratio corresponding to each account in the account set;

[0271] According to the information entropy of the risk indicator, determining the information entropy redundancy of the risk indicator;

[0272] According to the information entropy redundancy of each risk indicator under the risk statistical dimension, the reference weight of each risk indicator under the risk statistical dimension is determined.

[0273] In some feasible implementations, the processor 901 is configured to:

[0274] Determine at least one target matrix element in the first judgment matrix, wherein each target matrix element and the reference weights of the two risk indicators corresponding to the target matrix element meet a preset condition;

[0275] According to the reference weights of the two risk indicators corresponding to each of the above target matrix elements, the corresponding target matrix elements are adjusted to obtain the second judgment matrix.

[0276] In some feasible implementations, the processor 901 is configured to:

[0277] Determine the target standardization method based on the statistical data of each risk indicator of the target account under the risk statistical dimension;

[0278] According to the above target standardization method, the statistical data of each risk indicator of the above target account under the risk statistical dimension is subjected to data standardization processing to obtain standardized data;

[0279] The risk score of the target account corresponding to the risk statistical dimension is determined based on the standardized data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension.

[0280] In some feasible implementations, the processor 901 is configured to:

[0281] Determine the data skewness corresponding to the target account according to the statistical data of each risk indicator of the target account under the risk statistical dimension, where the data skewness is used to describe the degree of skewness of data distribution of the target account;

[0282] Based on the above data skewness, determine the target normalization method.

[0283] In some feasible implementations, the processor 901 is configured to:

[0284] For each risk indicator under the risk statistical dimension, determine the risk score of the target account corresponding to the risk indicator according to the standardized data of the target account in the risk indicator and the risk weight of the risk indicator;

[0285] According to the risk score of the target account corresponding to each risk indicator under the risk statistical dimension, the risk score of the target account corresponding to the risk statistical dimension is determined.

[0286] In some feasible implementations, the processor 901 is configured to:

[0287] According to each of the above risk statistical dimensions, a third judgment matrix is ​​determined, wherein each matrix element in the above third judgment matrix is ​​used to represent the relative importance of every two of the above risk statistical dimensions;

[0288] In response to the third judgment matrix passing the validity verification, determining the risk weight of each of the risk statistical dimensions according to the third judgment matrix;

[0289] The comprehensive risk score of the target account is determined according to the risk weight of each of the risk statistical dimensions and the risk score of the target account corresponding to each of the risk statistical dimensions.

[0290] It should be understood that in some feasible implementations, the processor 901 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the type of device.

[0291] In a specific implementation, the electronic device 900 can execute the above-mentioned functions through its built-in functional modules. Figure 2 For the implementation methods provided in each step, please refer to the implementation methods provided in the above steps for details, which will not be repeated here.

[0292] The present application also provides a computer-readable storage medium, which stores a computer program and is executed by a processor to implement Figure 2 For the methods provided in each step, please refer to the implementation methods provided in the above steps for details, which will not be repeated here.

[0293] The computer-readable storage medium may be an internal storage unit of a data processing device or an electronic device provided in any of the aforementioned embodiments, such as a hard disk or memory of an electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (smart media card, SMC), a secure digital (secure digital, SD) card, a flash card (flash card), etc. equipped on the electronic device. The computer-readable storage medium may also include a magnetic disk, an optical disk, a read-only memory (read-only memory, ROM) or a random access memory (random access memory, RAM), etc. Further, the computer-readable storage medium may also include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by the electronic device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0294] The present invention provides a computer program product, which includes a computer program. The computer program is executed by a processor. Figure 2 The methods provided in each step.

[0295] The terms "first", "second", etc. in the claims, specification and drawings of the present application are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or electronic device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or electronic devices. Mentioning "embodiment" in this article means that the specific features, structures or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present application. Displaying the phrase at various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments. The term "and / or" used in the present application specification and the attached claims refers to any combination of one or more of the items listed in association and all possible combinations, and includes these combinations.

[0296] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Professional and technical personnel may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0297] The above disclosure is only the preferred embodiment of the present application, which cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A data processing method, characterized in that: The method comprises: Determine the statistical data of each risk indicator of the target account under at least one risk statistical dimension; For each of the risk statistical dimensions, determine the risk score of the target account corresponding to the risk statistical dimension according to the relative importance of each risk indicator under the risk statistical dimension and the statistical data of each risk indicator of the target account under the risk statistical dimension; Determining a comprehensive risk score of the target account according to the relative importance of each of the risk statistical dimensions and the risk score of the target account corresponding to each of the risk statistical dimensions; The risk level of the target account is determined based on the comprehensive risk score.

2. The method according to claim 1, characterized in that For each of the risk statistical dimensions, determining the risk score of the target account corresponding to the risk statistical dimension according to the relative importance of each risk indicator under the risk statistical dimension and the statistical data of each risk indicator of the target account under the risk statistical dimension includes: Determine a first judgment matrix according to each risk indicator under the risk statistical dimension, wherein each matrix element in the first judgment matrix is ​​used to represent the relative importance of two risk indicators under the risk statistical dimension; In response to the first judgment matrix passing the validity verification, determining the risk weight of each risk indicator under the risk statistical dimension according to the first judgment matrix; The risk score of the target account corresponding to the risk statistical dimension is determined based on the statistical data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension.

3. The method according to claim 2, characterized in that Determining that the first judgment matrix passes the validity verification includes: Determine the maximum eigenvalue of the first judgment matrix; Determining a consistency index of the first judgment matrix according to the maximum eigenvalue of the first judgment matrix and a first number of risk indicators under the risk statistical dimension; Determining a reference consistency index of the first judgment matrix according to the first quantity; A first ratio of a consistency index of the first judgment matrix to a reference consistency index is determined, and in response to the first ratio being less than a first threshold, it is determined that the first judgment matrix passes validity verification.

4. The method according to claim 2, characterized in that: Determining the risk weight of each risk indicator under the risk statistical dimension according to the first judgment matrix includes: Determine a reference weight of each risk indicator under the risk statistical dimension according to the statistical data of each risk indicator under the risk statistical dimension for each account in the account set; According to the reference weight of each risk indicator under the risk statistical dimension, the matrix elements of the first judgment matrix are adjusted to obtain a second judgment matrix; According to the second judgment matrix, the risk weight of each risk indicator under the risk statistical dimension is determined.

5. The method according to claim 4, characterized in that For the risk statistical dimension, determining the reference weight of each risk indicator under the risk statistical dimension according to the statistical data of each risk indicator under the risk statistical dimension of each account in the account set includes: For each risk indicator under the risk statistical dimension, determine a second ratio of the statistical data of each account in the account set on the risk indicator to the target statistical data, where the target statistical data is the sum of the statistical data of all accounts in the account set on the risk indicator; Determining the information entropy of the risk indicator according to the second ratio corresponding to each account in the account set; According to the information entropy of the risk indicator, determining the information entropy redundancy of the risk indicator; According to the information entropy redundancy of each risk indicator under the risk statistical dimension, the reference weight of each risk indicator under the risk statistical dimension is determined.

6. The method according to claim 4, characterized in that The step of adjusting the matrix elements of the first judgment matrix according to the reference weight of each risk indicator under the risk statistical dimension to obtain the second judgment matrix includes: Determine at least one target matrix element in the first judgment matrix, wherein each of the target matrix elements and the reference weights of the two risk indicators corresponding to the target matrix element meet a preset condition; According to the reference weights of the two risk indicators corresponding to each of the target matrix elements, the corresponding target matrix elements are adjusted to obtain a second judgment matrix.

7. The method according to claim 2, characterized in that: The step of determining the risk score of the target account corresponding to the risk statistical dimension according to the statistical data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension comprises: Determine a target normalization method according to the statistical data of each risk indicator of the target account under the risk statistical dimension; According to the target standardization method, data standardization processing is performed on the statistical data of each risk indicator of the target account under the risk statistical dimension to obtain standardized data; The risk score of the target account corresponding to the risk statistical dimension is determined based on the standardized data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension.

8. The method according to claim 7, characterized in that The determining of the target normalization method according to the statistical data of each risk indicator of the target account under the risk statistical dimension includes: Determine the data skewness corresponding to the target account according to the statistical data of each risk indicator of the target account under the risk statistical dimension, where the data skewness is used to describe the degree of skewness of data distribution of the target account; According to the data skewness, the target normalization method is determined.

9. The method according to claim 7, characterized in that: The step of determining the risk score of the target account corresponding to the risk statistical dimension according to the standardized data of each risk indicator of the target account under the risk statistical dimension and the risk weight of each risk indicator under the risk statistical dimension comprises: For each risk indicator under the risk statistical dimension, determine the risk score of the target account corresponding to the risk indicator according to the standardized data of the target account in the risk indicator and the risk weight of the risk indicator; According to the risk score of the target account corresponding to each risk indicator under the risk statistical dimension, the risk score of the target account corresponding to the risk statistical dimension is determined.

10. The method according to claim 1, characterized in that Determining the comprehensive risk score of the target account according to the relative importance of each of the risk statistical dimensions and the risk score of the target account corresponding to each of the risk statistical dimensions includes: Determine a third judgment matrix according to each of the risk statistical dimensions, wherein each matrix element in the third judgment matrix is ​​used to characterize the relative importance of every two of the risk statistical dimensions; In response to the third judgment matrix passing the validity verification, determining the risk weight of each of the risk statistical dimensions according to the third judgment matrix; The comprehensive risk score of the target account is determined according to the risk weight of each risk statistical dimension and the risk score of the target account corresponding to each risk statistical dimension.

11. A data processing device, characterized in that: The device comprises: A data acquisition module, used to determine the statistical data of each risk indicator of the target account under at least one risk statistical dimension; A risk calculation module, for determining, for each risk statistical dimension, a risk score of the target account corresponding to the risk statistical dimension according to the relative importance of each risk indicator under the risk statistical dimension and the statistical data of each risk indicator of the target account under the risk statistical dimension; The risk calculation module is used to determine the comprehensive risk score of the target account according to the relative importance of each of the risk statistical dimensions and the risk score of the target account corresponding to each of the risk statistical dimensions; The level determination module is used to determine the risk level of the target account according to the comprehensive risk score.

12. An electronic device, characterized in that: comprising a processor and a memory, wherein the processor and the memory are connected to each other; The memory is used to store computer programs; The processor is configured to execute the method according to any one of claims 1 to 10 when calling the computer program.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method according to any one of claims 1 to 10.

Images