Risk assessment method

Abstract

The invention provides a risk assessment method, and the method comprises the steps: receiving a first evaluation index of each target attribute variable in a variable decision tree which is transmitted by a server and is associated with a target business of a financial institution, and displaying the variable decision tree and each first evaluation index; obtaining first attribute data associated with a target object of the to-be-executed target service; performing risk assessment on the target object according to the first attribute data and the risk control threshold value of each target attribute variable; therefore, through interaction between the client and the server, the risk control threshold is dynamically adjusted and optimized, the accuracy of risk assessment can be remarkably improved, the variable decision tree and the first evaluation index are displayed, so that business personnel can better understand the influence of each variable on risk assessment, business rules are conveniently formulated and optimized, and the risk assessment efficiency is improved. And the risk management capability of the financial institution is further improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of risk management, and in particular to a risk assessment method. Background Art

[0002] With the rapid development of the financial industry, the importance of financial risk management has become increasingly prominent. Implementing accurate risk assessment for target objects of financial services such as loans can not only help financial institutions effectively identify and manage potential credit risks, but also optimize resource allocation, improve capital utilization efficiency, and enhance market competitiveness. Therefore, it is very important to accurately conduct risk assessment for target objects of financial services. Summary of the invention

[0003] The present disclosure provides a risk assessment method to solve one of the technical problems in the related art at least to a certain extent. The technical solution of the present disclosure is as follows:

[0004] According to a first aspect of an embodiment of the present disclosure, a risk assessment method is provided, which is applied to a client, and includes: receiving a first evaluation index of each target attribute variable in a variable decision tree associated with a target business of a financial institution sent by a server, and displaying the variable decision tree and each first evaluation index; wherein the first evaluation index is determined according to risk control tags of multiple sample objects associated with the first attribute value of the corresponding target attribute variable, and the first attribute value is generated by the client in response to a first adjustment operation triggered by a threshold adjuster of the target attribute variable and sent to the server; the risk control tag is used to indicate the risk level of the sample object in executing the target business; in response to the difference between the first evaluation index and the second evaluation index of each target attribute variable being less than a set index threshold, determining the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable; wherein the second evaluation index is generated according to the client in response to a second adjustment operation triggered by the threshold adjuster and sent to the server; the second adjustment operation is triggered before the first adjustment operation; acquiring first attribute data associated with a target object to be executed with the target business; and performing risk assessment on the target object according to the first attribute data and the risk control threshold of each target attribute variable.

[0005] According to a second aspect of an embodiment of the present disclosure, another risk estimation method is provided, which is applied to a server, and includes: receiving a first attribute value of each target attribute variable in a variable decision tree associated with a target business of a financial institution sent by a client; wherein the first attribute value is generated and sent by the client in response to a first adjustment operation triggered by a threshold adjuster of the target attribute variable; determining a first evaluation indicator of each target attribute variable according to risk control labels of multiple sample objects associated with the first attribute value of each target attribute variable; sending the first evaluation indicator of each target attribute variable to the client, so as to display the variable decision tree and each first evaluation indicator through the client; wherein, The first evaluation indicator is used to determine the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable in response to the difference between the first evaluation indicator and the second evaluation indicator of each target attribute variable being less than the set indicator threshold; obtain the first attribute data associated with the target object to be executed for the target business; and perform risk assessment on the target object according to the first attribute data and the risk control threshold of each target attribute variable; wherein the second evaluation indicator is generated according to the second attribute value of the target attribute variable generated by the client in response to the second adjustment operation triggered by the threshold adjuster and sent to the server; the second adjustment operation is triggered before the first adjustment operation.

[0006] According to a third aspect of an embodiment of the present disclosure, there is provided a risk assessment device, which is applied to a client, and includes: a receiving module, which is used to receive a first evaluation indicator of each target attribute variable in a variable decision tree associated with a target business of a financial institution sent by a server, and to display the variable decision tree and each of the first evaluation indicators; wherein the first evaluation indicator is determined based on risk control labels of multiple sample objects associated with a first attribute value of the corresponding target attribute variable, and the first attribute value is generated by the client in response to a first adjustment operation triggered by a threshold adjuster of the target attribute variable and sent to the server; the risk control label is used to indicate the risk level of the sample object in executing the target business; A determination module, for determining the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable in response to the difference between the first evaluation indicator and the second evaluation indicator of each target attribute variable being less than the set indicator threshold; wherein the second evaluation indicator is generated and sent to the server according to the second adjustment operation triggered by the threshold adjuster in response to the client; the second adjustment operation is triggered before the first adjustment operation; a first acquisition module, for acquiring first attribute data associated with the target object to be executed the target business; an evaluation module, for performing risk assessment on the target object according to the first attribute data and the risk control threshold of each target attribute variable.

[0007] According to a fourth aspect of an embodiment of the present disclosure, a risk assessment device is provided, which is applied to a server, and includes: a receiving module, which is used to receive a first attribute value of each target attribute variable in a variable decision tree associated with a target business of a financial institution sent by a client; wherein the first attribute value is generated and sent by the client in response to a first adjustment operation triggered by a threshold adjuster of the target attribute variable; a determining module, which is used to determine a first evaluation index of each target attribute variable according to risk control labels of multiple sample objects associated with the first attribute value of each target attribute variable; and a sending module, which is used to send the first evaluation index of each target attribute variable to the client, so that the variable decision tree and each first evaluation index can be evaluated by the client. Indicators are displayed; wherein, the first evaluation indicator is used to determine the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable in response to the difference between the first evaluation indicator and the second evaluation indicator of each target attribute variable being less than the set indicator threshold; obtain the first attribute data associated with the target object to be executed for the target business; perform risk assessment on the target object according to the first attribute data and the risk control threshold of each target attribute variable; wherein, the second evaluation indicator is generated according to the second attribute value of the target attribute variable generated by the client in response to the second adjustment operation triggered by the threshold adjuster and sent to the server; the second adjustment operation is triggered before the first adjustment operation.

[0008] According to a fifth aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement a risk assessment method as described in any of the above embodiments of the present disclosure.

[0009] According to a sixth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the risk assessment method as described in any of the above embodiments of the present disclosure.

[0010] According to a seventh aspect of an embodiment of the present disclosure, a computer program product is provided, comprising: a computer program, wherein when the computer program is executed by a processor, the risk assessment method as described in any of the above embodiments of the present disclosure is implemented.

[0011] The technical solution provided by the embodiments of the present disclosure brings at least the following beneficial effects:

[0012] In the present technical scheme, the first evaluation index of each target attribute variable in the variable decision tree associated with the target business of the financial institution sent by the server is received, and the variable decision tree and each first evaluation index are displayed; in response to the difference between the first evaluation index and the second evaluation index of each target attribute variable being less than the set index threshold, the risk control threshold of each target attribute variable is determined according to the first attribute value of each target attribute variable; the first attribute data associated with the target object of the target business to be executed is obtained; and the target object is risk assessed according to the first attribute data and the risk control threshold of each target attribute variable. Thus, through the interaction between the client and the server, the risk control threshold is dynamically adjusted and optimized, which can significantly improve the accuracy of risk assessment, and the variable decision tree and the first evaluation index are displayed so that business personnel can better understand the impact of each variable on risk assessment, facilitate the formulation and optimization of business rules, and further enhance financial The risk management capability of the institution; wherein, according to the target attribute variables among multiple candidate attribute variables, a variable decision tree associated with the target business is constructed, so that the most influential target attribute variables can be screened out from the candidate attribute variables, and the rationality, accuracy and reliability of risk assessment are improved. The variable decision tree is constructed through the target attribute variables, so that relevant business personnel can intuitively understand the impact of each variable on risk assessment and better perform risk management. In addition, for any target attribute variable, the first evaluation indicator of each target attribute variable is determined according to the first evaluation indicator of each third bin of any target attribute variable, so that the first evaluation indicator of each bin can be quickly calculated through automated binning and statistical methods, saving the time and cost of manual calculation, and realizing refined assessment of risks in different intervals, providing scientific data support for business decisions, reducing errors in subjective judgment, and improving the scientificity and reliability of decisions.

[0013] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute improper limitations on the present disclosure.

[0015] Figure 1 is a flowchart of a risk assessment method shown in the first embodiment of the present disclosure;

[0016] Figure 2 is a flow chart of a risk assessment method according to a second embodiment of the present disclosure;

[0017] Figure 3 is a flow chart of a risk assessment method shown in the third embodiment of the present disclosure;

[0018] Figure 4 is a flowchart of a risk assessment method shown in the fourth embodiment of the present disclosure;

[0019] Figure 5 is a schematic diagram of the structure of a risk assessment device shown in the fifth embodiment of the present disclosure;

[0020] Figure 6 is a schematic diagram of the risk assessment principle shown in an embodiment of the present disclosure;

[0021] Figure 7 is a schematic diagram of variable threshold adjustment shown in an embodiment of the present disclosure;

[0022] Figure 8 is a schematic diagram of the structure of a risk assessment device shown in the sixth embodiment of the present disclosure;

[0023] Fig. 9 is a schematic diagram of the structure of a risk assessment device shown in the seventh embodiment of the present disclosure;

[0024] Fig.10 It is a schematic structural diagram of an electronic device shown in an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0027] It should be noted that in the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information are all carried out with the user's consent, and are in compliance with the relevant laws and regulations and do not violate public order and good morals.

[0028] At present, the risk strategy tuning technology solution, as an important part of enterprise risk management, uses a series of systematic and standardized processes to ensure that enterprises can effectively deal with various risks in a complex and changing market environment; risk strategy tuning emphasizes comprehensive risk identification; enterprises use a variety of methods, such as questionnaires, expert consultations, data analysis, etc., to comprehensively sort out and identify internal and external risks. This process not only focuses on direct business risks, but also covers multi-dimensional risks such as market and reputation to ensure the comprehensiveness and accuracy of risk identification; risk assessment is the core link of the tuning technology solution.

[0029] In related technologies, a combination of qualitative and quantitative methods is used to conduct in-depth assessments of identified risks. For example, by building a risk assessment model, the probability of risk occurrence, the degree of impact, and potential losses are analyzed to provide a scientific basis for formulating risk response strategies. For another example, in terms of risk monitoring, real-time monitoring and early warning of risks can be achieved by setting monitoring indicators and establishing an early warning mechanism.

[0030] In addition, in related technologies, according to the risk assessment results, formulate targeted risk response strategies, including risk avoidance, mitigation, transfer and acceptance, etc. At the same time, maintain the flexibility and adaptability of the strategy, and adjust and optimize the risk strategy in a timely manner according to market changes and business development needs.

[0031] However, the traditional strategy framework is rigid and difficult to adapt to market changes quickly, which affects the timeliness of investment decisions. The diversified investment of funds to reduce risks may limit the in-depth exploration of specific areas and cannot meet the diversity and flexibility of investment portfolios. Complex management processes and data analysis greatly increase the optimization costs, such as time costs and operating costs. The traditional risk strategy optimization process relies on advanced technical means and requires high professional quality of relevant personnel. In actual operation, it also requires multi-departmental collaboration to overcome communication barriers and conflicts of interest, which makes implementation difficult. In addition, the optimization of traditional risk strategies is highly dependent on historical data and model assumptions, both of which have limitations. Historical data is difficult to fully reflect future market changes, and model assumptions may deviate from reality, resulting in inaccurate forecast results.

[0032] Therefore, in response to the above problems, the present disclosure proposes a risk assessment method.

[0033] The risk assessment method according to the embodiment of the present disclosure is described below with reference to the accompanying drawings.

[0034] Figure 1 1 is a flow chart of the risk assessment method shown in the first embodiment of the present disclosure. It should be noted that the risk assessment method of the embodiment of the present disclosure is applied to the client.

[0035] like Figure 1 As shown, the risk assessment method includes the following steps:

[0036] Step 101, receiving the first evaluation index of each target attribute variable in the variable decision tree associated with the target business of the financial institution sent by the server, and displaying the variable decision tree and each first evaluation index.

[0037] Among them, the first evaluation indicator is determined according to the risk control labels of multiple sample objects associated with the first attribute value of the corresponding target attribute variable, and the first attribute value is generated by the client in response to the first adjustment operation triggered by the threshold adjuster of the target attribute variable and sent to the server; the risk control label is used to indicate the risk level of the sample object in executing the target business.

[0038] In order to improve the accuracy of risk assessment and reduce the difficulty of operation, as a possible implementation method, through real-time interaction between the client and the server, the client displays the variable decision tree associated with the target business and the first evaluation index of each target attribute variable, so as to dynamically adjust and optimize the risk control threshold.

[0039] As an example, the client generates a first attribute value in response to a first adjustment operation triggered by a threshold adjuster of a target attribute variable, and sends the first attribute value to the server. The server determines a first evaluation indicator of each target attribute variable based on risk control labels of multiple sample objects associated with the first attribute value of each target attribute variable, and sends the first evaluation indicator of each target attribute variable to the client so that the variable decision tree and each first evaluation indicator can be displayed through the client.

[0040] Step 102, in response to the difference between the first evaluation index and the second evaluation index of each target attribute variable being less than the set index threshold, determine the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable.

[0041] The second evaluation index is generated and sent to the server according to the second adjustment operation triggered by the client response to the threshold adjuster; the second adjustment operation is triggered before the first adjustment operation.

[0042] In order to improve the explainability of business decisions, as a possible implementation method, financial institutions can select the optimal risk control threshold after adjusting the threshold multiple times, thereby ensuring the stability and accuracy of risk assessment.

[0043] As an example, through two different threshold adjustment operations (the first adjustment operation and the second adjustment operation), the difference between the first evaluation index and the second evaluation index obtained is small, and the first evaluation index reaches a relatively stable level. In this case, the final risk control threshold can be determined according to the first attribute value of each target attribute variable. Among them, the second evaluation index is generated according to the second attribute value of the target attribute variable generated by the client in response to the second adjustment operation triggered by the threshold adjuster and sent to the server; the second adjustment operation is triggered before the first adjustment operation.

[0044] For example, taking the target attribute variable "income level" as an example, the business personnel adjusts the threshold adjuster of "income level" for the first time and sets the threshold to 6589. The client sends this threshold (6589) to the server. The server calculates the second evaluation indicator based on this threshold and sends the first evaluation indicator to the client. The business personnel adjusts the threshold adjuster of "income level" for the second time and adjusts the threshold to 6500. The client sends the new threshold (6500) to the server. The server calculates the first evaluation indicator based on the new threshold and sends the first evaluation indicator to the client. The client compares the first evaluation indicator with the second evaluation indicator and calculates the difference between the first evaluation indicator and the second evaluation indicator. When the difference between the first evaluation indicator and the second evaluation indicator is less than the set indicator threshold, the first attribute value (6500) is used as the risk control threshold of "income level".

[0045] Step 103: Acquire first attribute data associated with the target object of the target business to be executed.

[0046] In order to conduct risk assessment accurately and comprehensively, as a possible implementation method, multiple dimensions of data associated with the target object that executes the target business of the financial institution (i.e., first attribute data) are obtained, for example, including the customer's personal information, financial status, historical transaction records, debt repayment ability, repayment record, number of defaults, etc., and then, risk assessment is conducted based on the first attribute data.

[0047] Step 104: Perform risk assessment on the target object based on the first attribute data and the risk control threshold of each target attribute variable.

[0048] In order to accurately conduct risk assessment, as a possible implementation method, the first attribute data of the target object (such as age, income, credit score, etc.) is comprehensively analyzed, combined with the risk control threshold, to evaluate the risk level of the target object, and finally divide it into different risk levels (such as low risk, medium risk, high risk).

[0049] As an example, for any target attribute variable, the attribute value of any target attribute variable is obtained from the first attribute data; in response to the attribute value of any target attribute variable being greater than or equal to the risk control threshold of any target attribute variable, it is determined that the risk level of the target object executing the target business is greater than the set risk level.

[0050] That is to say, by obtaining the attribute value of any target attribute variable from the first attribute data, and judging whether the attribute value of any target attribute variable is greater than or equal to the corresponding risk control threshold, when the attribute value of any target attribute variable is greater than or equal to the corresponding risk control threshold, it is determined that the risk level of the target object performing the target business is greater than the set risk level, that is, it is determined that the risk level of the target object performing the target business is high risk.

[0051] As another example, in response to the fact that the attribute values ​​of multiple target attribute variables in the first attribute data and the first risk control attribute data are all less than the risk control thresholds of the multiple target attribute variables, it is determined that the risk level of the target object performing the target business is not greater than the set risk level, that is, the risk level of the target object performing the target business is determined to be low risk.

[0052] That is to say, when the attribute values ​​of multiple target attribute variables in the first attribute data and the first risk control attribute data are all less than the corresponding risk control thresholds, it is determined that the risk level of the target object executing the target business is not greater than the set risk level.

[0053] In summary, through the interaction between the client and the server, the risk control threshold can be dynamically adjusted and optimized, which can significantly improve the accuracy of risk assessment. The variable decision tree and the first evaluation indicator are displayed so that business personnel can better understand the impact of each variable on risk assessment, facilitate the formulation and optimization of business rules, and further enhance the risk management capabilities of financial institutions.

[0054] In order to clearly illustrate how the variable decision tree in the above embodiment is generated, the present disclosure proposes another risk assessment method.

[0055] Figure 2 It is a flowchart of the risk assessment method shown in the second embodiment of the present disclosure.

[0056] like Figure 2 As shown, based on any of the above embodiments, the variable decision tree is generated by following steps 201 to 203:

[0057] Step 201: Obtain sample attribute data and risk control labels of multiple sample objects that execute the target business.

[0058] In order to improve the accuracy of risk assessment, as an example, based on sample attribute data and risk control labels of a large number of sample objects performing target business, the risk control threshold of the target attribute variable is determined, and then the risk assessment of the target object is performed based on the risk control threshold of the target attribute variable.

[0059] Therefore, sample attribute data and risk control labels of multiple sample objects are obtained from the business database. The sample attribute data includes basic information of the sample object, external risk control data of the sample object outside the financial institution, and business risk control data within the financial institution, such as age, gender, occupation, credit records of the sample object in other financial institutions or third-party data providers, historical transaction records of the sample object in the financial institution, repayment records, overdue conditions, number of defaults, etc.

[0060] Step 202: based on the risk control label, select the target attribute variable from multiple candidate attribute variables.

[0061] Among them, multiple candidate attribute variables are located in sample attribute data of multiple sample objects.

[0062] In order to improve the pertinence of risk assessment, as a possible implementation method, target attribute variables with greater influence are screened out from multiple candidate attribute variables in the sample attribute data, and risk assessment is performed based on the risk control threshold of the target attribute variable.

[0063] As an example, based on the risk control label, the information values ​​of multiple candidate attribute variables are calculated, so as to screen out the target attribute variable from the multiple candidate attribute variables according to the information values ​​of the multiple candidate attribute variables. The multiple candidate attribute variables are located in the sample attribute data of the multiple sample objects, and the multiple candidate attribute variables include, for example, age, credit rating, income, debt, etc.

[0064] The calculation steps of the information values ​​of multiple candidate attribute variables are as follows:

[0065] 1. For any candidate attribute variable, obtain multiple candidate bins of the attribute value of any candidate attribute variable;

[0066] As an example, the attribute value of any candidate attribute variable is binned according to a set binning strategy to obtain multiple candidate bins for the attribute value of any candidate attribute variable. For example, if the candidate attribute variable is age, binning is performed according to the set age value, with those under 35 years old as one candidate bin and those over 35 years old as another candidate bin, etc.; or, binning is performed according to set age intervals; for another example, if the candidate attribute variable is occupation, binning is performed according to the type of occupation, and the occupations in each candidate bin are the same.

[0067] 2. Determine the number of positive sample objects and the number of negative sample objects in any candidate sub-box according to the risk control label of each sample object in any candidate sub-box;

[0068] That is, based on the risk control label of each sample object in any candidate bin, the first number of positive sample objects and the second number of negative sample objects in any candidate bin are counted. The positive sample object is a sample object with a positive risk control label, and the negative sample object is a sample object with a negative risk control label. The positive label is used to indicate that the risk level is lower than the set threshold, and the negative label is used to indicate that the risk level is not lower than the set threshold.

[0069] 3. Calculate the information value of any candidate attribute variable based on the number of positive sample objects and the number of negative sample objects in each candidate bin;

[0070] As an example, according to the number of positive sample objects and the number of negative sample objects in each candidate bin, the specific formula for calculating the information value of any candidate attribute variable is as follows:

[0071]

[0072] Among them, i is the i-th candidate bin of the attribute value of any candidate attribute variable, good i is the number of positive sample objects in the i-th candidate bin, bad i is the number of negative sample objects in the i-th candidate bin, good is the number of positive sample objects in multiple sample objects, bad is the number of negative sample objects in multiple sample objects, and n is the number of bins for the attribute value of any candidate attribute variable.

[0073] As another example, based on the credibility of each candidate attribute variable, a target attribute variable is screened out from multiple candidate attribute variables. It should be noted that the credibility of the candidate attribute variable is determined according to the source of the attribute value of the candidate attribute variable. For example, the attribute value of the candidate attribute variable is directly or indirectly inferred based on the information on the certificate issued by the relevant department, and the credibility is relatively high; for example, the candidate attribute variables are gender, age, etc., and the attribute value of the candidate attribute variable can be determined by the information on the identity card of the corresponding sample object; for another example, the attribute value of the candidate attribute variable is obtained through a third-party data query, and the credibility is second; for example, the candidate attribute variables are working city, consumption habits, commuting distance, etc.; for another example, the information entered by the sample object in the relevant interface has the lowest credibility, such as marital status, education, occupation, etc.

[0074] Step 203: construct a variable decision tree associated with the target business according to the target attribute variables.

[0075] In order to intuitively display the relationship and importance of each target attribute variable, as a possible implementation method, each variable combination corresponds to a variable decision tree, and the first-level nodes and non-first-level nodes of the corresponding variable decision tree are determined based on the information value of each target attribute variable in the variable combination.

[0076] In summary, by obtaining sample attribute data and risk control labels of multiple sample objects that execute the target business; based on the risk control labels, the target attribute variables are screened out from multiple candidate attribute variables; according to the target attribute variables, a variable decision tree associated with the target business is constructed, thereby screening out the most influential target attribute variables from the candidate attribute variables, thereby improving the rationality, accuracy and reliability of risk assessment, and constructing a variable decision tree through the target attribute variables, which is convenient for relevant business personnel to intuitively understand the impact of each variable on risk assessment and better perform risk management.

[0077] In order to clearly illustrate how to construct a variable decision tree associated with a target business according to target attribute variables in the above embodiment, the present disclosure proposes another risk assessment method.

[0078] Figure 3 It is a flowchart of the risk assessment method shown in the third embodiment of the present disclosure.

[0079] like Figure 3 As shown, based on any of the above embodiments, the specific steps of constructing a variable decision tree associated with the target business according to the target attribute variable are as follows:

[0080] Step 301, combining target attribute variables to obtain multiple variable combinations.

[0081] Among them, the number of target attribute variables in the variable combination is less than the set quantity threshold.

[0082] For example, the number of target attribute variables is m. There are m possibilities to select 1 variable from m variables, m(m-1) / 2 possibilities to select 2 variables from m variables, and m(m-1)…(m-n+1) / n! possibilities to select n variables from m variables. The above m+m(m-1) / 2+…+m(m-1)…(m-n+1) / n! possible variable combinations together constitute multiple variable combinations. It should be noted that in order to avoid too many variable combinations, the number of target attribute variables in the variable combination is less than the set number threshold, for example, n is less than or equal to 5.

[0083] Step 302: for any variable combination, generate a variable decision tree according to the information value of each target attribute variable in the any variable combination.

[0084] In order to improve the accuracy and interpretability of risk assessment, as a possible implementation method, the variables with the highest information value are selected from multiple target attribute variables as nodes of each layer of the variable decision tree.

[0085] As an example, the third attribute value of each target attribute variable in any variable combination is obtained; for any target attribute variable in any variable combination, multiple sample objects that execute the target business are binned according to the third attribute value of any target attribute variable to obtain multiple first bins corresponding to the third attribute value of any target attribute variable; for any first bin, the first number of positive sample objects and the second number of negative sample objects in any first bin are determined according to the risk control labels of each sample object in any first bin; the first information value of any target attribute variable is determined according to the first number and the second number in each first bin; and the first-level node in the variable decision tree to be generated is determined from each target attribute variable in any variable combination according to each first information value.

[0086] That is to say, the server can obtain the impact of different attribute values ​​on risk assessment, such as binning multiple sample objects that execute the target business according to different attribute values, and taking the attribute value corresponding to the highest strategy improvement degree (positive sample rate of a certain bin / positive sample rate of the corresponding variable combination) as the third attribute value. For any target attribute variable in any variable combination, obtain the third attribute value to obtain multiple first bins obtained by binning multiple sample objects that execute the target business; for each first bin, count the risk control labels of each sample object in the bin; calculate the first number of positive sample objects in the bin; calculate the second number of negative sample objects in the bin; determine the first information value of any target attribute variable based on the first number and the second number in each first bin; and select the variable with the highest information value as the first-level node of the variable decision tree based on the first information value of each target attribute variable.

[0087] The first information value of any target attribute variable can be determined in the following way:

[0088] 1. For any first sub-bin, calculate the positive sample ratio of any first sub-bin according to the first number of any first sub-bin; wherein the positive sample ratio is the ratio of the number of positive sample objects in the corresponding first sub-bin to the number of positive sample objects in all first sub-bins;

[0089] 2. Calculate the negative sample ratio of any first sub-bin according to the second number of any first sub-bin; wherein the negative sample ratio is the ratio of the number of negative sample objects in the corresponding first sub-bin to the number of negative sample objects in all first sub-bins;

[0090] 3. Calculate the first information value of any target attribute variable based on the positive sample ratio and negative sample ratio of each first bin.

[0091] For example, obtain the first difference between the positive sample ratio and the negative sample ratio, and the first ratio between the positive sample ratio and the negative sample ratio; determine the first information sub-value of any first bin based on the first difference and the first ratio; add the first information sub-values ​​corresponding to each first bin of any target attribute variable to obtain the first information value of any target attribute variable.

[0092] The specific formula for determining the first information value may refer to the formula for determining the information value of the candidate attribute variable in step 202, and will not be described in detail in this disclosure.

[0093] For example, the third attribute value of "income level" is 5,000 yuan, the third attribute value of "credit score" is 600 points, and the third attribute value of "years of work experience" is 5 years. For "income level", the binning interval is assumed to be [0,5000], (5000,8000]. For "credit score", the binning interval is assumed to be [0,600], (600,800]; for "years of work experience", the binning interval is assumed to be [0,5], (5,10]. The information value of "income level" is 0.3, the information value of "credit score" is 0.5, and the information value of "years of work experience" is 0.2. The information value of "credit score" is the highest (0.5), so "credit score" is selected as the first-level node of the variable decision tree.

[0094] As another example, a fourth attribute value of each target attribute variable in any variable combination sent by a client is received; wherein the fourth attribute value is generated by the client in response to a fourth adjustment operation triggered by a threshold adjuster of the target attribute variable and sent to the server; for any target attribute variable in any variable combination, according to the fourth attribute value of any target attribute variable, the sample objects of the first bin of the target attribute variable of the i-1th layer having a parent-child relationship with the i-th layer are binned to obtain multiple second bins corresponding to the fourth attribute value of any target attribute variable; wherein i is a positive integer greater than 1; for any second bin, according to the risk control label of the sample object in any second bin, the third number of positive sample objects and the fourth number of negative sample objects in any second bin are determined; according to the third number and the fourth number corresponding to each second bin, the second information value of any target attribute variable is determined; according to each second information value, the i-th layer node is determined from each target attribute variable in any variable combination.

[0095] That is to say, the server can obtain the impact of non-attribute values ​​on risk assessment, and bin the sample objects in the first bin of the target attribute variable of the i-1th layer that has a parent-child relationship with the i-th layer according to different attribute values, and take the attribute value corresponding to the highest strategy improvement degree (positive sample rate of a certain bin / positive sample rate of the corresponding variable combination) as the fourth attribute value; the server further bins the sample objects in the first bin of the target attribute variable of the i-1th layer that has a parent-child relationship according to the fourth attribute value of any target attribute variable; obtain multiple second bins; for each second bin, count the risk control labels of each sample object in the second bin; calculate the number of positive sample objects in the second bin (third number), and calculate the number of negative sample objects in the bin (fourth number); determine the second information value of any target attribute variable according to the third number and fourth number corresponding to each second bin; according to the second information value of each target attribute variable, select the variable with the highest information value as the i-th layer node.

[0096] For example, suppose the node of the i-1th layer is "credit score" and the sample objects have been binned into [0,600], (600,800]; select the sample objects in the interval of (600,800] and further bin these sample objects. For the first bin [0,600] of "credit score", suppose the bin interval of "income level" is [0,5000], (5000,8000], the second information value of "income level" is 1.386; suppose the bin interval of "years of work experience" is [0,5], (5,10], the second information value of "years of work experience" is 1.099; for the first bin [0,600] of "credit score", select "income level" as the child node.

[0097] In summary, by combining the target attribute variables, multiple variable combinations are obtained; for any variable combination, a variable decision tree is generated according to the information value of each target attribute variable in any variable combination. Thus, by combining multiple target attribute variables, the interaction between different variables is comprehensively considered, so as to have a more comprehensive understanding of the key factors affecting risk assessment. The variable decision tree generated based on actual data and information value provides scientific data support for business decisions, reduces the error of subjective judgment, and improves the scientificity and reliability of decision-making.

[0098] In order to clearly explain how the first evaluation index in the above embodiment is generated, the present disclosure proposes another risk assessment method.

[0099] Figure 4 It is a flowchart of the risk assessment method shown in the fourth embodiment of the present disclosure.

[0100] like Figure 4As shown, based on any of the above embodiments, the first evaluation index is generated by following steps 401 to 403:

[0101] Step 401 : for any target attribute variable, obtain a fifth number of positive sample objects and a sixth number of negative sample objects in any third bin corresponding to a first attribute value of any target attribute variable.

[0102] In order to facilitate the calculation of the first evaluation index and improve the interpretability of the first evaluation index, as a possible implementation method, the first evaluation index of each target attribute variable is calculated based on the bins corresponding to the attribute values ​​of each target attribute variable.

[0103] As an example, multiple third bins corresponding to the first attribute value of any target attribute variable are obtained; for any third bin, according to the risk control label of the sample object in any third bin, the number of positive sample objects and negative sample objects in any third bin is counted to obtain the fifth number of positive sample objects and the sixth number of negative sample objects in any third bin.

[0104] For example, taking the target attribute variable as "age", the first attribute value is 36.5 years old, and "age" is divided into less than 36.5 years old and greater than or equal to 36.5 years old; among them, the number of sample objects less than 36.5 years old is 10,000, the number of samples greater than or equal to 36.5 years old is 40,000, the number of positive sample objects less than 36.5 years old is 8,000, the number of negative sample objects is 2,000, the number of positive sample objects greater than or equal to 36.5 years old is 35,000, and the number of negative sample objects is 5,000.

[0105] Step 402: Determine a first evaluation index for any third sub-box based on the fifth quantity and the sixth quantity corresponding to any third sub-box.

[0106] It should be noted that the first evaluation index includes positive evaluation indicators and negative evaluation indicators. As shown in Table 1, Table 1 is the first evaluation index table. The positive evaluation indicators include: the number of positive samples, the positive sample rate, the strategy coverage rate, the IV value, the strategy improvement rate, etc. The negative evaluation indicators include: the number of negative samples, the negative sample rate, etc.

[0107] The larger the positive evaluation index, the better; the smaller the negative evaluation index, the better.

[0108] Table 1 The first evaluation index table

[0109] Evaluation indicators Evaluation Criteria Number of negative samples The smaller the total sample size, the better. Number of positive samples Given a certain total sample size, the larger the better Negative sample rate The smaller the better Positive sample rate Bigger is better Strategy coverage The higher the better IV Value The higher the better Strategy improvement The higher the better

[0110] Among them, the number of negative samples is the number of negative sample objects concerned by the corresponding target variable in each variable combination, the number of positive samples is the number of positive sample objects concerned by the corresponding target variable in each variable combination, and the number of samples of the target variable is the sum of the number of positive sample objects concerned by the target variable and the number of negative sample objects; negative sample rate = number of negative samples / number of samples of the corresponding target variable; positive sample rate = number of positive samples / number of samples of the corresponding target variable; strategy coverage rate = number of samples covered by a certain bin / total number of all samples of the corresponding variable combination; strategy lift = positive sample rate of a certain bin / positive sample rate of the corresponding variable combination.

[0111] For example, the total number of all samples of a certain variable combination is 50,000. Taking the target attribute variable "age" as an example, the number of sample objects under 36.5 years old is 10,000, and the number of samples greater than or equal to 36.5 years old is 40,000. The number of positive sample objects under 36.5 years old is 8,000, and the number of negative sample objects is 2,000. The number of positive sample objects greater than or equal to 36.5 years old is 35,000, and the number of negative sample objects is 5,000. The positive sample rate corresponding to the bin of less than 36.5 years old is 8,000 / 10,000 = 80%, the negative sample rate is 2,000 / 10,000 = 20%, the strategy coverage rate is 10,000 / 50,000 = 20%, and the strategy improvement is 80% / 86% = 93%.

[0112] Step 403: Determine the first evaluation index of each target attribute variable according to the first evaluation index of each third sub-box.

[0113] In the embodiment of the present disclosure, each third bin corresponds to a first evaluation index, and if the target attribute variable includes multiple third bins, then the target attribute variable corresponds to multiple first evaluation indexes.

[0114] In summary, by obtaining, for any target attribute variable, the fifth number of positive sample objects and the sixth number of negative sample objects in any third bin corresponding to the first attribute value of any target attribute variable; determining the first evaluation index of any third bin based on the fifth number and the sixth number corresponding to any third bin; determining the first evaluation index of each target attribute variable according to the first evaluation index of each third bin, thereby, through automated binning and statistical methods, the first evaluation index of each bin can be quickly calculated, saving the time and cost of manual calculation, and realizing refined assessment of risks in different intervals, providing scientific data support for business decisions, reducing errors in subjective judgment, and improving the scientificity and reliability of decisions.

[0115] In order to implement the above embodiment, the present disclosure proposes another risk estimation method.

[0116] Figure 51 is a flow chart of the risk assessment method shown in the fifth embodiment of the present disclosure. It should be noted that the risk assessment method of the embodiment of the present disclosure is applied to the server.

[0117] like Figure 5 As shown, the risk assessment method includes the following steps:

[0118] Step 501: receiving a first attribute value of each target attribute variable in a variable decision tree associated with a target business of a financial institution sent by a client.

[0119] The first attribute value is generated and sent by the client in response to a first adjustment operation triggered by a threshold adjuster of the target attribute variable.

[0120] Step 502: Determine a first evaluation index for each target attribute variable according to the risk control labels of multiple sample objects associated with the first attribute value of each target attribute variable.

[0121] Step 503: Send the first evaluation index of each target attribute variable to the client, so that the variable decision tree and each first evaluation index are displayed through the client.

[0122] Among them, the first evaluation indicator is used to determine the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable in response to the difference between the first evaluation indicator and the second evaluation indicator of each target attribute variable being less than the set indicator threshold; obtain the first attribute data associated with the target object of the target business to be executed; perform risk assessment on the target object according to the first attribute data and the risk control threshold of each target attribute variable; wherein the second evaluation indicator is generated according to the second attribute value of the target attribute variable generated by the client in response to the second adjustment operation triggered by the threshold adjuster and sent to the server; the second adjustment operation is triggered before the first adjustment operation.

[0123] The risk assessment method of the disclosed embodiment receives the first attribute value of each target attribute variable in the variable decision tree associated with the target business of the financial institution sent by the client; wherein the first attribute value is generated and sent by the client in response to the first adjustment operation triggered by the threshold adjuster of the target attribute variable; according to the risk control labels of multiple sample objects associated with the first attribute value of each target attribute variable, the first evaluation index of each target attribute variable is determined; the first evaluation index of each target attribute variable is sent to the client, so that the variable decision tree and each first evaluation index are displayed through the client; wherein the first evaluation index is used to determine each target attribute variable according to the first attribute value of each target attribute variable in response to the difference between the first evaluation index and the second evaluation index being less than the set index threshold. The risk control threshold of the target attribute variable is obtained; the first attribute data associated with the target object of the target business to be executed is obtained; according to the first attribute data and the risk control threshold of each target attribute variable, the risk assessment is performed on the target object; wherein the second evaluation index is generated according to the second attribute value of the target attribute variable generated by the client in response to the second adjustment operation triggered by the threshold adjuster and sent to the server; the second adjustment operation is triggered before the first adjustment operation, thereby, through the interaction between the client and the server, the risk control threshold is dynamically adjusted and optimized, which can significantly improve the accuracy of risk assessment, and the variable decision tree and the first evaluation index are displayed so that business personnel can better understand the impact of each variable on risk assessment, facilitate the formulation and optimization of business rules, and further enhance the risk management capabilities of financial institutions.

[0124] It should be noted that the implementation method of the embodiment of the present disclosure may adopt the implementation method of any of the above embodiments, and the present disclosure will not elaborate on them.

[0125] Based on any of the above embodiments, Figure 6 As shown, the risk assessment method of the embodiment of the present disclosure is implemented based on the following steps.

[0126] Step 1: Select some indicators from all indicators and include them in the variable pool;

[0127] Step 2: Construct variable combinations;

[0128] Step 3: Calculate the evaluation index of each bin of each target variable in all variable combinations

[0129] Commonly used evaluation indicators include the number of negative samples, the number of positive samples, the negative sample rate, the positive sample rate, the strategy coverage rate, IV value, missing rate and strategy improvement rate.

[0130] Step 4: Filter the thresholds with better rating indicators corresponding to the bins of the target variable as output;

[0131] It should be noted that, Figure 7As shown in the figure, when you need to adjust the threshold of a variable in the strategy, you can adjust the threshold of some or all variables in the variable combination by sliding the front-end joystick (threshold adjuster), and check whether the data on the key indicator dashboard is reached. If it reaches the expectation, stop adjusting, otherwise continue adjusting. Generally speaking, for discrete variables, there is no need to adjust the threshold, while for numerical variables, the purpose of adjustment is to make the threshold more business-interpretable. For example, adjust the age threshold to 35 or 40, and adjust the income to 7,000 or 6,000, etc.

[0132] In addition, it should be noted that variable combinations are generally displayed in a tree structure, so that the detailed process of constructing variable combinations can be intuitively seen; through the process of constructing variable combinations, combined with the key indicator values ​​in the interactive operation interface, it is easier to adjust thresholds with high business interpretability and good key indicators.

[0133] With the above Figures 1 to 4 Corresponding to the risk assessment method provided in the embodiment, the present disclosure further provides a risk assessment device. Since the risk assessment device provided in the embodiment of the present disclosure corresponds to the risk assessment method provided in the above embodiment, the implementation method of the risk assessment method is also applicable to the risk assessment device provided in the embodiment of the present disclosure, and will not be described in detail in the embodiment of the present disclosure.

[0134] Figure 8 1 is a schematic diagram of the structure of a risk assessment device according to a sixth embodiment of the present disclosure, wherein the risk assessment device is applied to a client.

[0135] like Figure 8 As shown, the risk assessment device 800 includes: a receiving module 810 , a first determining module 820 , a first acquiring module 830 and an assessing module 840 .

[0136] The receiving module 810 is used to receive the first evaluation index of each target attribute variable in the variable decision tree associated with the target business of the financial institution sent by the server, and display the variable decision tree and each first evaluation index; wherein the first evaluation index is determined according to the risk control labels of multiple sample objects associated with the first attribute value of the corresponding target attribute variable, and the first attribute value is generated by the client in response to the first adjustment operation triggered by the threshold regulator of the target attribute variable and sent to the server; the risk control label is used to indicate the risk level of the sample object in executing the target business; the first determination module 820 is used to determine the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable in response to the difference between the first evaluation index and the second evaluation index of each target attribute variable being less than the set index threshold; wherein the second evaluation index is generated according to the client in response to the second adjustment operation triggered by the threshold regulator and sent to the server; the second adjustment operation is triggered before the first adjustment operation; the first acquisition module 830 is used to acquire the first attribute data associated with the target object of the target business to be executed; the evaluation module 840 is used to perform risk evaluation on the target object according to the first attribute data and the risk control threshold of each target attribute variable.

[0137] As a possible implementation of the embodiment of the present disclosure, the variable decision tree is generated using the following modules: a second acquisition module, a screening module, and a construction module.

[0138] Among them, the second acquisition module is used to obtain sample attribute data and risk control labels of multiple sample objects that execute the target business; the screening module is used to screen out the target attribute variables from multiple candidate attribute variables based on the risk control labels; wherein the multiple candidate attribute variables are located in the sample attribute data of multiple sample objects; the construction module is used to construct a variable decision tree associated with the target business based on the target attribute variables.

[0139] As a possible implementation method of the embodiment of the present disclosure, a construction module is used to combine each of the target attribute variables to obtain multiple variable combinations; wherein the number of target attribute variables in the variable combination is less than a set quantity threshold; for any variable combination, a variable decision tree is generated according to the information value of each target attribute variable in the any variable combination.

[0140] As a possible implementation method of an embodiment of the present disclosure, the information value includes a first information value, and a construction module is used to obtain the third attribute value of each target attribute variable in any variable combination; for any target attribute variable in any variable combination, according to the third attribute value of any target attribute variable, multiple sample objects that execute the target business are binned to obtain multiple first bins corresponding to the third attribute value of any target attribute variable; for any first bin, according to the risk control label of each sample object in any first bin, determine the first number of positive sample objects and the second number of negative sample objects in any first bin; determine the first information value of any target attribute variable according to the first number and the second number in each first bin; and determine the first-level node in the variable decision tree to be generated from each target attribute variable in any variable combination according to each first information value.

[0141] As a possible implementation method of the embodiment of the present disclosure, a module is constructed to calculate, for any first sub-bin, the proportion of positive samples in any first sub-bin according to the first quantity of any first sub-bin; calculate the proportion of negative samples in any first sub-bin according to the second quantity of any first sub-bin; and calculate the first information value of any target attribute variable according to the positive sample proportion and the negative sample proportion of each first sub-bin.

[0142] As a possible implementation method of an embodiment of the present disclosure, the information value also includes a second information value, and the construction module is also used to obtain the fourth attribute value of each target attribute variable in any variable combination; for any target attribute variable in any variable combination, according to the fourth attribute value of any target attribute variable, the sample objects of the first bin of the target attribute variable of the i-1th layer having a parent-child relationship with the i-th layer are binned to obtain multiple second bins corresponding to the fourth attribute value of any target attribute variable; wherein i is a positive integer greater than 1; for any second bin, according to the risk control label of the sample object in any second bin, determine the third number of positive sample objects and the fourth number of negative sample objects in any second bin; determine the second information value of any target attribute variable according to the third number and the fourth number corresponding to each second bin; and determine the i-th layer node from each target attribute variable in any variable combination according to each second information value.

[0143] As a possible implementation manner of the embodiment of the present disclosure, the first evaluation index is generated by using the following modules: a third acquisition module and a second determination module.

[0144] Among them, the third acquisition module is used to obtain, for any target attribute variable, the fifth number of positive sample objects and the sixth number of negative sample objects in any third bin corresponding to the first attribute value of any target attribute variable; the second determination module is used to determine the first evaluation index of any third bin based on the fifth number and sixth number corresponding to any third bin; and determine the first evaluation index of each target attribute variable according to the first evaluation index of each third bin.

[0145] As a possible implementation method of the embodiment of the present disclosure, the third acquisition module is used to obtain multiple third bins corresponding to the first attribute value of any target attribute variable; for any third bin, according to the risk control label of the sample object in any third bin, the number of positive sample objects and negative sample objects in any third bin is counted to obtain the fifth number of positive sample objects and the sixth number of negative sample objects in any third bin.

[0146] As a possible implementation method of the embodiment of the present disclosure, the evaluation module 840 is used to obtain the attribute value of any target attribute variable from the first attribute data for any target attribute variable; in response to the attribute value of any target attribute variable being greater than or equal to the risk control threshold of any target attribute variable, determine that the risk level of the target object executing the target business is greater than the set risk level.

[0147] The risk assessment device of the disclosed embodiment receives the first evaluation index of each target attribute variable in the variable decision tree associated with the target business of the financial institution sent by the server, and displays the variable decision tree and each first evaluation index; wherein the first evaluation index is determined according to the risk control labels of multiple sample objects associated with the first attribute value of the corresponding target attribute variable, and the first attribute value is generated by the client in response to the first adjustment operation triggered by the threshold adjuster of the target attribute variable and sent to the server; the risk control label is used to indicate the risk level of the sample object in executing the target business; in response to the difference between the first evaluation index and the second evaluation index of each target attribute variable being less than the set index threshold, the risk control label of each target attribute variable is determined according to the first attribute value of each target attribute variable. The risk control threshold of the target attribute variable; wherein the second evaluation index is generated and sent to the server according to the second adjustment operation triggered by the threshold adjuster in response to the client; the second adjustment operation is triggered before the first adjustment operation; the first attribute data associated with the target object of the target business to be executed is obtained; according to the first attribute data and the risk control threshold of each target attribute variable, the risk assessment of the target object is performed, thereby, through the interaction between the client and the server, the risk control threshold is dynamically adjusted and optimized, which can significantly improve the accuracy of risk assessment, and the variable decision tree and the first evaluation index are displayed so that business personnel can better understand the impact of each variable on risk assessment, facilitate the formulation and optimization of business rules, and further enhance the risk management capabilities of financial institutions.

[0148] With the above Figure 5 Corresponding to the risk assessment method provided in the embodiment, the present disclosure further provides a risk assessment device. Since the risk assessment device provided in the embodiment of the present disclosure corresponds to the risk assessment method provided in the above embodiment, the implementation method of the risk assessment method is also applicable to the risk assessment device provided in the embodiment of the present disclosure, and will not be described in detail in the embodiment of the present disclosure.

[0149] Fig. 9 1 is a schematic diagram of the structure of a risk assessment device according to the seventh embodiment of the present disclosure, wherein the risk assessment device is applied to a server.

[0150] like Fig. 9 As shown, the risk assessment device 900 includes: a receiving module 910 , a determining module 920 and a sending module 930 .

[0151] Among them, the receiving module 910 is used to receive the first attribute value of each target attribute variable in the variable decision tree associated with the target business of the financial institution sent by the client; wherein the first attribute value is generated and sent by the client in response to the first adjustment operation triggered by the threshold adjuster of the target attribute variable; the determining module 920 is used to determine the first evaluation index of each target attribute variable according to the risk control labels of multiple sample objects associated with the first attribute value of each target attribute variable; the sending module 930 is used to send the first evaluation index of each target attribute variable to the client, so as to display the variable decision tree and each first evaluation index through the client; wherein The first evaluation indicator is used to determine the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable in response to the difference between the first evaluation indicator and the second evaluation indicator of each target attribute variable being less than the set indicator threshold; obtain the first attribute data associated with the target object of the target business to be executed; perform risk assessment on the target object according to the first attribute data and the risk control threshold of each target attribute variable; wherein the second evaluation indicator is generated according to the second attribute value of the target attribute variable generated by the client in response to the second adjustment operation triggered by the threshold adjuster and sent to the server; the second adjustment operation is triggered before the first adjustment operation.

[0152] The risk assessment device of the disclosed embodiment receives the first attribute value of each target attribute variable in the variable decision tree associated with the target business of the financial institution sent by the client; wherein the first attribute value is generated and sent by the client in response to the first adjustment operation triggered by the threshold adjuster of the target attribute variable; determines the first evaluation index of each target attribute variable according to the risk control labels of multiple sample objects associated with the first attribute value of each target attribute variable; sends the first evaluation index of each target attribute variable to the client, so as to display the variable decision tree and each first evaluation index through the client; wherein the first evaluation index is used to determine each target attribute variable according to the first attribute value of each target attribute variable in response to the difference between the first evaluation index and the second evaluation index being less than the set index threshold. The risk control threshold of the target attribute variable is obtained; the first attribute data associated with the target object of the target business to be executed is obtained; according to the first attribute data and the risk control threshold of each target attribute variable, the risk assessment is performed on the target object; wherein the second evaluation index is generated according to the second attribute value of the target attribute variable generated by the client in response to the second adjustment operation triggered by the threshold adjuster and sent to the server; the second adjustment operation is triggered before the first adjustment operation, thereby, through the interaction between the client and the server, the risk control threshold is dynamically adjusted and optimized, which can significantly improve the accuracy of risk assessment, and the variable decision tree and the first evaluation index are displayed so that business personnel can better understand the impact of each variable on risk assessment, facilitate the formulation and optimization of business rules, and further enhance the risk management capabilities of financial institutions.

[0153] In an exemplary embodiment, an electronic device is also provided.

[0154] Among them, electronic equipment includes:

[0155] processor;

[0156] a memory for storing processor-executable instructions;

[0157] The processor is configured to execute instructions to implement the risk assessment method proposed in any of the aforementioned embodiments.

[0158] As an example, Fig.10 is a schematic diagram of the structure of an electronic device 1000 shown in an exemplary embodiment of the present disclosure, such as Fig.10 As shown, the electronic device 1000 may further include:

[0159] The memory 1010 and the processor 1020, the bus 1030 connecting different components (including the memory 1010 and the processor 1020), the memory 1010 stores a computer program, and when the processor 1020 executes the program, the risk assessment method described in the embodiment of the present disclosure is implemented.

[0160] Bus 1030 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.

[0161] The electronic device 1000 typically includes a variety of electronic device readable media. These media can be any available media that can be accessed by the electronic device 1000, including volatile and non-volatile media, removable and non-removable media.

[0162] The memory 1010 may also include computer system readable media in the form of volatile memory, such as random access memory (RAM) 1040 and / or cache memory 1050. The server 1000 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 1060 may be used to read and write non-removable, non-volatile magnetic media ( Fig.10 not shown, usually called a "hard drive"). Although Fig.10 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 1030 via one or more data medium interfaces. The memory 1010 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present disclosure.

[0163] A program / utility 1080 having a set (at least one) of program modules 1070 may be stored, for example, in the memory 1010, such program modules 1070 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 1070 generally perform the functions and / or methods of the embodiments described in the present disclosure.

[0164] The electronic device 1000 may also communicate with one or more external devices 1090 (e.g., keyboard, pointing device, display 1091, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 1000, and / or communicate with any device that enables the electronic device 1000 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed through an input / output (I / O) interface 1092. In addition, the electronic device 1000 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) through a network adapter 1093. As shown, the network adapter 1093 communicates with other modules of the electronic device 1000 through a bus 1030. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1000, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0165] The processor 1020 executes various functional applications and data processing by running the programs stored in the memory 1010 .

[0166] It should be noted that the implementation process and technical principles of the electronic device of this embodiment refer to the aforementioned explanation of the risk assessment method of the embodiment of the present disclosure, and will not be repeated here.

[0167] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory including instructions, and the above instructions can be executed by a processor of an electronic device to complete the risk assessment method proposed in any of the above embodiments. Optionally, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0168] In an exemplary embodiment, a computer program product is also provided, including a computer program / instruction, wherein the computer program / instruction implements the risk assessment method proposed in any of the above embodiments when executed by a processor.

[0169] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0170] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A risk assessment method, characterized in that: Applied to the client, including: Receive a first evaluation index of each target attribute variable in a variable decision tree associated with a target business of a financial institution sent by a service end, and display the variable decision tree and each of the first evaluation indexes; wherein the first evaluation index is determined according to risk control tags of a plurality of sample objects associated with a first attribute value of the corresponding target attribute variable, the first attribute value is generated by the client in response to a first adjustment operation triggered by a threshold adjuster of the target attribute variable and sent to the service end; the risk control tag is used to indicate the risk level of the sample object in executing the target business; In response to the difference between the first evaluation index and the second evaluation index of each target attribute variable being less than a set index threshold, determining the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable; wherein the second evaluation index is generated and sent to the server according to the second adjustment operation triggered by the threshold adjuster in response to the client; the second adjustment operation is triggered before the first adjustment operation; Acquire first attribute data associated with a target object to be executed with the target business; Perform risk assessment on the target object according to the first attribute data and the risk control threshold of each of the target attribute variables.

2. The method according to claim 1, characterized in that The variable decision tree is generated by the following steps: Acquire sample attribute data and risk control tags of multiple sample objects that execute the target business; Based on the risk control label, a target attribute variable is screened out from a plurality of candidate attribute variables; wherein the plurality of candidate attribute variables are located in the sample attribute data of the plurality of sample objects; A variable decision tree associated with the target business is constructed according to the target attribute variable.

3. The method according to claim 2, characterized in that The step of constructing a variable decision tree associated with the target business according to the target attribute variable includes: Combining the target attribute variables to obtain a plurality of variable combinations; wherein the number of the target attribute variables in the variable combinations is less than a set number threshold; For any variable combination, a variable decision tree is generated according to the information value of each target attribute variable in the any variable combination.

4. The method according to claim 3, characterized in that The information value includes a first information value, and for any variable combination, generating a variable decision tree according to the information value of each target attribute variable in the any variable combination includes: Obtaining a third attribute value of each of the target attribute variables in any variable combination; For any target attribute variable in any variable combination, binning is performed on a plurality of sample objects that execute the target business according to a third attribute value of the any target attribute variable, so as to obtain a plurality of first bins corresponding to the third attribute value of the any target attribute variable; For any first sub-box, determining a first number of positive sample objects and a second number of negative sample objects in any first sub-box according to the risk control label of each sample object in the any first sub-box; Determine a first information value of any target attribute variable according to the first quantity and the second quantity in each of the first bins; According to each of the first information values, a first-level node in a variable decision tree to be generated is determined from each of the target attribute variables in any of the variable combinations.

5. The method according to claim 4, characterized in that The determining, according to the first quantity and the second quantity in each of the first sub-boxes, the first information value of any target attribute variable comprises: For any first sub-bin, calculating the positive sample ratio of any first sub-bin according to the first quantity of any first sub-bin; Calculate the negative sample ratio of any of the first sub-bins according to the second number of any of the first sub-bins; The first information value of any target attribute variable is calculated according to the positive sample ratio and the negative sample ratio of each of the first bins.

6. The method according to claim 4, characterized in that The information value further includes a second information value, and the method further includes: Obtaining the fourth attribute value of each of the target attribute variables in any variable combination; For any target attribute variable in any variable combination, according to the fourth attribute value of any target attribute variable, the sample objects of the first bin of the target attribute variable of the i-1th layer having a parent-child relationship with the i-th layer are binned to obtain a plurality of second bins corresponding to the fourth attribute value of any target attribute variable; wherein i is a positive integer greater than 1; For any second sub-bin, determine a third number of positive sample objects and a fourth number of negative sample objects in any second sub-bin according to the risk control labels of the sample objects in the second sub-bin; determine a second information value of any target attribute variable according to the third number and the fourth number corresponding to each of the second sub-bins; According to each of the second information values, the i-th layer node is determined from each of the target attribute variables in any of the variable combinations.

7. The method according to claim 1, characterized in that The first evaluation index is generated by the following steps: For any target attribute variable, obtaining a fifth number of positive sample objects and a sixth number of negative sample objects in any third bin corresponding to the first attribute value of the any target attribute variable; Determine a first evaluation index of any third sub-box based on the fifth quantity and the sixth quantity corresponding to any third sub-box; According to the first evaluation index of each of the third bins, the first evaluation index of each of the target attribute variables is determined.

8. The method according to claim 7, characterized in that The step of obtaining, for any target attribute variable, a fifth number of positive sample objects and a sixth number of negative sample objects in any third bin corresponding to the first attribute value of any target attribute variable comprises: Obtain a plurality of third bins corresponding to the first attribute value of any target attribute variable; For any third sub-box, according to the risk control labels of the sample objects in the any third sub-box, the positive sample objects and the negative sample objects in the any third sub-box are counted to obtain the fifth number of positive sample objects and the sixth number of negative sample objects in the any third sub-box.

9. The method according to claim 1, characterized in that: The step of performing risk assessment on the target object according to the first attribute data and the risk control threshold of each target attribute variable includes: For any target attribute variable, obtaining an attribute value of the any target attribute variable from the first attribute data; In response to the attribute value of any target attribute variable being greater than or equal to the risk control threshold of any target attribute variable, it is determined that the risk level of the target object executing the target business is greater than a set risk level.

10. A risk estimation method, characterized in that: Applied to the server, including: Receiving a first attribute value of each target attribute variable in a variable decision tree associated with a target business of a financial institution sent by a client; wherein the first attribute value is generated and sent by the client in response to a first adjustment operation triggered by a threshold adjuster of the target attribute variable; Determining a first evaluation index for each of the target attribute variables according to the risk control labels of the plurality of sample objects associated with the first attribute value of each of the target attribute variables; The first evaluation indicator of each target attribute variable is sent to the client so that the variable decision tree and each first evaluation indicator are displayed through the client; wherein the first evaluation indicator is used to determine the risk control threshold of each target attribute variable according to the first attribute value of each target attribute variable in response to the difference between the first evaluation indicator and the second evaluation indicator of each target attribute variable being less than the set indicator threshold; obtain the first attribute data associated with the target object of the target business to be executed; perform risk assessment on the target object according to the first attribute data and the risk control threshold of each target attribute variable; wherein the second evaluation indicator is generated according to the second attribute value of the target attribute variable generated by the client in response to the second adjustment operation triggered by the threshold adjuster and sent to the server; the second adjustment operation is triggered before the first adjustment operation.

Images