Data processing methods, apparatus, computer-readable storage media and electronic devices

Abstract

This application discloses a data processing method, apparatus, computer-readable storage medium, and electronic device. Relating to the field of artificial intelligence, the method includes: acquiring target data of a target user, wherein the target user is a user who has taken out a loan, and the target data includes at least the target user's loan data; performing predictive processing on the target data of the target user using a random forest model to obtain the target user's probability of overdue repayment, wherein the model parameters of the random forest model are determined based on the Black Terrier Optimization Algorithm; and determining target information based on the target user's probability of overdue repayment, wherein the target information is used to characterize whether the target user has a risk of overdue repayment. This application solves the problem in related technologies where the random forest model's model parameters are randomly determined when determining a user's overdue repayment risk, resulting in low accuracy in predicting the user's repayment risk.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more specifically, to a data processing method, apparatus, computer-readable storage medium, and electronic device. Background Technology

[0002] With economic development and fluctuations in the financial market, there are instances of loan delinquency, ranging from personal consumer loans to business operating loans. In order to control the risk of loan delinquency, it is necessary to predict the probability of loan delinquency for users.

[0003] Currently, when predicting the risk of loan delinquency based on random forest models in related technologies, the model parameters of the random forest model are usually determined randomly. This results in low efficiency in obtaining the optimal solution of the model parameters and a large degree of randomness, which makes it impossible to efficiently solve the problem of predicting bank loan delinquency and thus affects the accuracy of predicting bank loan delinquency for users.

[0004] There is currently no effective solution to the problem that when determining a user's overdue repayment risk based on the random forest model, the model parameters of the random forest model are randomly determined, resulting in low accuracy in predicting the user's repayment risk. Summary of the Invention

[0005] The main objective of this application is to provide a data processing method, apparatus, computer-readable storage medium, and electronic device to solve the problem in related technologies where the random forest model is randomly determined when determining a user's overdue repayment risk, resulting in low accuracy in predicting the user's repayment risk.

[0006] According to one aspect of the present invention, a data processing method is provided, comprising: acquiring target data of a target user, wherein the target user is a user who has taken out a loan, and the target data includes at least the target user's loan data; performing predictive processing on the target data of the target user using a random forest model to obtain the target user's overdue repayment probability, wherein the model parameters of the random forest model are determined based on the Black Terrier Optimization Algorithm; and determining target information based on the target user's overdue repayment probability, wherein the target information is used to characterize whether the target user has an overdue repayment risk.

[0007] Furthermore, the data processing method also includes: obtaining a sample set, wherein the training samples in the sample set are sample data of sample users, and the sample data includes at least sample loan data of sample users, and the true labels of the training samples characterize whether the sample users have the risk of overdue repayment; determining the model parameters of the random forest model to be generated based on the Black Terrier Optimization Algorithm; and training the random forest model based on the model parameters and the sample set.

[0008] Furthermore, the data processing method also includes: initializing the tern population in the tern optimization algorithm, wherein the tern population includes multiple tern positions, and each tern position is a set of model parameters corresponding to the random forest model; in the tern optimization algorithm, the mean squared error of the sample set is used as the fitness function, and the target position of each tern in the tern population is determined based on the fitness function to update the tern population; if the iteration of the tern population does not meet the preset iteration conditions, the step of determining the target position of each tern in the tern population based on the fitness function is repeated until the iteration meets the preset iteration conditions, and the optimal tern is determined from the tern population according to the fitness of each tern in the tern population, and the position of the optimal tern is determined as the model parameters used by the random forest model.

[0009] Furthermore, the data processing method also includes: for each tern, calculating the fitness of the tern based on the fitness function; determining the current best tern in the tern population based on the fitness of each tern; and for each tern, determining the target position of the tern in the tern population based on the tern's position and the position of the current best tern.

[0010] Furthermore, the data processing method also includes: determining the model parameters of the candidate random forest model to be generated based on the location of the black tern; training the candidate random forest model based on the model parameters and the sample set; and determining the fitness of the black tern by the average mean square error of the sample set processed by the candidate random forest model.

[0011] Furthermore, the data processing method also includes identifying the tern with the lowest fitness in the tern population as the optimal tern in the tern population.

[0012] Furthermore, the data processing method also includes: when the probability of a target user's overdue repayment is greater than a preset value, determining that the target information indicates that the target user has a risk of overdue repayment; when the probability of a target user's overdue repayment is less than or equal to a preset value, determining that the target information indicates that the target user does not have a risk of overdue repayment.

[0013] To achieve the above objectives, according to another aspect of this application, a data processing apparatus is provided. The apparatus includes: a first acquisition module for acquiring target data of a target user, wherein the target user is a user who has taken out a loan, and the target data includes at least the target user's loan data; a processing module for performing predictive processing on the target data of the target user using a random forest model to obtain the target user's probability of overdue repayment, wherein the model parameters of the random forest model are determined based on the Black-tailed Gull optimization algorithm; and a first determination module for determining target information based on the target user's probability of overdue repayment, wherein the target information is used to characterize whether the target user has a risk of overdue repayment.

[0014] Furthermore, the data processing device also includes: a second acquisition module for acquiring a sample set, wherein the training samples in the sample set are sample data of sample users, the sample data includes at least sample loan data of sample users, and the true labels of the training samples characterize whether the sample users have the risk of overdue repayment; a second determination module for determining the model parameters of the random forest model to be generated based on the Black-tailed Gull optimization algorithm; and a training module for training the random forest model based on the model parameters and the sample set.

[0015] Furthermore, the second determining module also includes: an initialization submodule, used to initialize the tern population in the tern optimization algorithm, wherein the tern population includes multiple tern positions, and each tern position is a set of model parameters corresponding to the random forest model; a first determining submodule, used in the tern optimization algorithm to use the mean squared error of the sample set as the fitness function, and to determine the target position of each tern in the tern population based on the fitness function, so as to update the tern population; and a second determining submodule, used to repeatedly execute the step of determining the target position of each tern in the tern population based on the fitness function when the iteration of the tern population does not meet the preset iteration conditions, until the iteration meets the preset iteration conditions, and then determine the optimal tern from the tern population based on the fitness of each tern in the tern population, and determine the position of the optimal tern as the model parameters adopted by the random forest model.

[0016] Furthermore, the first determining submodule also includes: a calculation unit, used to calculate the fitness of each tern based on a fitness function; a first determining unit, used to determine the current optimal tern from the tern population based on the fitness of each tern; and a second determining unit, used to determine the target position of each tern in the tern population based on the tern's position and the position of the current optimal tern.

[0017] Furthermore, the computing unit also includes: a first determining subunit, used to determine the model parameters of the candidate random forest model to be generated based on the location of the black tern; a training subunit, used to train the candidate random forest model based on the model parameters of the candidate random forest model and the sample set; and a second determining subunit, used to determine the fitness of the black tern by the average mean square error of the sample set processed by the candidate random forest model.

[0018] Furthermore, the second determining submodule also includes a third determining unit, used to determine the tern with the lowest fitness in the tern population as the optimal tern in the tern population.

[0019] Furthermore, the first determining module also includes: a third determining submodule, used to determine that the target information indicates that the target user has a risk of defaulting on repayment when the probability of defaulting on repayment by the target user is greater than a preset value; and a fourth determining submodule, used to determine that the target information indicates that the target user does not have a risk of defaulting on repayment when the probability of defaulting on repayment by the target user is less than or equal to a preset value.

[0020] To achieve the above objectives, according to another aspect of this application, a computer-readable storage medium is provided. The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device where the computer-readable storage medium is located to perform the aforementioned data processing method.

[0021] To achieve the above objectives, according to another aspect of this application, an electronic device is provided. The electronic device includes: a memory storing an executable program; and a processor for running the program, wherein the program executes the data processing method described above during runtime.

[0022] To achieve the above objectives, according to another aspect of this application, a computer program product is provided, which includes computer instructions that, when executed by a processor, implement the steps of the data processing method described above.

[0023] In this embodiment, the model parameters of a random forest model are determined using the Black-tailed Tern optimization algorithm, and the model is then used to predict the risk of user delinquency. By using the Black-tailed Tern optimization algorithm to determine the model parameters, accurate determination of the random forest model's parameters is achieved, avoiding the instability caused by randomly determining the model parameters, which would affect the model's training performance and thus its prediction accuracy. By predicting the target user's target data using a random forest model with optimized model parameters, the probability of delinquency for the target user is obtained. This achieves the goal of predicting user delinquency risk based on a random forest model with better model parameters, thereby improving the accuracy of the prediction.

[0024] Therefore, the solution provided in this application achieves the goal of determining the model parameters of the random forest model based on the tern optimization algorithm and predicting the user's overdue repayment risk based on the random forest model, thereby improving the technical effect of prediction accuracy. It also solves the technical problem in related technologies where the model parameters of the random forest model are randomly determined when determining the user's overdue repayment risk, resulting in low accuracy of predicting the user's repayment risk. Attached Figure Description

[0025] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0026] Figure 1 This is a hardware structure block diagram of a computer terminal provided according to an embodiment of this application;

[0027] Figure 2 This is a schematic diagram of a data processing method provided according to an embodiment of this application;

[0028] Figure 3 This is a flowchart illustrating the determination of model parameters for a random forest model according to embodiments of this application;

[0029] Figure 4 This is a schematic diagram of a data processing apparatus provided according to an embodiment of this application;

[0030] Figure 5 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0031] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0032] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0033] It should be noted that the information collected in this application (including but not limited to user device information, user personal information, etc.) and data (including but not limited to user loan data, data used for display, data for analysis, etc.) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding access points are provided for users to choose to authorize or refuse. For example, interfaces are set up between this system and relevant users or institutions, providing users with corresponding access points to choose to agree to or refuse automated decision-making results; if the user chooses to refuse, the process proceeds to the expert decision-making stage.

[0034] Example 1

[0035] According to an embodiment of this application, an embodiment of a data processing method is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0036] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing a data processing method is shown. Figure 1As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0037] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0038] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the data processing method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby implementing the aforementioned data processing method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0039] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0040] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0041] Under the aforementioned operating environment, this application provides the following: Figure 2 The data processing method shown. Figure 2 This is a schematic diagram of a data processing method provided according to an embodiment of this application.

[0042] Step S201: Obtain target data of the target user, wherein the target user is a user who has taken out a loan, and the target data includes at least the loan data of the target user.

[0043] Optionally, electronic devices, application systems, servers, and other devices can be used as the execution subject of this application. In this embodiment, a data processing system is used as the execution subject to execute the above-described data processing method.

[0044] Optionally, the target user's target data may include the target user's loan data. Loan data may include the user's loan amount, loan term, etc.

[0045] Optionally, the target user's target data may also include the target user's personal information, asset information, historical repayment information, and credit information, which serve as the basis for subsequent risk prediction.

[0046] Step S202: The target data of the target user is predicted and processed using a random forest model to obtain the overdue repayment probability of the target user. The model parameters of the random forest model are determined based on the Black Terrier Optimization Algorithm.

[0047] Optionally, a random forest model is a learning algorithm consisting of multiple decision trees. Model parameters for a random forest model may include the number of trees, the minimum number of leaf nodes, the maximum depth of each tree, and the feature selection ratio.

[0048] Optionally, the data processing system can determine the model parameters of the random forest model based on the tern optimization algorithm, and then train the random forest model based on the determined model parameters and the pre-acquired sample set, thereby using the trained random forest model to predict and process the target data of the target user.

[0049] Step S203: Determine target information based on the target user's probability of defaulting on repayments, wherein the target information is used to characterize whether the target user has a risk of defaulting on repayments.

[0050] Optionally, in the process of determining whether a user has a risk of defaulting on payments based on the predicted probability of default, a threshold can be set, such as 0.6. If the predicted probability (e.g., 0.67) is higher than the set threshold, the user is considered to have a risk of defaulting on payments and may default on payments. If the predicted probability (e.g., 0.23) is lower than the set threshold, the user is considered not to have a risk of defaulting on payments and is unlikely to default on payments.

[0051] In this embodiment, the model parameters of the random forest model are determined based on the Black Terns Optimization Algorithm, and the risk of user delinquency is predicted based on the random forest model. By determining the model parameters of the random forest model based on the Black Terns Optimization Algorithm, the model parameters of the random forest model are accurately determined, avoiding the instability of the training effect of the random forest model caused by randomly determining the model parameters, which would affect the prediction effect of the random forest model. By predicting the target data of the target user based on the random forest model with optimized model parameters, the probability of delinquency of the target user is obtained. This realizes the prediction of user delinquency risk based on the random forest model with better model parameters, thereby improving the accuracy of prediction.

[0052] Therefore, the solution provided in this application achieves the goal of determining the model parameters of the random forest model based on the tern optimization algorithm and predicting the user's overdue repayment risk based on the random forest model, thereby improving the technical effect of prediction accuracy. It also solves the technical problem in related technologies where the model parameters of the random forest model are randomly determined when determining the user's overdue repayment risk, resulting in low accuracy of predicting the user's repayment risk.

[0053] In one optional embodiment, the data processing system can determine the random forest model by: acquiring a sample set, wherein the training samples in the sample set are sample data of sample users, the sample data includes at least sample loan data of sample users, and the true labels of the training samples characterize whether the sample users have the risk of overdue repayment; determining the model parameters of the random forest model to be generated based on the Black Terrier Optimization Algorithm; and training the random forest model based on the model parameters and the sample set.

[0054] Optionally, the training samples in the sample set are sample data of sample users. The sample data includes at least sample loan data of sample users, and may also include sample personal information, sample asset information, sample historical repayment information, and sample credit information of sample users.

[0055] Optionally, the true label of the training sample represents whether the sample user has a risk of defaulting on repayment, that is, whether the sample user has defaulted on repayment.

[0056] Optionally, after obtaining the sample set and determining the model parameters of the random forest model, the sample set can be divided into a training set and a test set in a certain ratio (e.g., 8:2). The training set is used as the input data for training the random forest model to obtain the model. The test set can be used to verify the training effect of the random forest model, and to retrain the random forest model if the training effect does not meet the preset conditions.

[0057] It should be noted that by training the random forest model with user data bearing real labels, the random forest model can learn the correlation between user data and the risk of overdue repayment; by training with optimized parameters, the random forest model can better fit the data in the sample, thereby improving the accuracy of predicting the probability of loan users' overdue repayment.

[0058] In one optional embodiment, during the process of determining the model parameters of the random forest model to be generated based on the tern optimization algorithm, the data processing system can initialize the tern population in the tern optimization algorithm, wherein the tern population includes multiple tern positions, and each tern position is a set of model parameters corresponding to the random forest model; in the tern optimization algorithm, the mean squared error of the sample set is used as the fitness function, and the target position of each tern in the tern population is determined based on the fitness function to update the tern population; if the iteration of the tern population does not meet the preset iteration conditions, the step of determining the target position of each tern in the tern population based on the fitness function is repeated until the iteration meets the preset iteration conditions, and the optimal tern is determined from the tern population according to the fitness of each tern in the tern population, and the position of the optimal tern is determined as the model parameters adopted by the random forest model.

[0059] Optionally, in the tern optimization algorithm, the position of each individual in the tern population represents an initial solution, that is, a set of model parameters corresponding to the random forest model. These parameters typically include key parameters of the random forest, such as the number of trees, the minimum number of leaf nodes, the maximum depth of each tree, and the feature selection ratio. The initialized tern population X can be represented as:

[0060]

[0061] In the formula, X i X represents the current position of the i-th black tern; i,j is the current position of the i-th tern in the j-th dimension; N is the population size of the tern population; m is the dimension of the problem being solved, which is the number of model parameters used in a random forest model.

[0062] Optionally, the tern optimization algorithm mainly includes two stages in one iteration: a global exploration stage (the migratory behavior of terns) and a local exploration stage (the aggressive behavior of terns). The global exploration stage further includes a conflict avoidance stage, a clustering stage, and an update stage. In these two stages, the position of the tern calculated in the local exploration stage is determined as the target position.

[0063] Optionally, in determining the target position of each Black-eared Terns within the Black-eared Tern population based on the fitness function, the data processing system can calculate the fitness of each Black-eared Tern at its current position according to the fitness function, and select the Black-eared Tern with the lowest fitness as the optimal Black-eared Tern in the current iteration. The position of the optimal Black-eared Tern is set as the direction of movement for other Black-eared Terns in the population, thereby determining the target position of each Black-eared Tern within the population based on the position of the optimal Black-eared Tern and the current position of each Black-eared Tern, and updating the Black-eared Tern population to reflect these changes.

[0064] In some embodiments, the mean squared error (MSE) of the sample set is used as the fitness function in the tern optimization algorithm to evaluate the performance of each tern (i.e., each set of model parameters). For example, a candidate random forest model is trained using the model parameters corresponding to the position of a tern (100 trees, 55 minimum leaf nodes, 25 maximum depths per tree, and a feature selection ratio of M / 3, where M is the number of data types in the user's sample data). The mean squared error of this model is calculated on both the training and test sets, and then averaged to obtain the average mean squared error for the sample set. If this average mean squared error is small, the combination is considered superior; if the average mean squared error is large, the tern optimization algorithm will adjust the tern's position to be closer to a position with a smaller mean squared error.

[0065] In some embodiments, the preset iteration condition for the Black-eared Terns population can be reaching the maximum number of iterations.

[0066] Optionally, in each iteration, the algorithm evaluates the positions of all terns based on the fitness function and updates their positions accordingly. This process continues until a preset iteration condition is met. Once the iteration ends, the algorithm selects the tern with the lowest fitness (i.e., the lowest mean squared error) from the current tern population; its position is the optimal parameter for the final random forest model.

[0067] For example, Figure 3 This is a flowchart illustrating the determination of model parameters for a random forest model according to embodiments of this application, such as... Figure 3 As shown, before starting the Black-eared Terns optimization algorithm, the algorithm parameters are set, including the number of iterations, population size, and parameter optimization boundary. An initial Black-eared Terns population is randomly generated, with each individual representing a set of model parameters for a random forest model. After starting the Black-eared Terns algorithm, fitness is calculated, and the target position of each Black-eared Tern is calculated based on the fitness to update the population. This process includes global exploration and local exploration. Global exploration simulates the migratory behavior of Black-eared Terns; local exploration simulates their attack behavior, performing a local search for each individual to refine the search for the optimal solution. After obtaining the current optimal solution, the position of each individual is updated, and the global optimal solution is recorded. This ensures that even if the preset iteration conditions (i.e., ...) are not met during iteration... Figure 3 The population of Black-eared Terns is continuously updated if the maximum number of iterations has not been reached, and the preset iteration conditions are met (i.e., ... Figure 3 If the maximum number of iterations is reached, the optimal position of the black tern is determined as the optimal model parameter for the random forest model. Then, the random forest model using these parameters is used to predict loan delinquency risk. Upon receiving a user's loan request, the data processing system can predict the target user's delinquency probability based on the target user's target data, and determine whether to issue a loan based on the determined delinquency probability.

[0068] It should be noted that the above optimization process of model parameters can simulate the natural search behavior of terns. By initializing the tern population, evaluating fitness based on the mean square error, and iteratively optimizing the process, the location of the tern with the best fitness is finally selected as the optimal parameter configuration of the random forest model. This can effectively improve the predictive ability of the random forest model, thereby improving the accuracy of determining the risk of users defaulting on payments.

[0069] In an optional embodiment, in the process of determining the target position of each tern in the tern population based on the fitness function, the data processing system can calculate the fitness of each tern based on the fitness function; determine the current optimal tern from the tern population based on the fitness of each tern; and determine the target position of each tern in the tern population based on its position and the position of the current optimal tern.

[0070] Optionally, in one iteration, the fitness of each tern can be determined based on the average mean square error of the sample set, and the tern with the lowest fitness is selected as the current optimal tern. Then, the position of the optimal tern is set as the direction of movement for other tern individuals in the tern population, thereby determining the target position of the tern in the tern population based on the tern's position and the position of the current optimal tern.

[0071] Optionally, in determining the target location of each tern within the tern population based on the fitness function, the data processing system can perform the following steps:

[0072] Step 1: Global exploration (migration behavior), mainly divided into three parts: conflict avoidance, clustering, and updating.

[0073] Step 1.1: Conflict Avoidance;

[0074] During the conflict avoidance phase, determine the position (C) of each tern in the ternary population that should be in order to avoid colliding with other terns. st This means ensuring that the positions of each black tern do not conflict with each other, and its mathematical expression is...

[0075] The formula is as follows:

[0076] C st =S A *P st (Z)

[0077] Where Z represents the current iteration number, P st (Z) represents the current position of the Black-tailed Terrier in the Zth iteration round, S A The variable representing collision avoidance is used to calculate the position of each black tern after avoiding a collision. The formula corresponding to its constraints is as follows:

[0078]

[0079] Among them, C f It is used to adjust S A The control variable, therefore S A From C f The iterations decrease linearly to 0, where Z represents the current iteration number. Assume C...f If the value of S is 2, then S A It will gradually decrease from 2 to 0, Max iterations This represents the maximum number of iterations.

[0080] Step 1.2: Gathering;

[0081] During the aggregation phase, each tern is guaranteed to move towards the best position among its neighbors, i.e., towards the optimal solution position, while avoiding conflict. The mathematical expression for this is as follows:

[0082] m st =C B *(P bst (Z)-P st (Z))

[0083] Where, m st Indicates the current position P of the Black Terrier. st (Z) Move to the current optimal position P of the tern bst (Z) movement

[0084] Cheng, C B This is a random variable that makes the global exploration more comprehensive, and its mathematical expression is as follows:

[0085] C B =0.5*R rand

[0086] Among them, R rand It is a random number between 0 and 1.

[0087] Step 1.3: Update;

[0088] During the update phase, the updated trajectory d of each Black Terrier is calculated as it moves toward the position of the optimal Black Terrier. st The mathematical expression is:

[0089] d st =C st +m st

[0090] Step 2: Local exploration (attack behavior);

[0091] During migration (i.e., the global exploration phase), terns increase their flight altitude by flapping their wings and adjust their speed and attack angle. When attacking prey (i.e., the local exploration phase), they circle continuously in the air to find and capture prey. The mathematical model of this behavior is as follows:

[0092] x ′ =R*sin i

[0093] y ′=R*cos i

[0094] z ′ =R*i

[0095] R = ue kv

[0096] Among them, (x ′ y ′ , z ′ ) represents the coordinates of the position of the tern in space, R represents the radius of each spiral formed by the tern during its circling process, i represents an angle variable with a value range of [0, 2π], u and v are constants that define the shape of the spiral, which are generally preset to 1, and e is the basis of the natural logarithm.

[0097] Ultimately, the location of the Black-tailed Tern will be continuously updated based on the above variables, using the following update formula:

[0098] P st (Z) ′ =(d st *(x ′ +y ′ +z ′ ))*P bst (Z)

[0099] Among them, P st (Z) ′ This indicates the target location of the black tern.

[0100] It should be noted that by using the position update strategy of the black tern, the algorithm moves each black tern closer to the current optimal position in each iteration, which improves the optimization efficiency of the black tern algorithm, balances global search and local refinement, facilitates the solution of better parameter configuration of the random forest model, and better realizes the adjustment of model parameters, thereby effectively improving the prediction ability of the random forest model.

[0101] In an optional embodiment, during the process of calculating the fitness of the black tern based on the fitness function, the data processing system can determine the model parameters of the candidate random forest model to be generated based on the location of the black tern; train the candidate random forest model based on the model parameters of the candidate random forest model and the sample set; and determine the fitness of the black tern by the average mean square error of the sample set processed by the candidate random forest model.

[0102] Optionally, the location of each black tern corresponds to a set of parameters for a random forest model. Based on the location of the black tern, the specific parameters of the candidate random forest model can be determined, such as the number of trees, the minimum number of leaf nodes, the maximum depth of each tree, and the feature selection ratio.

[0103] Optionally, during the process of training a candidate random forest model based on the determined candidate model parameters and sample set, the random forest model constructs multiple decision trees based on the data in the sample set, enabling it to learn the relationship between sample features and true labels.

[0104] Optionally, the fitness of each black tern is the average of the mean squared errors of the candidate random forest model on the test and training sets. The formula for calculating the mean squared error is as follows:

[0105]

[0106] Where n is the number of samples in the sample set, y i z is the actual value of the i-th sample. i This is the predicted value for the i-th sample. The lower the mean squared error, the higher the prediction accuracy and the better the fitness of the model.

[0107] It should be noted that the Black Terrier Optimization Algorithm determines the fitness of each Black Terrier by averaging the mean squared error of the candidate random forest models generated for each Black Terrier on the sample set. This accurately assesses the quality of each random forest model's parameters, ensuring the optimization process moves towards reducing the mean squared error. Consequently, the random forest model becomes more accurate in predicting user default repayment risk, effectively improving loan risk management capabilities.

[0108] In an optional embodiment, during the process of determining the optimal tern from the tern population based on the fitness of each tern in the tern population, the data processing system can identify the tern with the lowest fitness in the tern population as the optimal tern in the tern population.

[0109] Optionally, the fitness of each tern is the mean squared error, and the smaller this value, the higher the prediction accuracy of the model. Therefore, the optimal tern in the tern population refers to the tern with the lowest fitness, i.e., the lowest mean squared error, which is also the tern with the best performance for the current random forest model parameter configuration.

[0110] It should be noted that the Black Terns optimization algorithm identifies the Black Terns with the lowest fitness as the optimal Black Terns, thus using the model parameters with the best corresponding model performance as the final model parameters to be used. This allows the random forest model using these model parameters to achieve higher prediction accuracy in predicting overdue repayment risks, effectively improving the control of user loan risks.

[0111] In one optional embodiment, during the process of determining target information based on the target user's overdue repayment probability, the data processing system can determine that the target information indicates that the target user has an overdue repayment risk if the target user's overdue repayment probability is greater than a preset value; and determine that the target information indicates that the target user does not have an overdue repayment risk if the target user's overdue repayment probability is less than or equal to the preset value.

[0112] Optionally, after predicting the probability of a target user's overdue repayment based on the target data, a threshold is set to classify the user's risk: when the predicted probability is greater than the threshold, it indicates that the target user has a risk of overdue repayment; when the predicted probability is less than the threshold, it indicates that the target user does not have a risk of overdue repayment.

[0113] In one optional embodiment, assuming a preset risk threshold of 0.6, the predicted overdue repayment probability for target user 1 using a random forest model with optimized model parameters is 0.65, for target user 2 it is 0.22, and for target user 3 it is 0.9. Therefore, it can be determined that the predicted overdue repayment risk probabilities for target users 1 and 3 are both greater than the preset threshold, indicating a risk of overdue repayment, while the predicted overdue repayment risk probability for target user 2 is less than the preset threshold, indicating no risk of overdue repayment.

[0114] It should be noted that by comparing the probability of overdue repayment with a preset threshold to determine a user's risk status, potential users with overdue repayment risk can be identified more efficiently. This helps banks conduct targeted risk management after lending, thereby reducing the probability of overdue risk and ensuring credit quality.

[0115] Therefore, the solution provided in this application achieves the goal of determining the model parameters of the random forest model based on the tern optimization algorithm and predicting the user's overdue repayment risk based on the random forest model, thereby improving the technical effect of prediction accuracy. It also solves the technical problem in related technologies where the model parameters of the random forest model are randomly determined when determining the user's overdue repayment risk, resulting in low accuracy of predicting the user's repayment risk.

[0116] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0117] Example 2

[0118] This application also provides a data processing apparatus. It should be noted that the data processing apparatus of this application can be used to execute the data processing method provided in this application. The data processing apparatus provided in this application will be described below.

[0119] According to an embodiment of this application, an apparatus for implementing the above-described data processing method is also provided, such as... Figure 4 As shown, the device includes:

[0120] The first acquisition module 401 is used to acquire target data of the target user, wherein the target user is a user who has taken out a loan, and the target data includes at least the loan data of the target user.

[0121] The processing module 402 is used to predict the target data of the target user through a random forest model to obtain the overdue repayment probability of the target user. The model parameters of the random forest model are determined based on the Black Terrier Optimization Algorithm.

[0122] The first determining module 403 is used to determine target information based on the probability of overdue repayment of the target user, wherein the target information is used to characterize whether the target user has the risk of overdue repayment.

[0123] It should be noted that the first acquisition module 401, processing module 402 and first determination module 403 mentioned above correspond to steps S201 to S203 in the above embodiments. The examples and application scenarios implemented by the three modules and the corresponding steps are the same, but are not limited to the content disclosed in the above embodiment 1.

[0124] In this embodiment, the model parameters of the random forest model are determined using the Black-tailed Tern optimization algorithm, and the risk of user delinquency is predicted based on this random forest model. This method, using the Black-tailed Tern optimization algorithm to determine the model parameters, achieves accurate determination of the random forest model's parameters, avoiding the instability of training performance caused by randomly determining the model parameters, which would negatively impact the prediction performance. By predicting the target user's target data using the random forest model with optimized model parameters, the probability of delinquency for the target user is obtained. This achieves the prediction of user delinquency risk based on a random forest model with better model parameters, thereby improving the accuracy of the prediction.

[0125] Therefore, the solution provided in this application achieves the goal of predicting user overdue repayment risk using a random forest model with optimal model parameters obtained from the Black-tailed Gull optimization algorithm, thereby improving the technical effect of prediction accuracy. This solves the technical problem in related technologies where the random forest model is randomly determined when determining user overdue repayment risk, resulting in low accuracy in predicting user repayment risk.

[0126] Optionally, in the data processing apparatus provided in this application embodiment, the data processing apparatus further includes: a second acquisition module, used to acquire a sample set, wherein the training samples in the sample set are sample data of sample users, the sample data includes at least sample loan data of sample users, and the real labels of the training samples characterize whether the sample users have the risk of overdue repayment; a second determination module, used to determine the model parameters of the random forest model to be generated based on the Black-tailed Gull optimization algorithm; and a training module, used to train the random forest model based on the model parameters and the sample set.

[0127] Optionally, in the data processing apparatus provided in this application embodiment, the second determining module further includes: an initialization submodule, used to initialize the tern population in the tern optimization algorithm, wherein the tern population includes multiple tern positions, and each tern position is a set of model parameters corresponding to the random forest model; a first determining submodule, used in the tern optimization algorithm to use the mean square error of the sample set as the fitness function, and to determine the target position of each tern in the tern population based on the fitness function, so as to update the tern population; and a second determining submodule, used to repeatedly execute the step of determining the target position of each tern in the tern population based on the fitness function when the iteration of the tern population does not meet the preset iteration conditions, until the iteration meets the preset iteration conditions, and then determine the optimal tern from the tern population based on the fitness of each tern in the tern population, and determine the position of the optimal tern as the model parameters adopted by the random forest model.

[0128] Optionally, in the data processing apparatus provided in the embodiments of this application, the first determining submodule further includes: a calculation unit, used to calculate the fitness of each tern based on a fitness function; a first determining unit, used to determine the current optimal tern from the tern population based on the fitness of each tern; and a second determining unit, used to determine the target position of each tern in the tern population based on the position of the tern and the position of the current optimal tern.

[0129] Optionally, in the data processing apparatus provided in this application embodiment, the computing unit further includes: a first determining subunit, used to determine the model parameters of the candidate random forest model to be generated based on the location of the black tern; a training subunit, used to train the candidate random forest model based on the model parameters of the candidate random forest model and the sample set; and a second determining subunit, used to determine the fitness of the black tern by the average mean square error of the sample set processed by the candidate random forest model.

[0130] Optionally, in the data processing apparatus provided in this application embodiment, the second determining submodule further includes: a third determining unit, used to determine the tern with the lowest fitness in the tern population as the optimal tern in the tern population.

[0131] Optionally, in the data processing apparatus provided in this application embodiment, the first determining module further includes: a third determining submodule, used to determine that the target information indicates that the target user has a risk of overdue repayment when the probability of overdue repayment of the target user is greater than a preset value; and a fourth determining submodule, used to determine that the target information indicates that the target user does not have a risk of overdue repayment when the probability of overdue repayment of the target user is less than or equal to a preset value.

[0132] It should be noted that the first acquisition module 401, processing module 402, and first determination module 403 mentioned above correspond to steps S201 to S203 in Embodiment 1. The three modules and their corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above modules can also be part of a device and run in the computer terminal 10 provided in Embodiment 1.

[0133] Example 3

[0134] Embodiments of this application may provide an electronic device. Figure 5 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 5 As shown, the electronic device may include: one or more ( Figure 5 (Only one is shown) processor 1002, memory 1004, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0135] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-described methods. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0136] The processor can access information and applications stored in memory via a transmission device to perform the following steps: acquiring target data of the target user, wherein the target user is a user who has taken out a loan, and the target data includes at least the target user's loan data; performing predictive processing on the target data of the target user using a random forest model to obtain the target user's probability of default, wherein the model parameters of the random forest model are determined based on the Black Terrier Optimization Algorithm; determining target information based on the target user's probability of default, wherein the target information is used to characterize whether the target user has a risk of default.

[0137] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: obtain a sample set, wherein the training samples in the sample set are sample data of sample users, the sample data includes at least sample loan data of sample users, and the real labels of the training samples characterize whether the sample users have the risk of overdue repayment; determine the model parameters of the random forest model to be generated based on the Black Terrier Optimization Algorithm; and train the random forest model based on the model parameters and the sample set.

[0138] The processor can also call the information and application program stored in the memory through the transmission device to perform the following steps: initializing the tern population in the tern optimization algorithm, wherein the tern population includes multiple tern positions, and each tern position is a set of model parameters corresponding to the random forest model; in the tern optimization algorithm, the mean square error of the sample set is used as the fitness function, and the target position of each tern in the tern population is determined based on the fitness function to update the tern population; if the iteration of the tern population does not meet the preset iteration conditions, the step of determining the target position of each tern in the tern population based on the fitness function is repeated until the iteration meets the preset iteration conditions, and the optimal tern is determined from the tern population according to the fitness of each tern in the tern population, and the position of the optimal tern is determined as the model parameters used by the random forest model.

[0139] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: for each tern, calculate the fitness of the tern based on the fitness function; determine the current best tern in the tern population based on the fitness of each tern; for each tern, determine the target position of the tern in the tern population based on the tern's position and the position of the current best tern.

[0140] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: determine the model parameters of the candidate random forest model to be generated based on the location of the black tern; train the candidate random forest model based on the model parameters and the sample set; and determine the fitness of the black tern by the average mean square error of the sample set processed by the candidate random forest model.

[0141] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: identify the tern with the lowest fitness in the tern population as the optimal tern in the tern population.

[0142] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: if the probability of the target user's overdue repayment is greater than a preset value, determine that the target information indicates that the target user has a risk of overdue repayment; if the probability of the target user's overdue repayment is less than or equal to the preset value, determine that the target information indicates that the target user does not have a risk of overdue repayment.

[0143] In this embodiment, the model parameters of the random forest model are determined using the Black-tailed Tern optimization algorithm, and the risk of user delinquency is predicted based on this random forest model. This method, using the Black-tailed Tern optimization algorithm to determine the model parameters, achieves accurate determination of the random forest model's parameters, avoiding the instability of training performance caused by randomly determining the model parameters, which would negatively impact the prediction performance. By predicting the target user's target data using the random forest model with optimized model parameters, the probability of delinquency for the target user is obtained. This achieves the prediction of user delinquency risk based on a random forest model with better model parameters, thereby improving the accuracy of the prediction.

[0144] Therefore, the solution provided in this application achieves the goal of predicting user overdue repayment risk using a random forest model with optimal model parameters obtained from the Black-tailed Gull optimization algorithm, thereby improving the technical effect of prediction accuracy. This solves the technical problem in related technologies where the random forest model is randomly determined when determining user overdue repayment risk, resulting in low accuracy in predicting user repayment risk.

[0145] Those skilled in the art will understand that Figure 5 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones (such as Android phones, iOS phones, etc.), tablets, PDAs, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 5 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 5 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 5 The different configurations shown.

[0146] A person skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0147] Example 4

[0148] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the data processing method provided in Embodiment 1.

[0149] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0150] This application also provides a computer program product, which, when executed on a data processing device, is a program adapted to perform data processing method steps.

[0151] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0152] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0153] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of units or modules may be electrical or other forms.

[0154] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0155] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0156] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0157] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A data processing method, characterized in that, include: Obtain target data of target users, wherein the target users are users who have taken out loans, and the target data includes at least the loan data of the target users; The target data of the target user is predicted and processed by a random forest model to obtain the probability of the target user's overdue repayment. The model parameters of the random forest model are determined based on the Black Terrier Optimization Algorithm. Target information is determined based on the probability of the target user's overdue repayment, wherein the target information is used to characterize whether the target user has a risk of overdue repayment; The random forest model is obtained in the following way: Obtain a sample set, wherein the training samples in the sample set are sample data of sample users, and the sample data includes at least the sample loan data of the sample users. The real labels of the training samples characterize whether the sample users have the risk of overdue repayment. The model parameters of the random forest model to be generated are determined based on the aforementioned tern optimization algorithm; The random forest model is trained based on the model parameters and the sample set. The model parameters of the random forest model to be generated are determined based on the aforementioned tern optimization algorithm, including: Initialize the black tern population in the black tern optimization algorithm, wherein the black tern population includes multiple black tern positions, and each black tern position is a set of model parameters corresponding to the random forest model; In the optimization algorithm for the black tern, the average mean square error of the sample set is used as the fitness function. Based on the fitness function, the target position of each black tern in the black tern population is determined to update the black tern population. If the iteration of the tern population does not meet the preset iteration conditions, the step of determining the target position of each tern in the tern population based on the fitness function is repeated until the iteration meets the preset iteration conditions. Then, the optimal tern is determined from the tern population based on the fitness of each tern in the tern population, and the position of the optimal tern is determined as the model parameter used by the random forest model.

2. The method according to claim 1, characterized in that, Determining the target location of each Black-eared Terrier within the Black-eared Terrier population based on the fitness function includes: For each tern, the fitness of the tern is calculated based on the fitness function. The optimal tern is determined from the tern population based on the fitness of each tern. For each tern, the target position of the tern in the tern population is determined based on the tern's position and the position of the current best tern.

3. The method according to claim 2, characterized in that, The fitness of the black tern is calculated based on the fitness function, including: The model parameters of the candidate random forest model to be generated are determined based on the location of the black tern; The candidate random forest model is obtained by training based on the model parameters of the candidate random forest model and the sample set; The fitness of the black tern is determined by the average mean squared error of the sample set processed by the candidate random forest model.

4. The method according to claim 1, characterized in that, The optimal Black Terns are determined from the Black Terns population based on the fitness of each Black Tern, including: The tern with the lowest fitness in the tern population was identified as the optimal tern in the tern population.

5. The method according to claim 1, characterized in that, Target information is determined based on the probability of delinquency for the target user, including: If the probability of a target user defaulting on a payment is greater than a preset value, the target information is determined to indicate that the target user has the risk of defaulting on a payment. If the probability of a target user defaulting on a payment is less than or equal to the preset value, the target information indicates that the target user does not have the risk of defaulting on a payment.

6. A data processing apparatus, characterized in that, For performing the data processing method according to any one of claims 1 to 5, comprising: The first acquisition module is used to acquire target data of a target user, wherein the target user is a user who has taken out a loan, and the target data includes at least the loan data of the target user; The processing module is used to predict the target data of the target user through a random forest model to obtain the overdue repayment probability of the target user, wherein the model parameters of the random forest model are determined based on the Black Terrier Optimization Algorithm. The first determining module is used to determine target information based on the probability of the target user's overdue repayment, wherein the target information is used to characterize whether the target user has a risk of overdue repayment.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the computer-readable storage medium is located to perform the data processing method according to any one of claims 1 to 5.

8. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the data processing method according to any one of claims 1 to 5.

9. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the data processing method according to any one of claims 1 to 5.

Images