Message pushing method and device, equipment, storage medium and program product
By evaluating the model to identify the user's repayment risk level and adjusting the prompt strategy, the problem of low effectiveness of the existing repayment prompt method is solved, and the user's enthusiasm for repayment behavior and credit risk management effect is improved.
Patent Information
- Application Number
- CN202411373986.7
- 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 existing repayment prompt method is relatively low and cannot effectively improve users' enthusiasm for repayment behavior and credit risk management effectiveness.
By obtaining the user's credit information, loan information, repayment information and transaction information, using the evaluation model to identify the repayment risk level, adjust the repayment prompt strategy, and personalize the prompts based on the user's historical repayment behavior and response information.
It improves the user's repayment effectiveness and the pertinence of credit risk management, and achieves more accurate risk assessment and dynamic optimization.
Smart Images
Figure CN119273457A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the technical field of big data processing, and in particular, relates to a message push method, device, equipment, storage medium and program product. 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 completes the application for replacement resources, the user can use the replacement resources, but the user needs to make repayments to repay the replacement resources used.
[0004] In the exemplary technology, a prompt message generated by a fixed message template is used to prompt the user to repay the loan, but this method is less effective for the user to repay the loan. Summary of the invention
[0005] The embodiments of the present application provide a message push method, apparatus, device, storage medium and program product for solving the problem of low effectiveness of existing repayment reminder methods.
[0006] In a first aspect, an embodiment of the present application provides a message push method, comprising:
[0007] Acquire first target information of a first user, where the first user is used to indicate a user who has applied for resource replacement, and the first target information includes credit information, loan information, repayment information, and transaction information of the first user;
[0008] Using an evaluation model to identify the first target information to obtain a repayment risk level of the first user, and determining a first repayment reminder strategy according to the repayment risk level;
[0009] Acquire historical repayment behavior information and historical repayment response information of the first user for the replacement resource, and adjust the first repayment reminder strategy according to the historical repayment behavior information and the historical repayment response information to obtain a second repayment reminder strategy;
[0010] A prompt message is generated according to the second repayment prompt strategy, and the prompt message is sent to a terminal device associated with the first user, wherein the prompt message is used to prompt the first user to repay the replaced resources.
[0011] In some possible implementations, the step of determining a first repayment reminder strategy according to the repayment risk level includes:
[0012] determining the repayment risk group to which the first user belongs according to the repayment risk level;
[0013] The repayment reminder strategy associated with the repayment risk group is determined as the first repayment reminder strategy.
[0014] In some possible implementations, the step of adjusting the first repayment reminder strategy according to the historical repayment behavior information and the historical repayment response information includes:
[0015] Determining a repayment enthusiasm parameter of the first user according to the historical repayment behavior information and the historical repayment response information, wherein the repayment enthusiasm parameter is used to indicate the repayment enthusiasm of the first user;
[0016] The first repayment reminder strategy is adjusted according to the repayment positive parameter.
[0017] In some possible implementations, the step of adjusting the first repayment reminder strategy according to the repayment positive parameter includes:
[0018] Determining a first strategy level corresponding to the first repayment reminder strategy;
[0019] In response to the repayment active parameter indicating that the first user is actively repaying, lowering the first policy level to obtain a second policy level, and determining the repayment reminder policy corresponding to the second policy level as the second repayment reminder policy;
[0020] In response to the repayment enthusiasm parameter indicating that the first user is passive in repayment, the first policy level is increased to obtain a third policy level, and the repayment reminder policy corresponding to the third policy level is determined as the second repayment reminder policy.
[0021] In some possible implementations, the step of using an evaluation model to identify the first target information to obtain the repayment risk level of the first user includes:
[0022] Acquire a plurality of first initial features according to the first target information;
[0023] Converting and deriving the first initial features to obtain a plurality of first target features;
[0024] Each of the first target features is input into the evaluation model to obtain the repayment risk level of the first user output by the evaluation model.
[0025] In some possible implementations, before the step of using the evaluation model to identify the first target information to obtain the repayment risk level of the first user, the step further includes:
[0026] Acquire each training sample, the training sample including second target information and a label, the second target information including credit information, borrowing information, repayment information and transaction information of the second user, the label being used to indicate the repayment risk level of the replacement resource applied for by the second user;
[0027] The preset model is trained according to each of the training samples to obtain the evaluation model.
[0028] In some possible implementations, the step of obtaining each training sample includes:
[0029] Acquire a plurality of second initial features according to the second target information of the second user;
[0030] Converting and deriving the second initial features to obtain a plurality of second target features;
[0031] A label is set for each of the second target features, and a training sample corresponding to the second user is constructed according to the second target features and the label.
[0032] In a second aspect, an embodiment of the present application provides a message push device, including:
[0033] A first acquisition module, used to acquire first target information of a first user, where the first user is used to indicate a user who has applied for resource replacement, and the first target information includes credit information, loan information, repayment information, and transaction information of the first user;
[0034] an identification module, configured to identify the first target information using an evaluation model to obtain a repayment risk level of the first user, and determine a first repayment reminder strategy according to the repayment risk level;
[0035] A second acquisition module is used to obtain the historical repayment behavior information and historical repayment response information of the first user for the replacement resources, and adjust the first repayment reminder strategy according to the historical repayment behavior information and the historical repayment response information to obtain a second repayment reminder strategy;
[0036] A generation module is used to generate a prompt message according to the second repayment prompt strategy, and send the prompt message to the terminal device associated with the first user, wherein the prompt message is used to prompt the first user to repay the replacement resources.
[0037] In a third aspect, an embodiment of the present application provides a message push device, comprising: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the message push method as described above is implemented.
[0038] 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 message push method as described above is implemented.
[0039] 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 execute the message push method as described above.
[0040] The message push method, device, equipment, storage medium and program product of the embodiment of the present application obtains target information including user credit information, loan information, repayment information and transaction information, uses an evaluation model to identify the target information to obtain the user's repayment risk level, and determines a first repayment reminder strategy based on the repayment risk level, adjusts the first repayment reminder strategy to obtain a second repayment reminder strategy through the user's historical repayment behavior information and historical repayment response information for the replacement resource, generates prompt information based on the second repayment prompt strategy, and then sends the prompt information to the terminal device associated with the user to prompt the user to repay the replacement resource. In the present application, the first repayment reminder strategy is preliminarily determined based on the user's repayment risk level, and then the first repayment reminder strategy is adjusted based on the user's historical repayment information behavior and historical repayment response to obtain the second repayment reminder strategy, thereby realizing personalized reminders for users to repay and improving the effectiveness of users' repayments. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] 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.
[0042] Figure 1 It is a schematic diagram of the scenario involved in the message push method of this application;
[0043] Figure 2 This is a flowchart of the first embodiment of the message push method of the present application;
[0044] Figure 3 This is a flow chart of the second embodiment of the message push method of this application;
[0045] Figure 4 This is a flow chart of the third embodiment of the message push method of the present application;
[0046] Figure 5 This is a flowchart of the fourth embodiment of the message push method of the present application;
[0047] Figure 6This is a schematic diagram of the functional modules of the message push device of this application;
[0048] Figure 7 It is a structural diagram of the message push device of this application. DETAILED DESCRIPTION
[0049] 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.
[0050] 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.
[0051] 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.
[0052] 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.
[0053] As people's consumption concepts change, more and more users are exchanging resources with financial institutions and using the exchanged resources for consumption.
[0054] As people's consumption concepts change, more and more users are exchanging resources with financial institutions and using the exchanged resources for consumption.
[0055] After the user completes the application for replacement resources, the user can use the replacement resources, and the user needs to make repayments to repay the replacement resources used.
[0056] The inventor of the present application has found that the prompt information generated by using a fixed message template to prompt the user to repay the loan is less effective for the user to repay the loan.
[0057] The inventor of the present application therefore came up with the idea of preliminarily determining a first repayment reminder strategy based on the user's repayment risk level, and then adjusting the first repayment reminder strategy based on the user's historical repayment information behavior and historical repayment responses to obtain a second repayment reminder strategy, thereby achieving personalized reminders for users to repay and improving the effectiveness of users' repayments.
[0058] Reference Figure 1 , Figure 1 This is a schematic diagram of the scenario of the message push method of this application. Figure 1 , the message push device 100 obtains the target information of the user to be repaid from the database 200. The target information includes the credit information, loan information, repayment information and transaction information of the user to be repaid. An evaluation model is provided in the message push device 100, and the repayment risk level of the user to be repaid is obtained by identifying the target information through the evaluation model, and a first repayment reminder strategy is determined based on the repayment risk level. The message push device 100 obtains the historical repayment behavior information and historical repayment response information of the user to be repaid from the target information, and adjusts the first repayment reminder strategy to obtain a second repayment reminder strategy based on the historical repayment behavior information and the historical repayment response information. The message push device 100 generates a reminder message based on the second repayment reminder strategy, thereby sending the reminder message to the terminal device 300 associated with the user to be repaid.
[0059] Furthermore, in the exemplary technology, there are problems such as insufficient data utilization, static models, and single repayment reminders that lack specificity.
[0060] Specifically, insufficient data utilization: exemplary technologies usually rely only on basic credit data and simple financial indicators, such as the number of approval inquiries for replacement resources and the utilization rate of account limits for replacement resources, etc., lacking analysis of more detailed and complex data. This leads to insufficient accuracy in risk assessment and prediction of users, and is unable to fully understand and predict users' financial behavior and credit risks.
[0061] Static model: The evaluation models in the exemplary technologies are often static, and once built, they are difficult to update in a timely manner according to changes in the market environment and user behavior. This inflexible model cannot accurately reflect the real-time risk situation, reducing the effectiveness of risk management.
[0062] Repayment reminders are single and lack specificity: The exemplary technologies often adopt a unified strategy for repayment reminder strategies, such as sending notification reminders, etc., and do not customize the repayment reminder strategy according to the user's risk level and specific characteristics. This one-size-fits-all approach is not only inefficient, but may also fail to effectively reduce credit risk.
[0063] In order to solve the above technical problems, the following Figure 1 , the message push method of this application is described in detail.
[0064] Reference Figure 2 , Figure 2 FIG. 1 is a flow chart of a first embodiment of a message push method of the present application. The message push method includes:
[0065] Step S201, obtaining first target information of a first user, where the first user is used to indicate a user who has applied for resource replacement, and the first target information includes the first user's credit information, loan information, repayment information, and transaction information.
[0066] In this embodiment, the execution subject is a message push device. For the sake of convenience, the device is used below to refer to the message push device. The device can be a server or a terminal device with a message push function. The exchanged resources refer to the resources that the user exchanges for assets. The assets can be physical assets, such as real estate, or intangible assets, such as credit. The resources can be digital currency.
[0067] The device can periodically poll the user who has applied for resource replacement to predict the user's repayment risk. The user can be an individual, or a group or group consisting of multiple people, and the user is defined as the first user.
[0068] The device obtains the first target information of the first user, and the first target information includes the credit information, loan information, repayment information and transaction information of the first user. The credit information refers to the credit score of the first user. The loan information refers to the number and amount of replacement resources applied by the first user. The repayment information includes the amount, time and overdue days of the repayment made by the first user. The transaction information includes the user's current deposit balance, the amount of the transaction, the transaction description and market dynamics.
[0069] Step S202: Use the evaluation model to identify the first target information to obtain the repayment risk level of the first user, and determine the first repayment reminder strategy according to the repayment risk level.
[0070] An evaluation model is provided in the device, and the evaluation model may be a logistic regression model, which can handle binary classification problems and give the probability of user repayment default.
[0071] The device identifies the first target information through the evaluation model to obtain the repayment risk level. Specifically, the first target information is input into the evaluation model, and the evaluation model outputs the probability of repayment default. The level associated with the interval in which the probability lies is the repayment risk level. The repayment risk level includes low risk level, medium risk level and high risk level. For example, if the probability of repayment default is less than 20%, the repayment risk level is a low risk level. If the probability of repayment default is greater than or equal to 20% and less than 50%, the repayment risk level is a medium risk level. If the probability of repayment default is greater than or equal to 50%, the repayment risk level is a high risk level.
[0072] In one example, different repayment risk levels are associated with corresponding repayment reminder strategies. For example, if the repayment risk level is a high risk level, the repayment reminder strategy includes an active reminder strategy, such as telephone contact and face-to-face interview; if the repayment risk level is a medium risk level, emails, text messages, etc. can be used as repayment reminder strategies; if the repayment risk level is a low risk level, an automated method can be used to remind users to repay, such as providing repayment reminders on the repayment date. The above repayment reminder strategies are only used as examples, but are not limited to the repayment reminder strategies in the above examples.
[0073] In another example, the repayment risk level is associated with a repayment risk group, and the device determines the repayment risk group to which the first user belongs based on the repayment risk level. The repayment risk group is, for example, a high default group, a medium default group, and a low default group. Each repayment risk group is associated with a repayment reminder strategy, and the device determines the repayment reminder strategy associated with the repayment risk group as the first repayment reminder strategy.
[0074] In order to distinguish, the device defines the repayment reminder strategy corresponding to the repayment risk level as the first repayment reminder strategy.
[0075] Step S203, obtaining the historical repayment behavior information and historical repayment response information of the first user for the replacement resource, and adjusting the first repayment reminder strategy according to the historical repayment behavior information and the historical repayment response information to obtain a second repayment reminder strategy.
[0076] The device obtains the historical repayment behavior information and historical repayment response information of the first user for the replacement resources. The historical repayment behavior information and historical repayment response information are the behaviors and responses made by the first user based on the previous repayment reminder strategy. Specifically, the device can obtain the historical repayment behavior information and historical repayment response information from the first target information. The historical repayment behavior information includes various historical repayment behaviors of the first user. The historical repayment behaviors are, for example, the repayment operation behaviors of the first user. The repayment operation behaviors include the operations of performing repayments in a certain channel, the operations of determining the repayment amount, etc., and the historical repayment response information includes the interval between the first user receiving the reminder information and performing the repayment. The length of the interval reflects whether the first user's attitude towards repayment is positive or negative.
[0077] The device adjusts the first repayment reminder strategy through historical repayment behavior information and historical repayment response information to obtain a second repayment reminder strategy.
[0078] Exemplarily, based on the historical repayment behavior information, the change in the repayment behavior of the first user is determined, and the first repayment reminder strategy is adjusted based on the change in the repayment behavior. For example, if the repayment behavior of the first user in the recent period changes to repayment near the repayment deadline, the first user may forget to repay, and the repayment risk level needs to be increased, that is, the first repayment reminder strategy is adjusted to the repayment reminder strategy associated with the repayment risk level of the previous level of the current risk level.
[0079] In addition, the device can determine the enthusiasm parameter of the first user's repayment based on the historical repayment response information, and adjust the first repayment reminder strategy based on the enthusiasm parameter. For example, the enthusiasm parameter refers to the average value of the interval between the first user receiving the reminder message and repaying. If the average value is less than the preset value, it can be determined that the first user is active in repayment, and the repayment risk level is reduced, that is, the first repayment reminder strategy is adjusted to the repayment reminder strategy associated with the repayment risk level of the next level of the current risk level. When the average value is greater than or equal to the preset value, it can be determined that the first user is passive in repayment, and the repayment risk level is increased, that is, the first repayment reminder strategy is adjusted to the repayment reminder strategy associated with the repayment risk level of the previous level of the current risk level.
[0080] As can be seen from the above, the historical repayment behavior information can adjust the first repayment reminder strategy, and the historical repayment response information can also adjust the first repayment reminder strategy. In one example, the first repayment reminder strategy is adjusted in combination with the first adjustment result of the historical repayment behavior information and the second adjustment result of the historical repayment response information. For example, the first adjustment result is that the first repayment reminder strategy is increased by one level (increasing one level means increasing the current risk level by one level), and the first adjustment result and the second adjustment result are both that the first repayment reminder strategy is increased by one level (increasing one level means increasing the current risk level by one level), then the second repayment reminder strategy is the repayment reminder strategy associated with the current repayment risk level after it is increased by two levels.
[0081] In this embodiment, the adjustment of the second repayment reminder strategy is based on the analysis of the execution effect of past strategies. By analyzing the impact of different strategies on customer repayment behavior, the most effective second repayment reminder method is identified. In addition, customer feedback is also an integral part of the adjustment strategy, especially in terms of customer response to certain second repayment reminder measures. Based on this information, the second repayment reminder strategy can be optimized to make it more targeted and effective.
[0082] Step S204: Generate a prompt message according to the second repayment prompt strategy, and send the prompt message to the terminal device associated with the first user, where the prompt message is used to prompt the first user to repay the replaced resources.
[0083] After generating the second repayment reminder strategy, the device generates a reminder message based on the second repayment reminder strategy, and sends the reminder message to the terminal device associated with the first user. The reminder message includes a reminder text and a sending channel. For example, if the sending channel is an email, the email containing the reminder text is sent to the terminal device associated with the first user by email.
[0084] In this embodiment, target information including user credit information, loan information, repayment information and transaction information is obtained, and the target information is identified by an evaluation model to obtain the user's repayment risk level, and a first repayment reminder strategy is determined according to the repayment risk level. The first repayment reminder strategy is adjusted to obtain a second repayment reminder strategy through the user's historical repayment behavior information and historical repayment response information for the replacement resource, and prompt information is generated based on the second repayment prompt strategy, and then the prompt information is sent to the terminal device associated with the user to prompt the user to repay the replacement resource. In this embodiment, the first repayment reminder strategy is preliminarily determined according to the user's repayment risk level, and then the first repayment reminder strategy is adjusted to obtain the second repayment reminder strategy according to the user's historical repayment information behavior and historical repayment response, so as to realize personalized reminders for users to repay and improve the effectiveness of users' repayments.
[0085] Reference Figure 3, Figure 3 This is a flow chart of the second embodiment of the message push method of this application. Based on the first embodiment, step S203 includes:
[0086] Step S301, determining a repayment enthusiasm parameter of the first user according to historical repayment behavior information and historical repayment response information, where the repayment enthusiasm parameter is used to indicate the repayment enthusiasm of the first user.
[0087] In this embodiment, the device determines the repayment positive parameter of the first user based on the historical repayment behavior information and the historical repayment response information.
[0088] Specifically, the first sub-data is extracted from the historical repayment behavior information, and the second sub-data is extracted from the historical repayment response information. The first sub-data is, for example, the repayment ratio, and the second sub-data is, for example, the average value of the interval between the first user receiving the reminder message and the repayment. The repayment ratio maps a first value, and the average value maps a second value. The higher the repayment ratio, the larger the first value, the smaller the average value, and the larger the second value. The device uses the sum of the first value and the second value as a repayment positivity parameter. The repayment positivity parameter is used to indicate the first user's repayment positivity, and the larger the repayment positivity parameter, the higher the first user's repayment positivity.
[0089] Step S302, adjusting the first repayment reminder strategy according to the repayment aggressiveness parameter.
[0090] The device adjusts the first repayment reminder strategy when determining the repayment positive parameter.
[0091] In one example, the first repayment reminder strategy has a corresponding upgrade bar. When the progress of the upgrade bar is filled with experience value, the first repayment reminder strategy is adjusted upward. The device subtracts the set value from the repayment enthusiasm parameter to obtain a difference, and the difference is converted into experience value. When the difference is positive, the converted experience value is positive, and when the difference is negative, the converted experience value is negative. The experience value in the upgrade bar plus the converted experience value can obtain the latest upgrade bar level, and the repayment reminder strategy associated with the upgrade bar level is the second repayment reminder strategy.
[0092] In another example, the device determines the first policy level corresponding to the first repayment reminder strategy, and the first policy level refers to the level of the first repayment reminder strategy. The device determines whether the repayment positive parameter is greater than the preset parameter. If the repayment positive parameter is greater than the preset parameter, it is determined that the first user is actively repaying, and the device lowers the first policy level to obtain the second policy level, and uses the repayment reminder strategy corresponding to the second policy level as the second repayment reminder strategy. When the repayment positive parameter is less than or equal to the preset parameter, it can be determined that the first user is passively repaying, and the first policy level is raised to obtain the third policy level, and the repayment reminder test corresponding to the third policy level is used as the second repayment reminder strategy.
[0093] In this embodiment, the device determines the repayment enthusiasm parameters of the first user based on the historical repayment behavior information and the historical repayment response information, and adjusts the first repayment reminder strategy according to the repayment enthusiasm parameters, that is, reasonably adjusts the first repayment reminder strategy according to the user's repayment enthusiasm.
[0094] Reference Figure 4 , Figure 4 This is a flow chart of the third embodiment of the message push method of the present application. Based on the first or second embodiment, step S202 includes:
[0095] Step S401, obtaining a plurality of first initial features according to first target information.
[0096] In this embodiment, the device first performs data cleaning on the first target information.
[0097] Exemplarily, the device processes missing values in the first target information, removes outliers, and processes duplicate records. For example, for missing values, an interpolation method can be used, such as linear interpolation or filling by the average value of historical data; the detection and processing of outliers can be achieved by calculating the interquartile range, thereby identifying and removing values that fall outside the normal data range.
[0098] Data cleaning also includes data preprocessing, which includes standardization and encoding conversion. All numerical data are standardized using the Z-score method to ensure that data of different magnitudes can be compared on the same scale. Categorical data, such as the range of overdue days, are converted into numerical format through one-hot encoding to facilitate algorithm processing.
[0099] By cleaning the data, the quality and consistency of the data are ensured, providing an accurate and reliable data basis for the accurate prediction of the subsequent evaluation model.
[0100] The device extracts the sub-data in the first target information to obtain multiple original features, each of which is defined as a first initial feature. The first initial feature is, for example, a loan-to-deposit ratio, a user's asset-loan coverage, and the like.
[0101] Step S402: convert and derive the first initial feature to obtain a plurality of first target features.
[0102] The device converts and derives the first initial feature to obtain multiple first target features. Conversion refers to converting non-numerical features into numerical features using unique hot encoding, and deriving, for example, calculating ratios, differences, and trend indicators. In addition, the growth rate of the number of approval inquiries for replacement resources in the past 12 months can be calculated, or the interactive features of the maximum loan-to-deposit ratio in the past 6 months and the deposit balance of the previous month can be constructed. These derived features can reveal the deep-level patterns of user repayment behavior and are of great value for the prediction of the evaluation model.
[0103] Furthermore, the device converts and derives the first initial feature to obtain the intermediate feature, and the device selects the intermediate features to eliminate those features that are unimportant or redundant for model prediction, and the remaining features are the first target features. Feature selection can not only reduce the complexity of the model, but also avoid overfitting and improve the generalization ability of the model. Model-based feature selection methods can be used, such as using a random forest or gradient boosting tree model with feature importance evaluation to identify and retain those features that contribute most to the prediction, that is, using a random forest or gradient boosting tree model to identify the importance parameters of the intermediate features, and the intermediate features with importance parameters greater than the threshold are used as the first target features.
[0104] Step S403: input each first target feature into the evaluation model to obtain the repayment risk level of the first user output by the evaluation model.
[0105] After obtaining each first target feature, each first target feature is input into the evaluation model to obtain the repayment risk level of the first user output by the evaluation model.
[0106] In this embodiment, the device converts and derives the initial features so that the obtained features can provide more dimensional information for the evaluation model, thereby making the repayment risk level output by the evaluation model more accurate.
[0107] Reference Figure 5 , Figure 5 This is a flow chart of the fourth embodiment of the message push method of the present application, based on any one of the first to third embodiments, before step S202, it also includes:
[0108] Step S501, obtaining various training samples, the training samples including second target information and labels, the second target information including the credit information, loan information, repayment information and transaction information of the second user, and the label is used to indicate the repayment risk level of the replacement resource applied for by the second user.
[0109] In this embodiment, the device obtains each training sample, and the training sample includes second target information and a label. The second target information includes the credit information, loan information, repayment information, and transaction information of the second user. The label is used to indicate the repayment risk level of the replacement resource applied by the second user.
[0110] In one example, the device obtains multiple second initial features based on the second target information of the second user. The second initial features are obtained in the same manner as the first initial features, which will not be described in detail. The device obtains the label set by the second user, thereby constructing a training sample through the label and the second initial feature.
[0111] In another example, after obtaining multiple second initial features, the second initial features are transformed and derived to obtain multiple second target features, and labels are set for each second target feature, so as to construct training samples based on each second target feature and label. The method for obtaining the second target feature is the same as the method for obtaining the first target feature, and will not be repeated here.
[0112] Step S502: training the preset model according to each training sample to obtain an evaluation model.
[0113] After obtaining each training sample, the preset model is trained based on each training sample to obtain the evaluation model.
[0114] Specifically, each training sample is divided into a training set and a test set. Usually 70% of the data is used as the training set and 30% of the data is used as the test set. In this way, the model can be built on the training set and the performance of the model can be evaluated on the test set.
[0115] Use the logistic regression algorithm to fit the training data. In this process, the model will learn the relationship between each feature and the probability of customer default. In order to optimize the performance of the model, the model's hyperparameters, such as regularization strength, can be adjusted through cross-validation.
[0116] The prediction performance of the model is evaluated on the test set, usually using indicators such as accuracy, AUC value, and recall rate. These indicators can help understand the generalization ability of the model on unseen data.
[0117] This application has the following advantages
[0118] Advantage 1: Comprehensive high-dimensional financial and behavioral data analysis
[0119] By comprehensively utilizing high-dimensional financial data and customer behavior data, such as the number of loan approval inquiries within the past twelve months, the maximum loan-to-deposit ratio in the past six months, and the transaction amount (inflow) of the enterprise in the past three months, complex features are constructed and analyzed. This approach provides deeper insights into understanding and predicting customer credit risk compared to existing technologies.
[0120] Advantage 2: Dynamic logistic regression model construction and update
[0121] The dynamically updated logistic regression model not only adjusts the model parameters according to the latest market data and customer behavior data, but also can flexibly adapt to market changes and the latest trends in customer behavior. This dynamic update mechanism is a significant improvement over the existing static risk assessment model.
[0122] Advantage 3: Implementation of personalized repayment reminder strategy
[0123] Based on the customer risk score and segmented group characteristics obtained through the dynamic logistic regression model, a personalized repayment reminder strategy is implemented. This strategy takes into account the customer's specific risk level and financial situation, and significantly improves the pertinence and efficiency of risk management compared to the generalized repayment reminder measures of existing technologies.
[0124] Advantage 4: Dynamic optimization of risk assessment and management
[0125] This application not only achieves innovation in the model building stage, but also realizes dynamic optimization of risk management strategies in the risk assessment and management process by continuously monitoring and analyzing market changes and customer behavior. This improvement enables more flexible and timely response to changes in credit risk.
[0126] These advantages together constitute the core innovation of this technical solution. They work together to significantly improve the accuracy and efficiency of users' risk assessment and management.
[0127] See also Figure 6 The message push device 600 provided in the embodiment of the present application includes:
[0128] A first acquisition module 610 is used to acquire first target information of a first user, where the first user is used to indicate a user who has applied for resource replacement, and the first target information includes credit information, loan information, repayment information, and transaction information of the first user;
[0129] An identification module 620 is used to identify the first target information using an evaluation model to obtain a repayment risk level of the first user, and determine a first repayment reminder strategy according to the repayment risk level;
[0130] The second acquisition module 630 is used to obtain the historical repayment behavior information and historical repayment response information of the first user for the replacement resource, and adjust the first repayment reminder strategy according to the historical repayment behavior information and the historical repayment response information to obtain a second repayment reminder strategy;
[0131] The generating module 640 is used to generate a prompt message according to the second repayment prompt strategy, and send the prompt message to the terminal device associated with the first user, where the prompt message is used to prompt the first user to repay the replacement resources.
[0132] In one embodiment, the message push device 600 is specifically used for:
[0133] Determining the repayment risk group to which the first user belongs according to the repayment risk level;
[0134] The repayment reminder strategy associated with the repayment risk group is determined as the first repayment reminder strategy.
[0135] In one embodiment, the message push device 600 is specifically used for:
[0136] Determine a repayment enthusiasm parameter of the first user according to the historical repayment behavior information and the historical repayment response information, where the repayment enthusiasm parameter is used to indicate the repayment enthusiasm of the first user;
[0137] Adjust the first repayment reminder strategy based on repayment aggressiveness parameters.
[0138] In one embodiment, the message push device 600 is specifically used for:
[0139] Determine a first strategy level corresponding to the first repayment reminder strategy;
[0140] In response to the repayment active parameter indicating that the first user actively repays, lowering the first policy level to obtain a second policy level, and determining the repayment reminder policy corresponding to the second policy level as the second repayment reminder policy;
[0141] In response to the repayment enthusiasm parameter indicating that the first user is passive in repayment, the first policy level is increased to obtain a third policy level, and the repayment reminder policy corresponding to the third policy level is determined as the second repayment reminder policy.
[0142] In one embodiment, the message push device 600 is specifically used for:
[0143] Acquire a plurality of first initial features according to the first target information;
[0144] Converting and deriving the first initial feature to obtain a plurality of first target features;
[0145] Each first target feature is input into the evaluation model to obtain the repayment risk level of the first user output by the evaluation model.
[0146] In one embodiment, the message push device 600 is specifically used for:
[0147] Acquire each training sample, the training sample includes second target information and a label, the second target information includes credit information, loan information, repayment information and transaction information of the second user, and the label is used to indicate the repayment risk level of the replacement resource applied by the second user;
[0148] The preset model is trained according to each training sample to obtain an evaluation model.
[0149] In one embodiment, the message push device 600 is specifically used for:
[0150] Acquire a plurality of second initial features according to second target information of a second user;
[0151] Converting and deriving the second initial feature to obtain a plurality of second target features;
[0152] A label is set for each second target feature, and a training sample corresponding to the second user is constructed according to the second target feature and the label.
[0153] Each module of the message push device provided in the embodiment of the present application can realize Figures 2 to 5 Any embodiment provides the functions of each step of the message push method and can achieve its corresponding technical effects. For the sake of concise description, they will not be repeated here.
[0154] Figure 7 A schematic diagram of the hardware structure of a message push device provided in an embodiment of the present application is shown.
[0155] The message push device may include a processor 701 and a memory 702 storing computer program instructions.
[0156] Specifically, the processor 701 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.
[0157] The memory 702 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 702 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 702 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 702 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 702 is a non-volatile solid-state memory.
[0158] In certain embodiments, the memory 702 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.
[0159] The processor 701 implements any one of the authentication methods in the above embodiments by reading and executing computer program instructions stored in the memory 702 .
[0160] In one example, the message push device may further include a communication interface 703 and a bus 710. Figure 7 As shown, the processor 701, the memory 702, and the communication interface 703 are connected via a bus 710 and communicate with each other.
[0161] The communication interface 703 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.
[0162] Bus 710 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 710 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.
[0163] 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.
[0164] 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.
[0165] 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.
[0166] 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 other programmable message push device to produce a machine so that these instructions executed by the processor of the computer or other programmable message push device 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.
[0167] 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 message push method, characterized in that: include: Acquire first target information of a first user, where the first user is used to indicate a user who has applied for resource replacement, and the first target information includes credit information, loan information, repayment information, and transaction information of the first user; Using an evaluation model to identify the first target information to obtain a repayment risk level of the first user, and determining a first repayment reminder strategy according to the repayment risk level; Acquire historical repayment behavior information and historical repayment response information of the first user for the replacement resource, and adjust the first repayment reminder strategy according to the historical repayment behavior information and the historical repayment response information to obtain a second repayment reminder strategy; A prompt message is generated according to the second repayment prompt strategy, and the prompt message is sent to a terminal device associated with the first user, wherein the prompt message is used to prompt the first user to repay the replaced resources.
2. The message push method according to claim 1, characterized in that: The step of determining the first repayment reminder strategy according to the repayment risk level comprises: determining the repayment risk group to which the first user belongs according to the repayment risk level; The repayment reminder strategy associated with the repayment risk group is determined as the first repayment reminder strategy.
3. The message push method according to claim 1, characterized in that: The step of adjusting the first repayment reminder strategy according to the historical repayment behavior information and the historical repayment response information includes: Determining a repayment enthusiasm parameter of the first user according to the historical repayment behavior information and the historical repayment response information, wherein the repayment enthusiasm parameter is used to indicate the repayment enthusiasm of the first user; The first repayment reminder strategy is adjusted according to the repayment positive parameter.
4. The message push method according to claim 3, characterized in that: The step of adjusting the first repayment reminder strategy according to the repayment positive parameter comprises: Determining a first strategy level corresponding to the first repayment reminder strategy; In response to the repayment active parameter indicating that the first user is actively repaying, lowering the first policy level to obtain a second policy level, and determining the repayment reminder policy corresponding to the second policy level as the second repayment reminder policy; In response to the repayment enthusiasm parameter indicating that the first user is passive in repayment, the first policy level is increased to obtain a third policy level, and the repayment reminder policy corresponding to the third policy level is determined as the second repayment reminder policy.
5. The message push method according to claim 1, characterized in that: The step of using the evaluation model to identify the first target information to obtain the repayment risk level of the first user includes: Acquire a plurality of first initial features according to the first target information; Converting and deriving the first initial features to obtain a plurality of first target features; Each of the first target features is input into the evaluation model to obtain the repayment risk level of the first user output by the evaluation model.
6. The message push method according to claim 1, characterized in that: Before the step of using the evaluation model to identify the first target information to obtain the repayment risk level of the first user, the method further includes: Acquire each training sample, the training sample including second target information and a label, the second target information including credit information, borrowing information, repayment information and transaction information of the second user, the label being used to indicate the repayment risk level of the 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.
7. The message push method according to claim 6, characterized in that: The step of obtaining each training sample comprises: Acquire a plurality of second initial features according to the second target information of the second user; Converting and deriving the second initial features to obtain a plurality of second target features; A label is set for each of the second target features, and a training sample corresponding to the second user is constructed according to the second target features and the label.
8. A message push device, characterized in that: include: A first acquisition module, used to acquire first target information of a first user, where the first user is used to indicate a user who has applied for resource replacement, and the first target information includes credit information, loan information, repayment information, and transaction information of the first user; an identification module, configured to identify the first target information using an evaluation model to obtain a repayment risk level of the first user, and determine a first repayment reminder strategy according to the repayment risk level; A second acquisition module is used to obtain the historical repayment behavior information and historical repayment response information of the first user for the replacement resources, and adjust the first repayment reminder strategy according to the historical repayment behavior information and the historical repayment response information to obtain a second repayment reminder strategy; A generation module is used to generate a prompt message according to the second repayment prompt strategy, and send the prompt message to the terminal device associated with the first user, wherein the prompt message is used to prompt the first user to repay the replacement resources.
9. A message push device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the message push method according to any one of claims 1 to 7 is implemented.
10. 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 message pushing method according to any one of claims 1 to 7 is implemented.
11. 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 message push method as described in any one of claims 1 to 7.