Risk assessment method and device for replacement resources, equipment, medium and program product
By obtaining multi-dimensional information of users, constructing derivative features and determining their importance parameters, the problem of low prediction accuracy of repayment default risk in the prior art is solved, and higher prediction accuracy is achieved.
Patent Information
- Application Number
- CN202411373109.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-29
- Publication Date
- 2025-01-07
AI Technical Summary
The prior art has low accuracy in predicting repayment default risks after user resource replacement, mainly due to the fewer information elements, which leads to the unreliable prediction results.
By obtaining user's asset information, transaction information, repayment information and loan information, extracting sub-data, constructing derivative characteristics and determining their importance parameters, and predicting repayment default risks based on these characteristics to improve the accuracy of the prediction.
Through multi-dimensional analysis and screening of importance parameters of derivative characteristics, the prediction accuracy of repayment default risk is significantly improved and more reliable risk assessment results are provided.
Smart Images
Figure CN119273453A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of big data processing technology, and in particular, relates to a risk assessment method, device, equipment, medium and program product for replacement resources. Background Art
[0002] As people's consumption concepts change, more and more users are exchanging resources with financial institutions and using the exchanged resources for consumption.
[0003] After the user replaces resources or replaces resources, it is necessary to predict the user's repayment default risk. In the exemplary technology, the user's repayment and borrowing information is used to monitor the user's repayment default risk, but this monitoring method uses fewer information elements and the prediction accuracy of the repayment default risk is low. Summary of the invention
[0004] The embodiments of the present application provide a risk assessment method, apparatus, device, medium and program product for replacement resources, which are used to solve the problem of low prediction accuracy of repayment default risk.
[0005] In a first aspect, an embodiment of the present application provides a risk assessment method for replacing resources, including:
[0006] Acquire first target information corresponding to the first user, where the first target information includes asset information, transaction information, repayment information, and loan information of the first user;
[0007] Extracting a plurality of first sub-data from the first target information, and acquiring a first feature corresponding to each of the first sub-data;
[0008] Creating a first derived feature corresponding to the first feature according to the attributes of each of the first features, and determining a first importance parameter of each of the first derived features;
[0009] Determining a first target feature from each of the first derived features according to a first importance parameter of the first derived features, wherein the first importance parameter of the first target feature is greater than a preset threshold;
[0010] According to each of the first target characteristics, a repayment default risk parameter of the replacement resource applied for by the first user is predicted.
[0011] In some possible implementations, the step of predicting the repayment default risk parameter of the replacement resource applied for by the first user according to each of the first target characteristics includes:
[0012] Acquire a first feature of a variable type from each of the first features as a second feature;
[0013] Based on each of the second characteristics and each of the first target characteristics, a repayment default risk parameter of the replacement resource applied for by the first user is predicted.
[0014] In some possible implementations, the step of determining a first importance parameter of each first derived feature includes:
[0015] According to each first attribute corresponding to the first feature to which the first derived feature belongs;
[0016] Among the first attributes, determine a first attribute belonging to the asset type as a first target attribute;
[0017] A first importance parameter of the first derived feature is determined according to the quantity and weight of each of the first target attributes.
[0018] In some possible implementations, the step of predicting the repayment default risk parameter of the replacement resource applied for by the first user according to each of the first target characteristics includes:
[0019] Each of the first target features is input into an evaluation model to obtain a repayment default risk parameter of the replacement resource applied for by the first user output by the evaluation model.
[0020] In some possible implementations, before the step of inputting each of the first target features into the evaluation model, the step further includes:
[0021] Acquire each training sample, the training sample including a plurality of second target features and a label, each second target feature is determined based on second target information of a second user, the second target information including asset information, transaction information, repayment information and borrowing information of the second user, and the label is used to indicate a repayment default risk parameter of a replacement resource applied for by the second user;
[0022] The preset model is trained according to each of the training samples to obtain the evaluation model.
[0023] In some possible implementations, the step of obtaining each training sample includes:
[0024] Acquire second target information of a second user, and extract a plurality of second sub-data from the second target information;
[0025] Acquire the third feature corresponding to each of the second sub-data, and create a second derived feature corresponding to the third feature according to the attribute of the third feature;
[0026] determining a second importance parameter of each of the second derived features, and determining a second target feature in each of the second derived features according to the second importance parameter of the second derived features;
[0027] A label corresponding to the second target feature is configured, and a training sample corresponding to the second user is constructed according to the second target feature and the label.
[0028] In some possible implementations, the step of constructing a training sample corresponding to the second user according to the second target feature and the label includes:
[0029] Obtaining a third feature of a variable type from each of the third features as a fourth feature;
[0030] A training sample is constructed according to each of the fourth features, each of the second target features and the label.
[0031] In some possible implementations, the step of determining a second importance parameter of each second derived feature includes:
[0032] According to each second attribute corresponding to the third feature to which the second derived feature belongs;
[0033] Among the second attributes, determine a second attribute belonging to the asset type as a second target attribute;
[0034] A second importance parameter of the second derived feature is determined according to the quantity and weight of each of the second target attributes.
[0035] In some possible implementations, the step of obtaining the second target information of the second user includes:
[0036] Obtaining the asset information, transaction information, repayment information, and loan information of the second user;
[0037] The asset information, transaction information, repayment information and loan information of the second user are processed to obtain the second target information, wherein the data processing includes at least one of data deduplication processing, missing value processing, outlier deletion, data format conversion and classification variable encoding.
[0038] In a second aspect, an embodiment of the present application provides a risk assessment device for replacing resources, including:
[0039] An acquisition module, configured to acquire first target information corresponding to a first user, wherein the first target information includes asset information, transaction information, repayment information, and loan information of the first user;
[0040] an extraction module, configured to extract a plurality of first sub-data from the first target information, and obtain a first feature corresponding to each of the first sub-data;
[0041] a creation module, configured to create a first derived feature corresponding to the first feature according to the attributes of each of the first features, and determine a first importance parameter of each of the first derived features;
[0042] a determination module, configured to determine a first target feature from each of the first derived features according to a first importance parameter of the first derived features, wherein the first importance parameter of the first target feature is greater than a preset threshold;
[0043] A prediction module is used to predict the repayment default risk parameters of the replacement resources applied for by the first user according to each of the first target characteristics.
[0044] In a third aspect, an embodiment of the present application provides a risk assessment device for replacing resources, the device comprising: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the risk assessment method for replacing resources as described above.
[0045] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, characterized in that computer program instructions are stored on the computer-readable storage medium, and when the computer program instructions are executed by a processor, the risk assessment method for replacement resources as described above is implemented.
[0046] In a fifth aspect, an embodiment of the present application provides a computer program product. When instructions in the computer program product are executed by a processor of an electronic device, the electronic device is enabled to perform the risk assessment method for replacing resources as described above.
[0047] The risk assessment method, device, equipment, medium and program product of the replacement resource of the embodiment of the present application obtains the target information of the user including asset information, transaction information, loan information and repayment information, extracts multiple sub-data from the target information, and obtains the first feature corresponding to each sub-data, constructs the derivative feature corresponding to the first feature based on the attributes of each first feature, and determines the importance parameter of the derivative feature, and then obtains the derivative feature whose importance parameter is greater than the preset threshold as the target feature, so as to predict the repayment default risk of the replacement resource applied by the user based on each target feature. In the present application, multiple derivative features are determined from multiple dimensions such as asset information, transaction information, repayment information and loan information, that is, the repayment default risk of the user for the replacement resource is predicted based on multiple factors, which improves the prediction accuracy of the repayment default risk. In addition, based on the importance parameter of the derivative feature, the derivative feature that has a significant contribution to the repayment default risk is determined among the multiple derivative features, and the repayment default risk is predicted, which further improves the prediction accuracy of the repayment default risk. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.
[0049] Figure 1 It is a schematic diagram of the scenario involved in the risk assessment method of replacing resources in this application;
[0050] Figure 2 It is a flowchart of the first embodiment of the risk assessment method for replacing resources of the present application;
[0051] Figure 3 It is a flowchart of the second embodiment of the risk assessment method for replacing resources of the present application;
[0052] Figure 4 It is a flowchart of the third embodiment of the risk assessment method for replacing resources of the present application;
[0053] Figure 5 is a flowchart of a fourth embodiment of the risk assessment method for replacing resources of the present application;
[0054] Figure 6 is a flowchart of a fifth embodiment of the risk assessment method for replacing resources of the present application;
[0055] Figure 7 It is a functional module schematic diagram of the risk assessment device for replacing resources in the present application;
[0056] Figure 8 It is a schematic diagram of the structure of the risk assessment equipment for replacing resources in this application. DETAILED DESCRIPTION
[0057] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.
[0058] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0059] It should be noted that the acquisition, storage, use and processing of data in the embodiments of the present application are in compliance with the relevant provisions of national laws and regulations.
[0060] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.
[0061] As people's consumption concepts change, more and more users are exchanging resources with financial institutions and using the exchanged resources for consumption.
[0062] As people's consumption concepts change, more and more users exchange resources with financial institutions and use the exchanged resources for consumption. After users exchange resources or replace resources, it is necessary to predict the repayment default risk of users.
[0063] The inventors of the present application have discovered that currently, the repayment and borrowing information of users is used to monitor the repayment default risk of users. However, this monitoring method uses fewer information elements and the prediction accuracy of the repayment default risk is low.
[0064] The inventor of the present application therefore came up with the idea of determining the derivative features that have a significant contribution to the repayment default risk among multiple derivative features based on the importance parameters of the derivative features, and predicting the repayment default risk, thereby further improving the prediction accuracy of the repayment default risk.
[0065] Reference Figure 1 , Figure 1 This is a schematic diagram of the scenario of the risk assessment method for replacing resources for this application. Figure 1 , the user 100 can send an application request for replacing resources to the risk assessment device 300 for replacing resources through the terminal device 200. The risk assessment device 300 for replacing resources obtains the target information of the user 100 including asset information, transaction information, repayment information and loan information. The risk assessment device 300 for replacing resources extracts multiple sub-data from the target information, obtains the first feature corresponding to each sub-data, creates all corresponding derivative features of the first feature based on the attributes of each first feature, and determines the importance parameter of each derivative feature. Based on the importance parameter of the derivative feature, the derivative feature with an importance parameter greater than a preset threshold is determined as the target feature in each derivative feature, and then the repayment default risk parameter of the replacement resource applied by the user is predicted based on each target feature. When the repayment default risk parameter is lower than the predicted risk parameter, the risk assessment device 300 for replacing resources agrees to the application request for replacing resources of the user 100, does not configure an account for the user 100, and outputs the login information of the account to the terminal device 200, so that the user 100 can log in to the account based on the login information to use the replacement resource.
[0066] The following combination Figure 1 , the risk assessment method for resource replacement in this application is described in detail.
[0067] Reference Figure 2 , Figure 2 FIG. 1 is a flow chart of a first embodiment of a risk assessment method for replacing resources of the present application. The risk assessment method for replacing resources includes:
[0068] Step S201, obtaining first target information corresponding to a first user, the first target information including asset information, transaction information, repayment information and loan information of the first user.
[0069] In this embodiment, the execution subject is a risk assessment device for replacing resources. For the sake of convenience, the term "device" is used below to refer to the risk assessment device for replacing resources. The device can be a physical device such as a server or a computer, a virtual device such as a cloud server, or a device, equipment, or system with a default risk prediction function.
[0070] The device obtains target information corresponding to a user, the user is defined as a first user, and the target information is defined as first target information. The first user can be an individual, or a group or group consisting of multiple users. The first user can be a user who has replaced resources, or a user who is applying for replacing resources.
[0071] When the first user is applying for a replacement resource, the first user sends an application request for a replacement resource to the device, and the application request includes the first target information of the first user, that is, the first user needs to apply for a replacement resource based on the first target information. The device can obtain the first target information by parsing the application request.
[0072] If the first user has completed the application for replacement resources, the device will then authorize the first user to use the replacement resources on the condition of obtaining the first user's repayment information, borrowing information, asset information and transaction information. Therefore, the device obtains the repayment information, borrowing information, asset information and transaction information as the first target information on the application authorized by the first user.
[0073] Step S202: extract multiple first sub-data from the first target information, and obtain a first feature corresponding to each first sub-data.
[0074] After obtaining the first target information, the device extracts a plurality of sub-data from the first target information, and the sub-data are defined as first sub-data. Each first sub-data is repayment information, loan information, asset information and transaction information.
[0075] Transaction information includes unstructured information, such as transaction descriptions, market dynamics and other information. Such unstructured information cannot be effectively processed and utilized in the prior art, so the repayment default risk prediction accuracy in the prior art is low. In this embodiment, the unstructured information in the transaction information is converted into features, which can be feature vectors. Positive values are assigned to positive unstructured information, and negative values are assigned to negative unstructured information. For example, if the transaction description is that the transaction is not smooth enough and the payment has been urged many times, it is negative unstructured information; for another example, if the market dynamics characterize that such transactions are stable, it is positive unstructured information. The transaction information also includes the transaction amount, etc.
[0076] Asset information includes the current deposit balance of the first user, the transaction amount in the past month, the transaction amount in the past six months, the valuation of real estate assets, etc. Repayment information includes the repayment record of the first user, which includes the repayment amount and repayment time. When the repayment is overdue, the repayment record also includes the late payment fee. Borrowing information includes the number of other replacement resources of the first user, the amount of other replacement resources, etc.
[0077] After obtaining the first target information, the device needs to perform data cleaning on the first target information. Data cleaning includes deduplication, missing value processing, and outlier identification and processing.
[0078] Deduplication: For the account number, customer number, etc. of the replacement resource, duplicate records are detected and deduplicated to ensure the uniqueness of each piece of data in the first target information.
[0079] Missing value processing: For missing values, adopt appropriate filling strategies. For example, for numerical variables such as current deposit balance, you can use the median or mean to fill, and for categorical variables such as guarantee method, you can use the mode to fill. In particular, if the missing ratio of key variables is too high, remove the variable or mark the missing state as a sub-data.
[0080] Identification and processing of outliers: Identify outliers through statistical analysis methods (such as box plots), and decide whether to modify, delete, or retain outliers with high or low transaction amounts based on business logic. For example, if the business logic requires the deletion of outliers, the outliers will be deleted.
[0081] After data cleaning is performed on the first target information, each first sub-data is extracted from the cleaned first target information, and a first feature corresponding to the first sub-data is determined.
[0082] Specifically, the first sub-data is standardized or categorically encoded. Standardization: Standardization is performed on numerical variables, such as transaction amounts, deposit balances, etc. For example, Z-score standardization is performed to eliminate the impact of dimensions and provide a certain degree of stability for repayment default risks.
[0083] Encoding of categorical variables: For categorical variables, such as guarantee methods, overdue variables, transaction descriptions, and market dynamics in unstructured information, one-hot encoding is used to convert them into numerical types, so that unstructured data can be used.
[0084] Step S203: creating first derived features corresponding to the first features according to the attributes of each first feature, and determining a first importance parameter of each first derived feature.
[0085] After obtaining each first feature, the device constructs a derivative feature corresponding to the first feature based on the attributes of each first feature, and the derivative feature is defined as the first derivative feature. This method is feature engineering, that is, based on business knowledge and data exploration results, new features are derived from the original variables. For example, if the first feature is the transaction amount in the past month, the first derivative feature of the average transaction amount can be derived, which can enhance the prediction ability of repayment default risk.
[0086] After determining each first derived feature, determine the importance parameter of each derived feature, and define the importance parameter as the first importance parameter. The first importance parameter is used to indicate the contribution of the first derived feature to the prediction accuracy of the repayment default risk. The greater the contribution, the greater the first importance parameter. Exemplarily, a value associated with the type of the first derived feature is obtained as the first importance parameter.
[0087] Step S204: determining a first target feature from each first derived feature according to a first importance parameter of the first derived feature, wherein the first importance parameter of the first target feature is greater than a preset threshold.
[0088] After determining the first importance parameter of the first derivative feature, the device determines a first target feature among the first derivative features. The first target feature is the first derivative feature whose first importance parameter is greater than a preset threshold value, that is, the first target feature is a feature that contributes more to the accuracy of prediction of repayment default risk.
[0089] Step S205: predicting the repayment default risk parameters of the replacement resources applied for by the first user according to each first target feature.
[0090] After determining each first target feature, a repayment default risk parameter of the replacement resource applied for by the first user is predicted based on each first target feature.
[0091] Exemplarily, multiple preset feature vectors are set in the device, each preset feature vector is associated with a numerical value used to characterize the repayment default risk parameter, the device constructs each first target feature into a target feature vector, calculates the similarity between the target feature vector and each preset feature vector, and the numerical value associated with the preset feature vector with the highest similarity is the predicted repayment default risk parameter.
[0092] The device can display the repayment default risk parameters so that the reviewer can determine whether to approve the application request of the first user based on the repayment default risk parameters or whether to adopt a collection strategy to collect the repayment from the first user who has applied for resource replacement.
[0093] It should be noted that the repayment default risk parameter can be a specific value, for example, the repayment default risk parameter is a default rate of 95%. In addition, the repayment default risk parameter can also be a risk level, and the risk level includes a low risk level, a medium risk level, and a high risk level. If the predicted probability value is less than 20%, it is a low risk level, if the predicted probability value is greater than or equal to 20% and less than 50%, it is a medium risk level, and if the predicted probability value is greater than or equal to 50%, it is a high risk level. In addition, the risk assessment method for replacing resources provided in this embodiment can also be used to replace resources applied by users. The resources must be reviewed, that is, the risk assessment of the replacement resources can be used to replace them. is the probability of overdue repayment after the user applies for resource replacement, so the repayment default risk parameter can be expressed as the risk probability parameter Make a replacement.
[0094] In this embodiment, the target information of the user including asset information, transaction information, loan information and repayment information is obtained, multiple sub-data are extracted from the target information, and the first feature corresponding to each sub-data is obtained, and the derivative feature corresponding to the first feature is constructed based on the attributes of each first feature, and the importance parameter of the derivative feature is determined, and then the derivative feature whose importance parameter is greater than the preset threshold is obtained as the target feature, so as to predict the repayment default risk of the replacement resource applied by the user based on each target feature. In this embodiment, multiple derivative features are determined from multiple dimensions such as asset information, transaction information, repayment information and loan information, that is, the repayment default risk of the user for the replacement resource is predicted based on multiple factors, which improves the prediction accuracy of the repayment default risk. In addition, based on the importance parameter of the derivative feature, the derivative feature that has a significant contribution to the repayment default risk is determined among the multiple derivative features, and the repayment default risk is predicted, which further improves the prediction accuracy of the repayment default risk.
[0095] Reference Figure 3 , Figure 3 This is a flow chart of a second embodiment of the risk assessment method for replacing resources in the present application. Based on the first embodiment, step S205 includes:
[0096] Step S301: acquiring a first feature of a variable type from each first feature as a second feature.
[0097] In this embodiment, in the prediction process, special attention should be paid to the selection and use of variable content. For example, the inflow transaction amount of the first user in the past month and the outflow transaction amount of the first user in the past six months play an important role in predicting overdue. These variables should be used to predict the repayment default risk, that is, the actual variables also play a greater role in the prediction accuracy of the repayment default risk. In this regard, the device obtains the first feature of the variable type from each first feature as the second feature. The second feature is, for example, the current deposit balance, the number of warnings before overdue, etc.
[0098] Step S302: predicting the repayment default risk parameters of the replacement resources applied for by the first user based on each second feature and each first target feature.
[0099] The device predicts the repayment default risk parameters of the replacement resources applied for by the first user based on each second feature and each first target feature.
[0100] Exemplarily, a plurality of preset feature vectors are set in the device, and each preset feature vector is associated with a value for characterizing the repayment default risk parameter. The device constructs each second feature and each first target feature into a target feature vector, calculates the similarity between the target feature vector and each preset feature vector, and the value associated with the preset feature vector with the highest similarity is the predicted repayment default risk parameter.
[0101] In this embodiment, the device obtains the first feature of the variable type from the first feature as the second feature, and thereby predicts the repayment default risk parameters of the replacement resources applied for by the first user based on the second feature and each first target feature, that is, based on the variable parameters and the highly important derivative features, accurately predicts the repayment default risk parameters of the first user.
[0102] Reference Figure 4 , Figure 4 This is a flow chart of a third embodiment of the risk assessment method for replacing resources in the present application. Based on the first or second embodiment, step S203 includes:
[0103] Step S401: first attributes corresponding to the first feature to which the first derived feature belongs.
[0104] In this embodiment, the device obtains each attribute corresponding to the first feature to which the first derived feature belongs, and the attribute is defined as a first attribute. The attribute is, for example, the type and identifier of the first feature.
[0105] Step S402: among the first attributes, determine the first attribute belonging to the asset type as the first target attribute.
[0106] The first user's assets contribute more to the accurate prediction of repayment default risk. In this regard, the device determines the first attribute belonging to the asset type as the first target attribute among the first attributes. For example, deposits, real estate, etc.
[0107] Step S403: determining a first importance parameter of the first derived feature according to the quantity and weight of each first target attribute.
[0108] After determining each first target attribute, the device can determine the first importance parameter of the first derived feature based on the quantity and weight of each first target attribute.
[0109] Exemplarily, the more attributes of the asset type, the more the amount that can be used for repayment by the first user can be determined, and thus, an initial value can be determined based on the quantity, and the larger the quantity, the larger the initial value. Furthermore, the asset type has a weight, and the product of the weight and the initial value can obtain the first importance parameter. The weight can be set based on the quantity, and the larger the quantity, the larger the weight, that is, the larger the number of asset types, the more the amount that can be used for repayment, and the greater the contribution to the accuracy of the repayment default risk prediction, so the weight is positively correlated with the quantity.
[0110] In this embodiment, the device determines the first attribute belonging to the asset type as the first target attribute among the first attributes based on the first attributes corresponding to the first feature to which the first derived feature belongs, thereby accurately determining the importance parameter of the first derived feature based on the number and weight of the first target attributes.
[0111] Reference Figure 5 , Figure 5 This is a flow chart of a fourth embodiment of the risk assessment method for replacing resources of the present application, based on any one of the first to third embodiments, step S205 includes:
[0112] Step S501, input each first target feature into the evaluation model, and obtain the repayment default risk parameter of the replacement resource applied for by the first user output by the evaluation model.
[0113] In this embodiment, an evaluation model is provided in the device. The device inputs each first target feature into the evaluation model, and the parameter output by the evaluation model is the repayment default risk parameter of the replacement resource applied for by the first user.
[0114] In this embodiment, the evaluation model is a random forest model. The random forest model includes multiple decision trees. The evaluation model integrates the prediction results of each decision tree and outputs the repayment default risk parameter. The random forest model can effectively capture the complex nonlinear relationship between data, so that the credit risk of users can be more accurately evaluated, especially for groups with complex operating environments and financial status.
[0115] In addition, the first target feature plays a relatively important role in the prediction of the model and can improve the accuracy and efficiency of the random forest model.
[0116] It should be noted that after obtaining each second feature, the device can input each second feature and the first target feature into the evaluation model, so that the evaluation model outputs more accurate repayment default risk parameters.
[0117] In this embodiment, the device inputs the first target feature into the evaluation model, thereby obtaining the accurate repayment default risk parameter output by the evaluation model.
[0118] Reference Figure 6 , Figure 6 This is a flowchart of the fifth embodiment of the risk assessment method for replacing resources in the present application. Based on the fourth embodiment, step S501 includes:
[0119] Step S601, obtaining each training sample, the training sample includes multiple second target features and labels, each second target feature is determined based on the second target information of the second user, the second target information includes the second user's asset information, transaction information, repayment information and borrowing information, and the label is used to indicate the repayment default risk parameter of the replacement resource applied for by the second user.
[0120] In this embodiment, the device obtains each training sample, the training sample includes multiple second target features and labels, each second target feature is determined based on the second target information of the second user, the second target information includes the asset information, transaction information, repayment information and loan information of the second user, and the label identifies the repayment default analysis parameter of the replacement resource applied by the second user. The asset information, transaction information, repayment information and loan information of the second user are the same as the information types contained in the asset information, transaction information, repayment information and loan information of the first user, and will not be described in detail.
[0121] In one example, the device obtains the second target information of the second user, and extracts the second sub-data from the second target information. The extraction method of the second sub-data is the same as the extraction method of the first sub-data, which will not be repeated here. The device obtains the third feature corresponding to each second sub-data, and creates a second derivative feature corresponding to the third feature based on the attribute of the third feature; the acquisition method of the third feature is the same as the acquisition method of the first feature, and the second derivative feature is determined in the same way as the first derivative feature, which will not be repeated here. After obtaining the second derivative feature, the device determines the second importance parameter of the second derivative feature, and determines the second target feature in each second derivative feature according to the second importance parameter of the second derivative feature. The second target feature is a second derivative feature whose second importance parameter is greater than a preset threshold. The determination method of the second importance parameter is the same as the determination method of the first importance parameter, which will not be repeated here. The device configures a label corresponding to the second target feature, and the label can be set by the device based on the user's input information on the second target feature, that is, the input information includes a repayment default risk parameter. The device can construct a training sample corresponding to the second user based on the second target feature and the label.
[0122] Further, the device obtains the third feature of the variable type from each third feature as the fourth feature, and then constructs a training sample according to each fourth feature, each second target feature and the label. The method for obtaining the fourth feature is the same as the method for obtaining the second feature, and will not be repeated here.
[0123] In addition, when determining the second importance parameter of the second derived feature, the device obtains the second attribute corresponding to the third feature to which the second derived feature belongs, determines the second attribute belonging to the asset type among the second attributes as the second target attribute, and thereby determines the second importance parameter of the second derived feature based on the number and weight of the second target attribute. The process of determining the second importance parameter of the second derived feature based on the number and weight of the second target attribute is the same as the process of determining the first importance parameter of the second derived feature based on the number and weight of the first target attribute, and is not repeated here.
[0124] Furthermore, the device obtains the asset information, transaction information, repayment information, and loan information of the second user; performs data processing on the asset information, transaction information, repayment information, and loan information of the second user to obtain the second target information, and the data processing includes at least one of data deduplication processing, missing value processing, outlier deletion, data format conversion, and categorical variable encoding. For details of deduplication processing, missing value processing, outlier deletion, data format conversion, and categorical variable encoding, refer to the above description.
[0125] Step S602: training the preset model according to each training sample to obtain an evaluation model.
[0126] After obtaining each training sample, the device can train the preset model to obtain the evaluation model. The preset model is a random forest model. In addition, the device will regularly update the evaluation model.
[0127] In this embodiment, the device accurately obtains training samples based on the second target information of the second user, thereby training the evaluation model through the training samples.
[0128] See also Figure 7 The risk assessment device 700 for replacing resources provided in the embodiment of the present application includes:
[0129] An acquisition module 710 is used to acquire first target information corresponding to a first user, where the first target information includes asset information, transaction information, repayment information, and loan information of the first user;
[0130] An extraction module 720 is used to extract a plurality of first sub-data from the first target information, and obtain a first feature corresponding to each first sub-data;
[0131] A creation module 730, configured to create first derived features corresponding to the first features according to the attributes of each first feature, and determine a first importance parameter of each first derived feature;
[0132] A determination module 740 is used to determine a first target feature from each first derived feature according to a first importance parameter of the first derived feature, wherein the first importance parameter of the first target feature is greater than a preset threshold;
[0133] The prediction module 750 is used to predict the repayment default risk parameters of the replacement resources applied for by the first user according to each first target feature.
[0134] In one embodiment, the risk assessment device 700 for replacing resources is specifically used for:
[0135] Obtaining the first feature of the variable type from each first feature as the second feature;
[0136] According to each second feature and each first target feature, a repayment default risk parameter of the replacement resource applied for by the first user is predicted.
[0137] In one embodiment, the risk assessment device 700 for replacing resources is specifically used for:
[0138] According to each first attribute corresponding to the first feature to which the first derived feature belongs;
[0139] Among the first attributes, determine the first attribute belonging to the asset type as the first target attribute;
[0140] A first importance parameter of the first derived feature is determined according to the quantity and weight of each first target attribute.
[0141] In one embodiment, the risk assessment device 700 for replacing resources is specifically used for:
[0142] Each first target feature is input into the evaluation model to obtain the repayment default risk parameter of the replacement resource applied for by the first user output by the evaluation model.
[0143] In one embodiment, the risk assessment device 700 for replacing resources is specifically used for:
[0144] Acquire each training sample, the training sample includes a plurality of second target features and a label, each second target feature is determined based on the second target information of the second user, the second target information includes the asset information, transaction information, repayment information and borrowing information of the second user, and the label is used to indicate the repayment default risk parameter of the replacement resource applied for by the second user;
[0145] The preset model is trained according to each training sample to obtain an evaluation model.
[0146] In one embodiment, the risk assessment device 700 for replacing resources is specifically used for:
[0147] Acquire second target information of a second user, and extract a plurality of second sub-data from the second target information;
[0148] Acquire the third feature corresponding to each second sub-data, and create a second derived feature corresponding to the third feature according to the attribute of the third feature;
[0149] Determine a second importance parameter of each second derived feature, and determine a second target feature in each second derived feature according to the second importance parameter of the second derived feature;
[0150] A label corresponding to the second target feature is configured, and a training sample corresponding to the second user is constructed according to the second target feature and the label.
[0151] In one embodiment, the risk assessment device 700 for replacing resources is specifically used for:
[0152] Obtain the third feature of the variable type from each third feature as the fourth feature;
[0153] A training sample is constructed according to each fourth feature, each second target feature and a label.
[0154] In one embodiment, the risk assessment device 700 for replacing resources is specifically used for:
[0155] According to each second attribute corresponding to the third feature to which the second derived feature belongs;
[0156] Among the second attributes, determine the second attribute belonging to the asset type as the second target attribute;
[0157] A second importance parameter of the second derived feature is determined according to the quantity and weight of each second target attribute.
[0158] In one embodiment, the risk assessment device 700 for replacing resources is specifically used for:
[0159] Obtaining the second user's asset information, transaction information, repayment information, and loan information;
[0160] Data processing is performed on the second user's asset information, transaction information, repayment information, and loan information to obtain second target information, wherein the data processing includes at least one of data deduplication processing, missing value processing, outlier deletion, data format conversion, and categorical variable encoding.
[0161] Each module of the risk assessment device for replacing resources provided in the embodiment of the present application can realize Figures 2 to 6Any of the embodiments provides the functions of each step of the risk assessment method for replacing resources and can achieve its corresponding technical effects. For the sake of concise description, they will not be repeated here.
[0162] Figure 8 A schematic diagram of the hardware structure of a risk assessment device for replacement resources provided in an embodiment of the present application is shown.
[0163] The risk assessment device for replacing resources may include a processor 801 and a memory 802 storing computer program instructions.
[0164] Specifically, the processor 801 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.
[0165] The memory 802 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 802 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 802 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 802 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 802 is a non-volatile solid-state memory.
[0166] In certain embodiments, the memory 802 may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage media device, an optical storage media device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.
[0167] The processor 801 implements any one of the authentication methods in the above embodiments by reading and executing computer program instructions stored in the memory 802 .
[0168] In one example, the risk assessment device for replacing resources may further include a communication interface 803 and a bus 810. Figure 8As shown, the processor 801, the memory 802, and the communication interface 803 are connected via a bus 810 and communicate with each other.
[0169] The communication interface 803 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.
[0170] Bus 810 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 810 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the application considers any suitable bus or interconnection.
[0171] The data transmission method in the above embodiment can be implemented by providing a computer storage medium in the present application embodiment. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any one of the data transmission methods in the above embodiment is implemented.
[0172] An embodiment of the present application further provides a computer program product, including a computer program, which implements any one of the data transmission methods in the above embodiments when the computer program is processed and executed.
[0173] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.
[0174] Aspects of the present disclosure are described above with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or a risk assessment device of other programmable permutation resources to produce a machine so that these instructions executed by the processor of the risk assessment device of the computer or other programmable permutation resources enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.
[0175] The above are only specific implementation methods of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.
Claims
1. A risk assessment method for replacement resources, characterized in that: include: Acquire first target information corresponding to the first user, where the first target information includes asset information, transaction information, repayment information, and loan information of the first user; Extracting a plurality of first sub-data from the first target information, and acquiring a first feature corresponding to each of the first sub-data; Creating a first derived feature corresponding to the first feature according to the attributes of each of the first features, and determining a first importance parameter of each of the first derived features; Determining a first target feature from each of the first derived features according to a first importance parameter of the first derived features, wherein the first importance parameter of the first target feature is greater than a preset threshold; According to each of the first target characteristics, a repayment default risk parameter of the replacement resource applied for by the first user is predicted.
2. The risk assessment method for replacement resources according to claim 1, characterized in that: The step of predicting the repayment default risk parameter of the replacement resource applied for by the first user according to each of the first target characteristics comprises: Acquire a first feature of a variable type from each of the first features as a second feature; Based on each of the second characteristics and each of the first target characteristics, a repayment default risk parameter of the replacement resource applied for by the first user is predicted.
3. The risk assessment method for replacement resources according to claim 1, characterized in that: The step of determining a first importance parameter for each of the first derived features comprises: According to each first attribute corresponding to the first feature to which the first derived feature belongs; Among the first attributes, determine a first attribute belonging to the asset type as a first target attribute; A first importance parameter of the first derived feature is determined according to the quantity and weight of each of the first target attributes.
4. The risk assessment method for replacement resources according to claim 1, characterized in that: The step of predicting the repayment default risk parameter of the replacement resource applied for by the first user according to each of the first target characteristics comprises: Each of the first target features is input into an evaluation model to obtain a repayment default risk parameter of the replacement resource applied for by the first user output by the evaluation model.
5. The risk assessment method for replacement resources according to claim 4, characterized in that: Before the step of inputting each of the first target features into the evaluation model, the method further includes: Acquire each training sample, the training sample including a plurality of second target features and a label, each second target feature is determined based on second target information of a second user, the second target information including asset information, transaction information, repayment information and borrowing information of the second user, and the label is used to indicate a repayment default risk parameter of a replacement resource applied for by the second user; The preset model is trained according to each of the training samples to obtain the evaluation model.
6. The risk assessment method for replacement resources according to claim 5, characterized in that: The step of obtaining each training sample comprises: Acquire second target information of a second user, and extract a plurality of second sub-data from the second target information; Acquire the third feature corresponding to each of the second sub-data, and create a second derived feature corresponding to the third feature according to the attribute of the third feature; determining a second importance parameter of each of the second derived features, and determining a second target feature in each of the second derived features according to the second importance parameter of the second derived features; A label corresponding to the second target feature is configured, and a training sample corresponding to the second user is constructed according to the second target feature and the label.
7. The risk assessment method for replacement resources according to claim 6, characterized in that: The step of constructing a training sample corresponding to the second user according to the second target feature and the label includes: Obtaining a third feature of a variable type from each of the third features as a fourth feature; A training sample is constructed according to each of the fourth features, each of the second target features and the label.
8. The risk assessment method for replacement resources according to claim 6, characterized in that: The step of determining a second importance parameter for each of the second derived features comprises: According to each second attribute corresponding to the third feature to which the second derived feature belongs; Among the second attributes, determine a second attribute belonging to the asset type as a second target attribute; A second importance parameter of the second derived feature is determined according to the quantity and weight of each of the second target attributes.
9. The risk assessment method for replacement resources according to claim 6, characterized in that: The step of obtaining the second target information of the second user comprises: Obtaining the asset information, transaction information, repayment information, and loan information of the second user; The asset information, transaction information, repayment information and loan information of the second user are processed to obtain the second target information, wherein the data processing includes at least one of data deduplication processing, missing value processing, outlier deletion, data format conversion and classification variable encoding.
10. A risk assessment device for replacing resources, characterized in that: include: An acquisition module, configured to acquire first target information corresponding to a first user, wherein the first target information includes asset information, transaction information, repayment information, and loan information of the first user; an extraction module, configured to extract a plurality of first sub-data from the first target information, and obtain a first feature corresponding to each of the first sub-data; a creation module, configured to create a first derived feature corresponding to the first feature according to the attributes of each of the first features, and determine a first importance parameter of each of the first derived features; a determination module, configured to determine a first target feature from each of the first derived features according to a first importance parameter of the first derived features, wherein the first importance parameter of the first target feature is greater than a preset threshold; A prediction module is used to predict the repayment default risk parameters of the replacement resources applied for by the first user according to each of the first target characteristics.
11. A risk assessment device for replacing resources, characterized in that: The device comprises: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the risk assessment method for replacement resources as claimed in any one of claims 1 to 9 is implemented.
12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the risk assessment method for replacement resources according to any one of claims 1 to 9 is implemented.
13. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device is enabled to execute the risk assessment method for replacement resources as described in any one of claims 1-9.