Risk prediction method for financial products, related device and computer storage medium

Abstract

The present application provides a risk prediction method for financial products, a related device and a computer storage medium, the method comprising: real-time collection of data of at least one data source set by a customer; for each data source, feature processing is performed on the data of the data source to obtain a feature value of the data source; the feature value of each data source is input into a risk prediction model to obtain a risk prediction result; the risk prediction model is obtained by training a random forest model with historical transaction data of training sample customers; the product of the risk prediction result and the risk sensitivity is used as the final risk value; finally, if the final risk value is greater than the return threshold, a reminder message is generated; the reminder message is used to remind the customer that the financial product held by the customer is risky. Thus, the purpose of risk prediction of the financial products currently held by the customer is achieved to avoid causing a large amount of economic losses to the customer.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a risk prediction method for a financial product, a related device and a computer storage medium. Background Art

[0002] With the development of the economy, in addition to deposit products, banks will also launch other financial products for customers to choose from, such as: funds, wealth management, precious metals, futures, etc. Customers purchase financial products in the hope of obtaining greater returns.

[0003] However, financial products all have both returns and risks. Under market economy conditions, once special circumstances occur, it is very likely that customers will lose all their money and the bank itself will suffer direct economic losses. Summary of the invention

[0004] In view of this, the present application provides a risk prediction method for financial products, related devices and computer storage media, which are used to predict the risks of financial products currently held by customers to avoid causing large economic losses to customers.

[0005] The first aspect of the present application provides a risk prediction method for a financial product, comprising:

[0006] Collect data from at least one data source set by the customer in real time;

[0007] For each data source, feature processing is performed on the data of the data source to obtain a feature value of the data source;

[0008] Inputting the characteristic value of each data source into the risk prediction model to obtain a risk prediction result; wherein the risk prediction model is obtained by training a random forest model with historical transaction data of training sample customers; the historical transaction data includes historical transaction operation data of the training sample customers and data of the data source corresponding to the historical transactions of the training sample customers;

[0009] The product of the risk prediction result and the risk sensitivity is used as the final risk value; wherein the risk sensitivity is the result of evaluating the risk tolerance of the customer in advance in the form of a questionnaire;

[0010] If the final risk value is greater than the return threshold, a reminder message is generated; wherein the return threshold includes a positive threshold and a negative threshold; the return threshold is set by the customer; the reminder message is used to remind the customer that the financial product held by the customer is risky.

[0011] Optionally, the method for constructing the risk prediction model includes:

[0012] Constructing a training sample set; wherein the training sample set includes historical transaction data of multiple training sample customers;

[0013] For each of the training sample customers, the data source corresponding to the historical transactions of the training sample customers is input into the random forest model to obtain a risk prediction result;

[0014] The random forest model is continuously adjusted by utilizing the error between the risk prediction result and the actual result until the error between the risk prediction result output by the adjusted random forest model and the actual result meets the preset convergence condition, and the adjusted random forest model is determined as the risk prediction model; wherein the actual result is the historical transaction operation corresponding to the data source corresponding to the historical transaction performed by the training sample customer.

[0015] Optionally, the method further includes: continuously adjusting the random forest model by using the error between the risk prediction result and the actual result until the error between the risk prediction result output by the adjusted random forest model and the actual result meets a preset convergence condition, and determining the adjusted random forest model as the risk prediction model.

[0016] Calculate the mean of the errors between the risk prediction results and the actual results of all the training sample customers in the training sample set as the correction parameter;

[0017] Taking the sum of the risk prediction result and the correction parameter as the final risk prediction result;

[0018] The method of continuously adjusting the random forest model by using the error between the risk prediction result and the actual result until the error between the prediction result output by the adjusted random forest model and the actual result meets a preset convergence condition, and determining the adjusted random forest model as the risk prediction model includes:

[0019] The error between the final risk prediction result and the actual result is used to continuously adjust the random forest model until the error between the final risk prediction result output by the adjusted random forest model and the actual result meets the preset convergence condition, and the adjusted random forest model is determined as the risk prediction model.

[0020] Optionally, the risk prediction method for the financial product further includes:

[0021] If the final risk value is less than the return threshold and greater than the system inherent error, the funds are redeemed according to the redemption settings set for the customer; wherein the system inherent error is the cumulative error generated when all customers in the system perform risk prediction.

[0022] A second aspect of the present application provides a risk prediction device for a financial product, comprising:

[0023] A collection unit, used for collecting data from at least one data source set by a customer in real time;

[0024] A feature processing unit, configured to perform feature processing on the data of each data source to obtain a feature value of the data source;

[0025] A first input unit is used to input the characteristic value of each data source into the risk prediction model to obtain a risk prediction result; wherein the risk prediction model is obtained by training a random forest model with historical transaction data of training sample customers; the historical transaction data includes historical transaction operation data performed by the training sample customers and data of the data source corresponding to the historical transactions performed by the training sample customers;

[0026] A risk determination unit, configured to use the product of the risk prediction result and the risk sensitivity as a final risk value; wherein the risk sensitivity is a result of evaluating the risk tolerance of the customer in advance in the form of a questionnaire;

[0027] A reminder unit is used to generate a reminder message if the final risk value is greater than a profit threshold; wherein the profit threshold includes a positive threshold and a negative threshold; and the profit threshold is set by the customer.

[0028] Optionally, the risk prediction model construction unit includes:

[0029] A construction unit, used to construct a training sample set; wherein the training sample set includes historical transaction data of multiple training sample customers;

[0030] The second input unit is used to input data of a data source corresponding to the historical transactions of each training sample customer into the random forest model to obtain a risk prediction result;

[0031] An adjustment unit is used to continuously adjust the random forest model by using the error between the risk prediction result and the actual result, until the error between the risk prediction result output by the adjusted random forest model and the actual result meets a preset convergence condition, and the adjusted random forest model is determined as the risk prediction model; wherein the actual result is the historical transaction operation corresponding to the data source corresponding to the historical transaction performed by the training sample customer.

[0032] Optionally, the risk prediction device for the financial product further includes:

[0033] The first calculation unit is used to calculate the mean of the errors between the risk prediction results and the actual results of all the training sample customers in the training sample set as a correction parameter;

[0034] A second calculation unit, used for taking the sum of the risk prediction result and the correction parameter as the final risk prediction result;

[0035] Wherein, the adjustment unit is used for:

[0036] The error between the final risk prediction result and the actual result is used to continuously adjust the random forest model until the error between the final risk prediction result output by the adjusted random forest model and the actual result meets the preset convergence condition, and the adjusted random forest model is determined as the risk prediction model.

[0037] Optionally, the risk prediction device for the financial product further includes:

[0038] A redemption unit is used to redeem the funds according to the redemption settings set for the customer if the final risk value is less than the return threshold and greater than the system inherent error; wherein the system inherent error is the cumulative error generated when all customers in the system perform risk prediction.

[0039] A third aspect of the present application provides an electronic device, including:

[0040] one or more processors;

[0041] a storage device having one or more programs stored thereon;

[0042] When the one or more programs are executed by the one or more processors, the one or more processors implement the risk prediction method for financial products as described in any one of the first aspects.

[0043] A fourth aspect of the present application provides a computer storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the risk prediction method for a financial product as described in any one of the first aspects is implemented.

[0044] It can be seen from the above scheme that in a risk prediction method, related device and computer storage medium of a financial product provided by the present application, the risk prediction method of the financial product includes: first, collecting data of at least one data source set by the customer in real time; then, for each data source, performing feature processing on the data of the data source to obtain the feature value of the data source; then, inputting the feature value of each data source into the risk prediction model to obtain a risk prediction result; wherein the risk prediction model is obtained by training a random forest model with historical transaction data of training sample customers; the historical transaction data includes historical transaction operation data performed by the training sample customers and data of the data source corresponding to the historical transactions performed by the training sample customers; then, taking the product of the risk prediction result and the risk sensitivity as the final risk value; wherein the risk sensitivity is the result of evaluating the risk tolerance of the customer in advance in the form of a questionnaire; finally, if the final risk value is greater than the return threshold, generating a reminder message; wherein the return threshold includes a positive threshold and a negative threshold; the return threshold is set by the customer; the reminder message is used to remind the customer that the financial product held by the customer is risky. This will help us predict the risks of the financial products currently held by customers, so as to avoid causing large financial losses to customers. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0046] Figure 1 A specific flow chart of a risk prediction method for a financial product provided in an embodiment of the present application;

[0047] Figure 2 A flowchart of a method for constructing a risk prediction model provided in another embodiment of the present application;

[0048] Figure 3 A schematic diagram of a risk prediction device for a financial product provided by another embodiment of the present application;

[0049] Figure 4A schematic diagram of a construction unit of a risk prediction model provided in another embodiment of the present application;

[0050] Figure 5 A schematic diagram of an electronic device for implementing a risk prediction method for financial products provided in another embodiment of the present application. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] It should be noted that the concepts of "first", "second", etc. mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units, and 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 sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0053] The present application embodiment provides a risk prediction method for a financial product, such as Figure 1 As shown, the specific steps include:

[0054] S101. Collect data from at least one data source set by a customer in real time.

[0055] Among them, the data source can be but is not limited to articles, gold prices, Douglas Index, etc. Taking the data source as an article as an example, the data of the data source can include but is not limited to the article title, article reading volume, comments on the article, positive emotional comments, negative emotional comments, etc., which are not limited here.

[0056] Specifically, the data from the data source may be collected by means of, but not limited to, crawlers.

[0057] S102: Perform feature processing on the data of each data source to obtain feature values ​​of the data source.

[0058] It should be noted that the methods for feature processing of data are very diverse and mature, and can include but are not limited to outlier filtering, mean filling, character replacement, normalization and other operations to finally obtain the feature values ​​of the data source.

[0059] S103: Input the characteristic value of each data source into the risk prediction model to obtain the risk prediction result.

[0060] Among them, the risk prediction model is obtained by training the random forest model with the historical transaction data of the training sample customers; the historical transaction data includes the historical transaction operation data of the training sample customers and the data source data corresponding to the historical transactions of the training sample customers.

[0061] Optionally, in another embodiment of the present application, an implementation of a method for constructing a risk prediction model includes:

[0062] S201: Construct a training sample set.

[0063] The training sample set includes historical transaction data of multiple training sample customers.

[0064] S202: For each training sample customer, the data source corresponding to the historical transactions of the training sample customer is input into the random forest model to obtain a risk prediction result.

[0065] S203: Determine whether the error between the risk prediction result and the actual result meets the preset convergence condition.

[0066] Specifically, if it is determined that the error between the risk prediction result and the actual result meets the preset convergence condition, step S205 is executed; if it is determined that the error between the risk prediction result and the actual result does not meet the preset convergence condition, step S204 is executed.

[0067] To improve the accuracy of the model, before step S203 is executed, an implementation of the risk prediction method for financial products provided by the present application further includes:

[0068] The mean value of the error between the risk prediction results and the actual results of all training sample customers in the training sample set is calculated as the correction parameter.

[0069] The sum of the risk prediction result and the correction parameter is taken as the final risk prediction result.

[0070] Then, an implementation of step S203 includes:

[0071] Determine whether the error between the final risk prediction result and the actual result meets the preset convergence condition.

[0072] S204. Using the error between the risk prediction result and the actual result, adjust the random forest model.

[0073] S205. Determine the random forest model as a risk prediction model.

[0074] The actual result is the historical transaction operation corresponding to the data source corresponding to the historical transaction performed by the training sample customer.

[0075] S104. The product of the risk prediction result and the risk sensitivity is taken as the final risk value.

[0076] Among them, risk sensitivity is the result of assessing the customer's risk tolerance in advance in the form of a questionnaire.

[0077] S105. Determine whether the final risk value is greater than the return threshold.

[0078] Among them, the profit threshold includes a positive threshold and a negative threshold; the profit threshold is set by the customer.

[0079] It is understandable that when the final risk value is greater than the positive threshold, it means that the current wealth management product’s return is too high, that is, the increase is too large. The greater the increase, the more likely it is to fall. Therefore, it is necessary to remind customers to act as promptly as possible to obtain maximum returns. When the final risk value is greater than the negative threshold, it means that the current wealth management product may continue to fall. Therefore, a negative threshold needs to be set to remind users to act in time and stop losses in time.

[0080] Specifically, if the final risk value is greater than the return threshold, step S106 is executed.

[0081] Optionally, in another embodiment of the present application, an implementation of the risk prediction method for a financial product further includes:

[0082] If the final risk value is less than the return threshold and greater than the system's inherent error, the funds will be redeemed according to the redemption settings set for the customer.

[0083] Among them, the redemption settings are pre-set by the customer, for example: when the final risk value is less than the return threshold and greater than the system's inherent error, 20% of the funds will be redeemed; when the final risk value is less than the return threshold and greater than the system's inherent error, 40% of the funds will be redeemed, etc. There is no limitation here.

[0084] Among them, the system inherent error is the cumulative error generated when all customers in the system perform risk prediction.

[0085] S106: Generate reminder information.

[0086] Among them, the reminder information is used to remind customers that the financial products they hold are risky, so that customers can make decisions in time to avoid causing large financial losses to customers.

[0087] Similarly, when generating reminder information, if the customer has pre-set redemption settings, the funds will be redeemed according to the redemption settings set for the customer. For example: if the final risk value is greater than the return threshold, 40% of the funds will be redeemed; if the final risk value is greater than the return threshold, 100% of the funds will be redeemed, etc.

[0088] It can be seen from the above scheme that the risk prediction method of a financial product provided by the present application includes: first, collecting data of at least one data source set by the customer in real time; then, for each data source, performing feature processing on the data of the data source to obtain the feature value of the data source; then, inputting the feature value of each data source into the risk prediction model to obtain the risk prediction result; wherein the risk prediction model is obtained by training the random forest model with the historical transaction data of the training sample customer; the historical transaction data includes the historical transaction operation data of the training sample customer and the data source corresponding to the historical transaction of the training sample customer; then, the product of the risk prediction result and the risk sensitivity is used as the final risk value; wherein the risk sensitivity is the result of evaluating the risk tolerance of the customer in advance in the form of a questionnaire; finally, if the final risk value is greater than the return threshold, a reminder message is generated; wherein the return threshold includes a positive threshold and a negative threshold; the return threshold is set by the customer; the reminder message is used to remind the customer that the financial product held by the customer is risky. Thus, the risk prediction of the financial product currently held by the customer is achieved to avoid causing a large amount of economic losses to the customer.

[0089] Another embodiment of the present application provides a risk prediction device for a financial product, such as Figure 3 As shown, specifically including:

[0090] The collection unit 301 is used to collect data from at least one data source set by a customer in real time.

[0091] The feature processing unit 302 is used to perform feature processing on the data of each data source to obtain a feature value of the data source.

[0092] The first input unit 303 is used to input the characteristic value of each data source into the risk prediction model to obtain a risk prediction result.

[0093] Among them, the risk prediction model is obtained by training the random forest model with the historical transaction data of the training sample customers; the historical transaction data includes the historical transaction operation data of the training sample customers and the data source data corresponding to the historical transactions of the training sample customers.

[0094] Optionally, in another embodiment of the present application, an implementation of the construction unit of the risk prediction model is as follows: Figure 4 As shown, including:

[0095] The construction unit 401 is used to construct a training sample set.

[0096] The training sample set includes historical transaction data of multiple training sample customers.

[0097] The second input unit 402 is used to input the data source corresponding to the historical transactions of each training sample customer into the random forest model to obtain a risk prediction result.

[0098] The adjustment unit 403 is used to continuously adjust the random forest model using the error between the risk prediction result and the actual result, until the error between the risk prediction result output by the adjusted random forest model and the actual result meets the preset convergence condition, and the adjusted random forest model is determined as the risk prediction model.

[0099] The actual result is the historical transaction operation corresponding to the data source corresponding to the historical transaction performed by the training sample customer.

[0100] For the specific working process of the units disclosed in the above embodiments of the present application, please refer to the corresponding method embodiments, such as Figure 2 As shown, no further description is given here.

[0101] The risk determination unit 304 is used to take the product of the risk prediction result and the risk sensitivity as the final risk value.

[0102] Among them, risk sensitivity is the result of assessing the customer's risk tolerance in advance in the form of a questionnaire.

[0103] The reminder unit 305 is used to generate a reminder message if the final risk value is greater than the return threshold.

[0104] Among them, the profit threshold includes a positive threshold and a negative threshold; the profit threshold is set by the customer.

[0105] For the specific working process of the units disclosed in the above embodiments of the present application, please refer to the corresponding method embodiments, such as Figure 1 As shown, no further description is given here.

[0106] Optionally, in another embodiment of the present application, an implementation of the risk prediction device for financial products further includes:

[0107] The first calculation unit is used to calculate the mean of the errors between the risk prediction results and the actual results of all the training sample customers in the training sample set as a correction parameter.

[0108] The second calculation unit is used to take the sum of the risk prediction result and the correction parameter as the final risk prediction result.

[0109] The adjustment unit 403 is used to:

[0110] The error between the final risk prediction result and the actual result is used to continuously adjust the random forest model until the error between the final risk prediction result output by the adjusted random forest model and the actual result meets the preset convergence condition, and the adjusted random forest model is determined as the risk prediction model.

[0111] The specific working process of the units disclosed in the above embodiments of the present application can be found in the corresponding method embodiments, which will not be repeated here.

[0112] Optionally, in another embodiment of the present application, an implementation of the risk prediction device for financial products further includes:

[0113] The redemption unit is used to redeem funds according to the redemption settings set for the customer if the final risk value is less than the return threshold and greater than the system's inherent error.

[0114] Among them, the system inherent error is the cumulative error generated when all customers in the system perform risk prediction.

[0115] The specific working process of the units disclosed in the above embodiments of the present application can be found in the corresponding method embodiments, which will not be repeated here.

[0116] It can be seen from the above scheme that a risk prediction device for a financial product provided by the present application includes: first, a collection unit 301 collects data of at least one data source set by a customer in real time; then, a feature processing unit 302 performs feature processing on the data of each data source to obtain a feature value of the data source; then, a first input unit 303 inputs the feature value of each data source into a risk prediction model to obtain a risk prediction result; wherein, the risk prediction model is obtained by training a random forest model with historical transaction data of training sample customers; the historical transaction data includes historical transaction operation data performed by the training sample customers and data of the data source corresponding to the historical transactions performed by the training sample customers; then, a risk determination unit 304 uses the product of the risk prediction result and the risk sensitivity as the final risk value; wherein, the risk sensitivity is the result of evaluating the risk tolerance of the customer in advance in the form of a questionnaire; finally, if the final risk value is greater than the return threshold, the reminder unit 305 generates a reminder message; wherein, the return threshold includes a positive threshold and a negative threshold; the return threshold is set by the customer; the reminder message is used to remind the customer that the financial product held by the customer is risky. This will help us predict the risks of the financial products currently held by customers, so as to avoid causing large financial losses to customers.

[0117] Another embodiment of the present application provides an electronic device, such as Figure 5 As shown, including:

[0118] One or more processors 501 .

[0119] The storage device 502 stores one or more programs.

[0120] When the one or more programs are executed by the one or more processors 501, the one or more processors 501 implement the risk prediction method for financial products as described in any one of the above embodiments.

[0121] Another embodiment of the present application provides a computer storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the risk prediction method for a financial product as described in any one of the above embodiments is implemented.

[0122] In the above embodiments disclosed in the present application, it should be understood that the disclosed apparatus and method can also be implemented in other ways. The apparatus and method embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the apparatus, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, program segment or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0123] In addition, each functional module in each embodiment of the present disclosure can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part. If the function is implemented in the form of a software functional module 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 the present disclosure is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a live broadcast device, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), disk or optical disk and other media that can store program code.

[0124] The present application can be implemented or used by professionals and technicians. Various modifications to these embodiments will be apparent to professionals and technicians in the field, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A risk prediction method for a financial product, characterized in that: include: Collect data from at least one data source set by the customer in real time; For each data source, feature processing is performed on the data of the data source to obtain a feature value of the data source; Inputting the characteristic value of each data source into the risk prediction model to obtain a risk prediction result; wherein the risk prediction model is obtained by training a random forest model with historical transaction data of training sample customers; the historical transaction data includes historical transaction operation data of the training sample customers and data of the data source corresponding to the historical transactions of the training sample customers; The product of the risk prediction result and the risk sensitivity is used as the final risk value; wherein the risk sensitivity is the result of evaluating the risk tolerance of the customer in advance in the form of a questionnaire; If the final risk value is greater than the return threshold, a reminder message is generated; wherein the return threshold includes a positive threshold and a negative threshold; the return threshold is set by the customer; the reminder message is used to remind the customer that the financial product held by the customer has risks; The method for constructing the risk prediction model comprises: Constructing a training sample set; wherein the training sample set includes historical transaction data of multiple training sample customers; For each of the training sample customers, the data source corresponding to the historical transactions of the training sample customers is input into the random forest model to obtain a risk prediction result; The random forest model is continuously adjusted by utilizing the error between the risk prediction result and the actual result until the error between the risk prediction result output by the adjusted random forest model and the actual result meets the preset convergence condition, and the adjusted random forest model is determined as the risk prediction model; wherein the actual result is the historical transaction operation corresponding to the data source corresponding to the historical transaction performed by the training sample customer.

2. The risk prediction method according to claim 1, characterized in that: The method further comprises: continuously adjusting the random forest model by using the error between the risk prediction result and the actual result until the error between the risk prediction result output by the adjusted random forest model and the actual result meets a preset convergence condition, and determining the adjusted random forest model as the risk prediction model. Calculate the mean of the errors between the risk prediction results and the actual results of all the training sample customers in the training sample set as the correction parameter; Taking the sum of the risk prediction result and the correction parameter as the final risk prediction result; The method of continuously adjusting the random forest model by using the error between the risk prediction result and the actual result until the error between the prediction result output by the adjusted random forest model and the actual result meets a preset convergence condition, and determining the adjusted random forest model as the risk prediction model includes: The error between the final risk prediction result and the actual result is used to continuously adjust the random forest model until the error between the final risk prediction result output by the adjusted random forest model and the actual result meets the preset convergence condition, and the adjusted random forest model is determined as the risk prediction model.

3. The risk prediction method according to claim 1, characterized in that: Also includes: If the final risk value is less than the return threshold and greater than the system inherent error, the funds are redeemed according to the redemption settings set for the customer; wherein the system inherent error is the cumulative error generated when all customers in the system perform risk prediction.

4. A risk prediction device for a financial product, characterized in that: include: A collection unit, used for collecting data from at least one data source set by a customer in real time; A feature processing unit, configured to perform feature processing on the data of each data source to obtain a feature value of the data source; A first input unit is used to input the characteristic value of each data source into the risk prediction model to obtain a risk prediction result; wherein the risk prediction model is obtained by training a random forest model with historical transaction data of training sample customers; the historical transaction data includes historical transaction operation data performed by the training sample customers and data of the data source corresponding to the historical transactions performed by the training sample customers; A risk determination unit, configured to use the product of the risk prediction result and the risk sensitivity as a final risk value; wherein the risk sensitivity is a result of evaluating the risk tolerance of the customer in advance in the form of a questionnaire; a reminder unit, configured to generate a reminder message if the final risk value is greater than a profit threshold; wherein the profit threshold includes a positive threshold and a negative threshold; and the profit threshold is set by the customer; The construction unit of the risk prediction model includes: A construction unit, used to construct a training sample set; wherein the training sample set includes historical transaction data of multiple training sample customers; The second input unit is used to input data of a data source corresponding to the historical transactions of each training sample customer into the random forest model to obtain a risk prediction result; An adjustment unit is used to continuously adjust the random forest model by using the error between the risk prediction result and the actual result, until the error between the risk prediction result output by the adjusted random forest model and the actual result meets a preset convergence condition, and the adjusted random forest model is determined as the risk prediction model; wherein the actual result is the historical transaction operation corresponding to the data source corresponding to the historical transaction performed by the training sample customer.

5. The risk prediction device according to claim 4, characterized in that: Also includes: The first calculation unit is used to calculate the mean of the errors between the risk prediction results and the actual results of all the training sample customers in the training sample set as a correction parameter; A second calculation unit, used for taking the sum of the risk prediction result and the correction parameter as the final risk prediction result; Wherein, the adjustment unit is used for: The error between the final risk prediction result and the actual result is used to continuously adjust the random forest model until the error between the final risk prediction result output by the adjusted random forest model and the actual result meets the preset convergence condition, and the adjusted random forest model is determined as the risk prediction model.

6. The risk prediction device according to claim 4, characterized in that: Also includes: A redemption unit is used to redeem the funds according to the redemption settings set for the customer if the final risk value is less than the return threshold and greater than the system inherent error; wherein the system inherent error is the cumulative error generated when all customers in the system perform risk prediction.

7. An electronic device, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the risk prediction method for financial products as described in any one of claims 1 to 3.

8. A computer storage medium, characterized in that A computer program is stored thereon, wherein when the computer program is executed by a processor, the risk prediction method for a financial product as described in any one of claims 1 to 3 is implemented.

Images