Monitoring and early warning method and device for financial risk control model, electronic equipment and medium

Abstract

The invention provides a monitoring and early warning method and device for a financial risk control model, electronic equipment and a medium, and the method comprises the steps: obtaining financial monitoring data, and carrying out the preprocessing of the financial monitoring data; wherein the preprocessing comprises encryption processing; inputting the preprocessed financial monitoring data into a pre-trained financial risk control model to obtain early warning information; wherein the early warning information at least comprises an event triggering risk early warning, risk occurrence time, a risk type and a risk level; and sending the early warning information to a worker, so that the worker processes an event triggering risk early warning. According to the invention, the timeliness and accuracy of risk early warning are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer machine learning technology, and in particular to a monitoring and early warning method, device, electronic equipment and medium for a financial risk control model. Background Art

[0002] In the field of financial risk control, there are multiple model monitoring and early warning platforms that use technologies such as big data, artificial intelligence, and machine learning to conduct real-time monitoring and early warning of risks in financial transactions, credit, and investment. However, existing financial risk control model early warning monitoring platforms still have problems such as low timeliness and accuracy of risk warnings. Summary of the invention

[0003] In view of this, an object of the present invention is to provide a monitoring and early warning method, device, electronic equipment and medium for a financial risk control model to improve the timeliness and accuracy of risk early warning.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present invention provides a monitoring and early warning method for a financial risk control model, comprising: obtaining financial monitoring data, and preprocessing the financial monitoring data; wherein the preprocessing comprises: encryption processing; inputting the preprocessed financial monitoring data into a pre-trained financial risk control model to obtain early warning information; wherein the early warning information comprises at least: an event triggering a risk warning, the time of risk occurrence, the risk type and the risk level; and sending the early warning information to a staff member so that the staff member can process the event triggering the risk warning.

[0006] Optionally, the preprocessed financial monitoring data is input into a pre-trained financial risk control model to obtain early warning information, including: determining monitoring indicators of the preprocessed financial monitoring data based on the pre-trained financial risk control model; if the monitoring indicators exceed a pre-set monitoring early warning threshold, generating early warning information based on the monitoring indicators.

[0007] Optionally, monitoring indicators of preprocessed financial monitoring data are determined based on a pre-trained financial risk control model, including: calculating initial indicators of preprocessed financial monitoring data based on the pre-trained financial risk control model; wherein the initial indicators include at least one of statistical indicators, distribution intervals and offset indicators; and determining monitoring indicators of preprocessed financial monitoring data based on the initial indicators.

[0008] Optionally, it also includes: generating monitoring reports based on monitoring indicators and visually displaying the monitoring indicators.

[0009] Optionally, it also includes: obtaining the memory utilization of the financial risk control model; if the memory utilization exceeds a preset memory threshold, generating a warning message.

[0010] Optionally, it also includes: obtaining the CPU utilization of the financial risk control model; if the CPU utilization exceeds a preset CPU threshold, generating a warning message.

[0011] Optionally, after sending the warning information to the staff, it also includes: calculating the accuracy of the warning information; if the accuracy of the warning information is lower than a preset accuracy threshold, updating the financial risk control model.

[0012] In a second aspect, the present invention provides a monitoring and early warning device for a financial risk control model, comprising: a data acquisition module, used to acquire financial monitoring data and pre-process the financial monitoring data; wherein the pre-processing includes: encryption processing; a monitoring and early warning module, used to input the pre-processed financial monitoring data into a pre-trained financial risk control model to obtain early warning information; wherein the early warning information includes at least: events that trigger risk warnings, risk occurrence time, risk type and risk level; an early warning processing module, used to send the early warning information to staff so that the staff can process the events that trigger risk warnings.

[0013] In a third aspect, the present invention provides an electronic device comprising a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the steps of any one of the methods provided in the first aspect above.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any one of the methods provided in the first aspect are executed.

[0015] The present invention brings the following beneficial effects:

[0016] The monitoring and early warning method, device, electronic device and medium of the above-mentioned financial risk control model provided by the present invention first obtain financial monitoring data and pre-process the financial monitoring data; wherein the pre-processing includes: encryption processing; then the pre-processed financial monitoring data is input into the pre-trained financial risk control model to obtain early warning information; wherein the early warning information at least includes: the event that triggers the risk warning, the time of risk occurrence, the risk type and the risk level; finally, the early warning information is sent to the staff so that the staff can handle the event that triggers the risk warning. In the above-mentioned method, by pre-processing the financial monitoring data, the data quality can be improved, and then based on the pre-processed financial monitoring data, the early warning information is obtained through the financial risk control model, which can improve the timeliness and accuracy of the risk warning; at the same time, in the above-mentioned method, the financial monitoring data is encrypted, which can prevent data leakage and ensure the confidentiality and integrity of the data.

[0017] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0018] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the preferred embodiments are specifically mentioned below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A flowchart of a monitoring and early warning method for a financial risk control model provided by an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of the structure of a monitoring and early warning device for a financial risk control model provided by an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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.

[0024] At present, although the existing financial risk control model early warning monitoring platform has made significant progress in many aspects, it still has some shortcomings and defects, as follows:

[0025] 1. Low timeliness of risk warning: The financial risk control model monitoring and warning platform needs to be able to detect and warn potential risks in the first place. However, due to factors such as delays in data acquisition and processing, errors in model prediction, etc., the timeliness of risk warning may be affected.

[0026] 2. Weak data processing capabilities: Complex data sources and formats increase the difficulty of data processing. Some platforms have deficiencies in data processing, which may affect the accuracy and timeliness of model warnings.

[0027] 3. Poor accuracy of risk identification: Due to the complexity and uncertainty of the financial market, as well as the diversity and hidden nature of risk characteristics, the accuracy of risk identification may be challenged. The lack of adaptability of the model may lead to a decline in early warning effectiveness and even false alarms and omissions.

[0028] 4. The monitoring and early warning mechanism is relatively simple: Currently, most of them focus on a single performance indicator (such as accuracy, response time, etc.), lack a comprehensive monitoring and early warning mechanism, and it is difficult to fully reflect the operating status of the model. In addition, existing monitoring solutions often rely on manually set thresholds for abnormal judgment, lack of adaptability and intelligence, resulting in untimely warnings or high false alarm rates.

[0029] 5. Low degree of service customization: Different financial institutions have different risk preferences and business models, but the services provided by some platforms lack sufficient customization and are difficult to meet the personalized needs of financial institutions.

[0030] Based on this, the monitoring and early warning method, device, electronic device and medium of a financial risk control model provided by the embodiment of the present invention can improve the timeliness and accuracy of risk early warning.

[0031] To facilitate understanding of this embodiment, a monitoring and early warning method for a financial risk control model disclosed in an embodiment of the present invention is first introduced in detail. The method can be executed by an electronic device, such as a smart phone, a computer, a tablet computer, etc. Figure 1The flowchart of a monitoring and early warning method of a financial risk control model shown in FIG. 1 illustrates that the method mainly includes the following steps S101 to S103:

[0032] Step S101: Acquire financial monitoring data and pre-process the financial monitoring data.

[0033] In one implementation, the financial monitoring data required for monitoring and early warning by financial institutions is first determined, including: data sources, such as the internal systems of financial institutions, third-party data providers with licensed credit reporting access, etc., and data types, which may include internal data (such as user profile data, transaction data, credit history, etc.) and external third-party data authorized by customers (such as e-commerce consumption, payment, operator, etc. data).

[0034] When acquiring data, in order to ensure the accuracy and timeliness of data capture and avoid data omissions or delays, you can write a data capture program to extract the required financial monitoring data from the data source; you can also obtain financial monitoring data from multiple data sources (such as user transaction data, credit history, social media behavior, etc.) through API interfaces, data crawling, file imports, etc.

[0035] In order to ensure the validity and accuracy of the data, after obtaining the financial monitoring data, the acquired data can be preprocessed by data cleaning, denoising and formatting. Specifically, distributed computing and parallel processing technology can be used for preprocessing, which includes: cleaning, deduplication, filling missing values, data formatting, outlier processing, variable name mapping, type conversion and other preprocessing operations on the captured data to improve data quality, ensure data integrity and accuracy, and reduce the impact of noise data on model stability; convert data into a format suitable for model training, such as converting text data into numerical data, encoding numerical data, such as using one-hot encoding or WOE encoding to process categorical variables, so that the model can better understand and process these data; standardize or normalize the data to facilitate the subsequent improvement of the model training effect; generate a wide table of input feature variables to provide a basis for subsequent modeling and analysis.

[0036] In addition, for large amounts of sensitive financial data, advanced data encryption and privacy protection technologies can be used to encrypt the acquired financial monitoring data to ensure the confidentiality and integrity of the data.

[0037] Step S102: input the pre-processed financial monitoring data into a pre-trained financial risk control model to obtain early warning information.

[0038] In one implementation, financial monitoring data can be monitored in real time using a pre-trained financial risk control model. The financial risk control model can generate warning information based on preset warning rules. The warning information includes at least: events that trigger risk warnings, risk occurrence time, risk type, risk level, impact range, etc.

[0039] In specific implementation, the training of financial risk control models includes the following processes:

[0040] First, obtain the financial monitoring data for model training; then preprocess the collected data; then extract the feature vector that can reflect the operating status of the model from the preprocessed data, use the feature selection method to screen out important features, select the appropriate model training based on business needs and data characteristics, and use cross-validation and other techniques to verify the trained model using the test set and output the final model.

[0041] Specifically, first, according to business needs and model performance requirements, extract feature vectors that can reflect the operating status of the model, such as time series features, statistical features, etc., and use feature selection methods (such as IV information value, information gain, correlation coefficient, etc.) to sort and screen the features, and screen out the features that contribute most to the model's prediction performance; then randomly divide the data set into a training set and a test set with a split ratio of 3:7. Combined with the risk type and target, consider factors such as model accuracy, stability, interpretability and computational efficiency, and select a suitable financial risk control model, such as: nonlinear risk control model, logistic regression model, GBDT or XGBoost decision tree model, neural network model, etc.; use the training set data to train the model, and use the grid search method to tune the parameters, use confusion matrix, AUC, KS and other indicators to evaluate the performance of the model, ensure the accuracy and stability of the model, and obtain a trained financial risk control model.

[0042] Furthermore, cross-validation and other techniques are used to verify the trained financial risk control model using a test set, and the model's accuracy, precision, recall rate and other indicators are evaluated. The model is verified and tested to ensure the model's stability, reliability and generalization ability; and the model is adjusted and optimized based on the verification results, or a third-party model verification team is introduced to conduct independent verification to ensure the objectivity and accuracy of the model.

[0043] Step S103: Send the warning information to the staff so that the staff can handle the event that triggers the risk warning.

[0044] In one implementation, the warning information is promptly notified to relevant staff via email, SMS, message push, etc., so that the staff can handle the event that triggers the risk warning, including: risk warning, risk resolution, risk transfer, etc., as well as adjusting model parameters or switching to backup models, etc., to minimize the impact of the anomaly on the business.

[0045] The monitoring and early warning method of the above-mentioned financial risk control model provided by the present invention can improve data quality by preprocessing the financial monitoring data, and then obtain early warning information based on the preprocessed financial monitoring data through the financial risk control model, which can improve the timeliness and accuracy of risk warning; at the same time, the financial monitoring data is encrypted in the above-mentioned method, which can prevent data leakage and ensure the confidentiality and integrity of the data.

[0046] In one embodiment, for the aforementioned step S102, that is, when the preprocessed financial monitoring data is input into a pre-trained financial risk control model to obtain early warning information, methods including but not limited to the following can be adopted: determining the monitoring indicators of the preprocessed financial monitoring data based on the pre-trained financial risk control model; if the monitoring indicators exceed a pre-set monitoring early warning threshold, generating early warning information based on the monitoring indicators.

[0047] In specific implementation, the financial monitoring data is scored, evaluated and monitored according to the trained financial risk control model, that is, the monitoring indicators of the financial monitoring data are calculated, and the monitoring warning threshold is pre-set according to the model training results and the risk tolerance of the financial institution. When the actual monitoring indicator exceeds the monitoring warning threshold, the warning signal is triggered, the warning information is generated, and the risk assessment report is generated.

[0048] Specifically, determining the monitoring indicators of the preprocessed financial monitoring data based on a pre-trained financial risk control model includes: calculating the initial indicators of the preprocessed financial monitoring data based on the pre-trained financial risk control model; wherein the initial indicators include at least one of a statistical indicator, a distribution interval and an offset indicator; and determining the monitoring indicators of the preprocessed financial monitoring data based on the initial indicators.

[0049] Among them, the statistical indicators include at least: the mean, standard deviation, maximum value and minimum value of the financial monitoring data, and the deviation indicators include at least: PSI and JS divergence.

[0050] In an embodiment of the present invention, the deviation index PSI and its changing trend within a certain period of time can be used as an early warning indicator of model stability. Specifically, it is generally the PSI of the model score on the day and the model score on the base day. The base day can be the model launch date, or the model score distribution during a verification period. The purpose is to measure whether the distribution of the model on the day is very different from that at the beginning. If there are significant differences between the training data set and the monitoring data, it is necessary to rebuild the model with different variables, segmentation and grouping, samples, etc.

[0051] PSI reflects the stability of the distribution of validation samples in each segment and the distribution of modeling samples. It is often used to screen feature variables and evaluate the stability of the model in modeling. Stability has a reference, so two distributions are required: actual distribution and expected distribution. Among them, the training set is used as the expected distribution during modeling, and the test set and cross-time validation set are used as the actual distribution.

[0052] The PSI calculation formula is as follows:

[0053]

[0054] Among them, Actuali% represents the number of samples in the i-th bin in the actual distribution, and Expectedi% is the number of samples in the i-th bin in the expected distribution.

[0055] The calculation process of PSI is as follows:

[0056] (1) Select benchmark samples: Select benchmark samples (Benchmark) as expected distribution (Excepted), generally using training samples as benchmarks.

[0057] (2) Binning the expected distribution: Bin the expected distribution and obtain the proportion of samples in each box (number of samples in the box / total number of samples). It is worth noting that:

[0058] ① Binning can be equal frequency, equal distance or other methods. Different binning methods will lead to slightly different calculation results;

[0059] ② For continuous variables (feature variables, model scores, etc.), the number of bins needs to be set reasonably, generally set to 10 or 20; for discrete variables, if there are too many bins, you can consider merging small bins in advance; too many bins may result in too few samples in each bin and lose statistical significance; too few bins will reduce the accuracy of the calculation results. Bin the actual distribution (Actual) according to the expected distribution bin threshold and calculate the proportion of samples in each bin.

[0060] (3) Calculate the PSI value of each box: Calculate the index of each box according to the formula (actual proportion - expected proportion) / ln(actual proportion / expected proportion);

[0061] (4) Accumulate the PSI value of each box: Accumulate the index of each box to get the final PSI.

[0062] The smaller the PSI value, the closer the actual distribution is to the expected distribution, and the more stable the model is. It is generally believed that when the PSI value is less than 0.1, the model stability is very good; when the PSI value is between 0.1 and 0.25, the model stability is good; when the PSI value is greater than 0.25, the model stability is poor and the model may need to be readjusted.

[0063] Based on this, in this embodiment, under the premise of sufficient incoming documents, the time granularity can be refined to hours, the PSI of the model score and the model score of the previous hour can be calculated hour by hour, and a threshold can be set. Once the PSI exceeds the threshold, an early warning message is issued.

[0064] In one implementation, the above method also supports the configuration of warning rules for service resource indicators such as memory occupancy rate and CPU occupancy rate. Specifically, it includes:

[0065] (1) Obtain the memory utilization of the financial risk control model; if the memory utilization exceeds the pre-set memory threshold, generate an early warning message. Specifically, support customized configuration of the early warning rule of the data deviation indicator "memory utilization", and send an early warning message when the trigger condition (i.e., exceeding the memory threshold) is triggered.

[0066] (2) Obtain the CPU utilization of the financial risk control model; if the CPU utilization exceeds the pre-set CPU threshold, generate an early warning message. Specifically, support customized configuration of the early warning rule of the data deviation indicator "CPU utilization", and send an early warning message when the trigger suite (i.e., exceeding the CPU threshold) is triggered.

[0067] Furthermore, the above method also includes: generating a monitoring report based on the monitoring indicators, and visually displaying the monitoring indicators.

[0068] In specific implementation, the above method also provides convenient early warning query and statistical analysis functions, including: generating monitoring reports, and using data visualization technology to display monitoring indicators in the form of charts, etc., thereby improving data readability and analysis efficiency.

[0069] In one embodiment, after obtaining the warning information, the warning information can be analyzed to evaluate the predictive performance of the financial risk control model, calculate the performance indicators and monitor their changing trends, and continuously update the data and models according to business needs and market changes, optimize the risk control strategy and warning rules to adapt to new risk characteristics. Based on this, after sending the warning information to the staff, the above method also includes: calculating the accuracy of the warning information; if the accuracy of the warning information is lower than the preset accuracy threshold, the financial risk control model is updated.

[0070] In the specific implementation, the warning information is analyzed, the accuracy of the warning information is calculated, and the accuracy is monitored. When the accuracy is lower than the pre-set accuracy threshold, the financial risk control model is updated, the model parameters are adjusted or the feature selection is optimized to improve the accuracy of the model.

[0071] In addition, in this embodiment, a model monitoring report can also be generated to monitor changes in monitoring indicators such as KS, AUC, PSI, and regularly maintain and update the financial risk control model. When monitoring finds that the model performance has declined or cannot meet business needs, the model iteration process will be triggered, and the model will be retrained with new data and features, and verified and tested to ensure that the model performance is improved and the model iteration is completed. At the same time, data and models can be continuously updated according to business needs and market changes, and risk control strategies and early warning rules can be optimized to adapt to new risk characteristics.

[0072] The above method provided by the embodiment of the present invention improves the comprehensiveness and accuracy of data by optimizing data acquisition channels and integrating multiple data sources. At the same time, advanced data processing technology is used to clean and preprocess data, reduce data noise, and improve data quality; distributed computing and parallel processing technology are used to improve data processing speed to ensure that the early warning system can respond to market changes in real time; nonlinear risk control models are used, combined with machine learning and deep learning algorithms, to improve the prediction accuracy and adaptability of the model. At the same time, through continuous learning and optimization, the model can continuously adapt to new risk characteristics; a comprehensive early warning indicator system is constructed, comprehensively considering multi-dimensional factors such as economy, market, credit, industry, etc., and a combination of subjective and objective weighting methods is used to reduce the subjectivity and processing errors in the construction process of early warning indicators; through modular design, flexible configuration and customization of the early warning system are realized, and financial institutions can choose suitable modules and parameters according to their own needs to improve the pertinence and practicality of the early warning system; advanced data encryption and privacy protection technology are used to ensure the confidentiality and integrity of data. At the same time, a strict data access permission control mechanism is established to prevent data leakage and abuse.

[0073] For the monitoring and early warning method of the financial risk control model provided in the above embodiment, the embodiment of the present invention also provides a monitoring and early warning device for the financial risk control model, see Figure 2 The schematic diagram of the structure of a monitoring and early warning device for a financial risk control model is shown, indicating that the device mainly includes the following parts:

[0074] The data acquisition module 201 is used to acquire financial monitoring data and pre-process the financial monitoring data; wherein the pre-processing includes: encryption processing;

[0075] The monitoring and warning module 202 is used to input the pre-processed financial monitoring data into the pre-trained financial risk control model to obtain warning information; wherein the warning information at least includes: the event triggering the risk warning, the time of risk occurrence, the risk type and the risk level;

[0076] The warning processing module 203 is used to send warning information to staff so that the staff can handle the event that triggers the risk warning.

[0077] The monitoring and early warning device of the above-mentioned financial risk control model provided by the present invention can improve data quality by preprocessing the financial monitoring data, and then obtain early warning information based on the preprocessed financial monitoring data through the financial risk control model, which can improve the timeliness and accuracy of risk warning; at the same time, the financial monitoring data is encrypted in the above-mentioned device, which can prevent data leakage and ensure the confidentiality and integrity of the data.

[0078] In one embodiment, the monitoring and early warning module 202 is specifically used to: determine the monitoring indicators of the preprocessed financial monitoring data based on a pre-trained financial risk control model; if the monitoring indicators exceed a pre-set monitoring and early warning threshold, generate early warning information based on the monitoring indicators.

[0079] In one embodiment, the above-mentioned monitoring and early warning module 202 is specifically used to: calculate the initial indicators of the preprocessed financial monitoring data based on a pre-trained financial risk control model; wherein the initial indicators include at least one of a statistical indicator, a distribution interval and an offset indicator; and determine the monitoring indicators of the preprocessed financial monitoring data based on the initial indicators.

[0080] In one embodiment, the above-mentioned device further includes: a monitoring module, which is used to generate a monitoring report based on the monitoring indicators and to visualize the monitoring indicators.

[0081] In one embodiment, the above-mentioned device also includes: a memory monitoring module, which is used to obtain the memory utilization of the financial risk control model; if the memory utilization exceeds a preset memory threshold, a warning message is generated.

[0082] In one embodiment, the above-mentioned device also includes: a CPU monitoring module, which is used to obtain the CPU utilization of the financial risk control model; if the CPU utilization exceeds a preset CPU threshold, an early warning message is generated.

[0083] In one embodiment, the above-mentioned device also includes: an optimization and updating module, which is used to calculate the accuracy of the early warning information; if the accuracy of the early warning information is lower than a preset accuracy threshold, the financial risk control model is updated.

[0084] It should be noted that the implementation principle and technical effects of the device provided in the embodiment of the present invention are the same as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment.

[0085] An embodiment of the present invention further provides an electronic device. Specifically, the electronic device includes a processor and a storage device. The storage device stores a computer program, and when the computer program is executed by the processor, it executes the method described in any one of the above implementation methods.

[0086] Figure 3 A structural diagram of an electronic device provided in an embodiment of the present invention, the electronic device 100 includes: a processor 30, a memory 31, a bus 32 and a communication interface 33, wherein the processor 30, the communication interface 33 and the memory 31 are connected via the bus 32; the processor 30 is used to execute an executable module stored in the memory 31, such as a computer program.

[0087] The memory 31 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 33 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0088] The bus 32 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0089] Among them, the memory 31 is used to store programs, and the processor 30 executes the program after receiving the execution instruction. The method executed by the device for flow process definition disclosed in any embodiment of the above-mentioned embodiments of the present invention can be applied to the processor 30 or implemented by the processor 30.

[0090] The processor 30 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 30. The above processor 30 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 31, and the processor 30 reads the information in the memory 31 and completes the steps of the above method in combination with its hardware.

[0091] The computer program product of the readable storage medium provided in the embodiment of the present invention includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the methods described in the previous method embodiments. The specific implementation can be referred to the previous method embodiments, which will not be repeated here.

[0092] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, 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. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0093] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-mentioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-mentioned embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A monitoring and early warning method for a financial risk control model, characterized in that: include: Acquire financial monitoring data, and pre-process the financial monitoring data; wherein the pre-processing includes: encryption processing; Input the pre-processed financial monitoring data into the pre-trained financial risk control model to obtain early warning information; wherein the early warning information at least includes: the event that triggers the risk warning, the time of risk occurrence, the risk type and the risk level; The warning information is sent to the staff so that the staff can handle the event that triggers the risk warning.

2. The method according to claim 1, characterized in that: Input the pre-processed financial monitoring data into the pre-trained financial risk control model to obtain early warning information, including: Determine the monitoring indicators of the pre-processed financial monitoring data based on the pre-trained financial risk control model; If the monitoring indicator exceeds a preset monitoring warning threshold, warning information is generated based on the monitoring indicator.

3. The method according to claim 2, characterized in that Based on the pre-trained financial risk control model, the monitoring indicators of the pre-processed financial monitoring data are determined, including: Calculating initial indicators of preprocessed financial monitoring data based on a pre-trained financial risk control model; wherein the initial indicators include at least one of a statistical indicator, a distribution interval, and a deviation indicator; The monitoring index of the preprocessed financial monitoring data is determined based on the initial index.

4. The method according to claim 2, characterized in that: Also includes: Generate a monitoring report based on the monitoring indicators, and visualize the monitoring indicators.

5. The method according to claim 1, characterized in that Also includes: Obtaining the memory utilization rate of the financial risk control model; If the memory utilization exceeds a preset memory threshold, an early warning message is generated.

6. The method according to claim 1, characterized in that Also includes: Obtaining the CPU utilization of the financial risk control model; If the CPU utilization exceeds a preset CPU threshold, an early warning message is generated.

7. The method according to claim 1, characterized in that After sending the warning information to the staff, it also includes: Calculating the accuracy of the warning information; If the accuracy of the early warning information is lower than a preset accuracy threshold, the financial risk control model is updated.

8. A monitoring and early warning device for a financial risk control model, characterized in that: include: A data acquisition module, used to acquire financial monitoring data and pre-process the financial monitoring data; wherein the pre-processing includes: encryption processing; The monitoring and early warning module is used to input the pre-processed financial monitoring data into the pre-trained financial risk control model to obtain early warning information; wherein the early warning information at least includes: the event that triggers the risk warning, the time of risk occurrence, the risk type and the risk level; The warning processing module is used to send the warning information to the staff so that the staff can process the event that triggers the risk warning.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are performed.

Images