A data processing method, apparatus and electronic device

Abstract

The present invention provides a data processing method, apparatus, and electronic device. In the present invention, for each functional module, there is a corresponding data processing sub-model, and thus the fault prediction value of the functional module can be determined according to the corresponding data processing sub-model, which can avoid the situation where the accuracy of the predicted fault value is relatively low due to manual prediction of the fault value based on experience, and improve the accuracy of the fault prediction value. In addition, in the present invention, the functional module includes at least one functional sub-module. When the fault prediction value is greater than a preset threshold, based on the fault prediction values of the respective functional sub-modules, the target functional sub-module that needs to be fault-repaired is determined, and the fault repair rule of the target functional sub-module is determined. That is, the present invention can determine the fault repair rule from a finer-grained level of the target functional sub-module, and can improve the accuracy of determining the fault repair rule compared with the method of manually determining the fault repair rule based on experience.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and more specifically, to a data processing method, apparatus, and electronic device. Background Art

[0002] With the continuous increase in the number of users, if an abnormality occurs in the banking system, it will affect the user experience.

[0003] Currently, in order to avoid the situation where the banking system crashes due to an abnormality and cannot provide services to users, it is common to manually predict possible fault points of the banking system based on experience and determine fault correction rules, so as to perform fault repair in advance based on the fault correction rules and avoid the occurrence of faults.

[0004] However, this method has low accuracy, resulting in low accuracy of the predicted fault points of the banking system and low accuracy of the determined fault correction rules. Summary of the Invention

[0005] In view of this, the present invention provides a data processing method, apparatus, and electronic device to solve the problem of low accuracy in manually predicting the fault points of the banking system based on experience and determining the fault correction rules.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions:

[0007] A data processing method is applied to a controller in a banking system. The banking system includes multiple functional modules, and each functional module includes at least one functional sub-module. The data processing method includes:

[0008] Obtain a data processing model of the banking system. The data processing model includes multiple data processing sub-models, and the data processing sub-models correspond to the corresponding functional modules in the banking system.

[0009] Use the data processing sub-model to perform data processing on the operation data of each functional sub-module of the corresponding functional module to obtain a fault prediction value of the functional sub-module, and calculate a fault prediction value of the functional module based on the fault prediction values of each functional sub-module of the functional module.

[0010] Based on the association relationship between different functional modules, determine the weight value of each functional module, and calculate the fault prediction value of the banking system according to the weight value and fault prediction value of each functional module.

[0011] When the failure prediction value of the bank system is greater than the preset prediction threshold, based on the failure prediction values of each of the functional sub-modules, determine the target functional sub-module that needs to be repaired, and determine the failure repair rule of the target functional sub-module.

[0012] Optionally, use the data processing sub-model to process the operation data of each functional sub-module of the corresponding functional module to obtain the failure prediction value of the functional sub-module, and calculate the failure prediction value of the functional module based on the failure prediction values of each functional sub-module of the functional module, including:

[0013] Obtain the operation data of each functional sub-module of the functional module;

[0014] Use the data processing sub-model corresponding to the functional module to process the operation data of each functional sub-module of the functional module to obtain the failure prediction value of the functional sub-module;

[0015] Sum the failure prediction values of each functional sub-module of the functional module to obtain the failure prediction value of the functional module.

[0016] Optionally, based on the association relationship between different functional modules, determine the weight value of each functional module, and calculate the failure prediction value of the bank system according to the weight value and failure prediction value of each functional module, including:

[0017] Obtain the total number of failure points of the functional module obtained based on the data processing sub-model corresponding to the functional module, and use the total number of failure points as the initial weight value of the functional module;

[0018] Based on the association relationship between different functional modules, correct the initial weight value of the functional module to obtain the weight value of the functional module;

[0019] Take the sum of the products of the weight values and failure prediction values of each functional module as the failure prediction value of the bank system.

[0020] Optionally, take the sum of the products of the weight values and failure prediction values of each functional module as the failure prediction value of the bank system, including:

[0021] Based on the association relationship between different functional modules, determine the functional sub-modules with overlapping functions and the functional sub-modules with related functions in different functional modules;

[0022] For functional sub - modules with overlapping functions, delete the failure prediction value of the functional sub - module with the smaller failure prediction value among the functional sub - modules with overlapping functions, and re - determine the failure prediction value of the functional module including the functional sub - module with the smaller failure prediction value;

[0023] For functional sub - modules with related functions, calculate the new failure prediction value of the functional module corresponding to the functional sub - modules with related functions after deleting the failure prediction value of the functional sub - modules with related functions;

[0024] Screen out the functional modules with smaller corresponding new failure prediction values, and use the new failure prediction value corresponding to the functional module as the failure prediction value of the functional module;

[0025] Take the sum of the products of the weight values of each functional module and the corresponding failure prediction values as the failure prediction value of the banking system.

[0026] Optionally, based on the failure prediction values of each functional sub - module, determine the target functional sub - modules that need to be repaired for failures, and determine the failure repair rules for the target functional sub - modules, including:

[0027] Obtain the number of failure points of each functional sub - module of the functional module based on the data - processing sub - model corresponding to the functional module;

[0028] Take the product of the number of failure points of the functional sub - module and the failure prediction value of the functional module corresponding to the functional sub - module as the failure repair parameter of the functional sub - module;

[0029] Screen out the functional sub - modules with failure repair parameters greater than the preset repair threshold as candidate functional sub - modules;

[0030] Based on the historical failure average value and the failure prediction value of the candidate functional sub - modules, determine the target functional sub - modules that need to be repaired for failures, and determine the failure repair rules for the target functional sub - modules.

[0031] Optionally, based on the historical failure average value and the failure prediction value of the candidate functional sub - modules, determine the target functional sub - modules that need to be repaired for failures, and determine the failure repair rules for the target functional sub - modules, including:

[0032] Group the candidate functional sub - modules to obtain multiple combinations; each combination includes a preset number of candidate functional sub - modules;

[0033] For each combination, calculate the deviation degree of the failure prediction value of the candidate functional sub - modules in the combination relative to the historical failure average value;

[0034] Determine the fault repair rule for adjusting the deviation of the candidate functional sub-module to the set deviation, and calculate the new fault prediction value of the bank system when the candidate functional sub-module in the combination is repaired;

[0035] When there is a new fault prediction value of the bank system that is less than the preset prediction threshold, take the candidate functional sub-module in the combination corresponding to the smallest new fault prediction value as the target functional sub-module, and output the fault repair rule corresponding to the target functional sub-module.

[0036] Optionally, when there is no new fault prediction value of the bank system that is greater than the preset prediction threshold, it further includes:

[0037] Adjust the size of the set deviation, and return to the step of determining the fault repair rule for adjusting the deviation of the candidate functional sub-module to the set deviation, and stop until there is a new fault prediction value of the bank system that is less than the preset prediction threshold. Take the candidate functional sub-module in the combination corresponding to the smallest new fault prediction value as the target functional sub-module, and output the fault repair rule corresponding to the target functional sub-module.

[0038] A data processing device is applied to a controller in a bank system. The bank system includes multiple functional modules, and the functional modules include at least one functional sub-module; the data processing device includes:

[0039] A model acquisition module is used to acquire the data processing model of the bank system. The data processing model includes multiple data processing sub-models, and the data processing sub-models correspond to the corresponding functional modules in the bank system;

[0040] The first fault prediction module is used to perform data processing on the operation data of each functional sub-module of the corresponding functional module using the data processing sub-model, obtain the fault prediction value of the functional sub-module, and calculate the fault prediction value of the functional module based on the fault prediction values of each functional sub-module of the functional module;

[0041] The second fault prediction module is used to determine the weight value of each functional module based on the association relationship between different functional modules, and calculate the fault prediction value of the bank system according to the weight value and fault prediction value of each functional module;

[0042] The fault repair module is used to, when the fault prediction value of the bank system is greater than the preset prediction threshold, determine the target functional sub-module that needs to be repaired based on the fault prediction values of each functional sub-module, and determine the fault repair rule of the target functional sub-module.

[0043] Optionally, the first fault prediction module is specifically configured to:

[0044] Obtain the operation data of each functional sub-module of the functional module;

[0045] Use the corresponding data processing sub-model of the functional module to process the operation data of each functional sub-module of the functional module to obtain the fault prediction value of the functional sub-module;

[0046] Sum the fault prediction values of each functional sub-module of the functional module to obtain the fault prediction value of the functional module.

[0047] An electronic device includes: a memory and a processor;

[0048] Wherein, the memory is used to store programs;

[0049] The processor calls the program and is used to execute the above data processing method.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention provides a data processing method, device and electronic device. In the present invention, for each functional module, there is a corresponding data processing sub-model, and thus the fault prediction value of the functional module can be determined according to the corresponding data processing sub-model, which can avoid the situation that the fault value predicted manually according to experience has low accuracy, and improve the accuracy of the fault prediction value. In addition, in the present invention, for different functional modules, corresponding data processing sub-models are configured, making the data processing sub-model more adaptable to the functional module, and further improving the accuracy of fault prediction. Moreover, in the present invention, the functional module includes at least one functional sub-module. When the fault prediction value is greater than a preset threshold, based on the fault prediction values of each functional sub-module, the target functional sub-module that needs to be repaired is determined, and the fault repair rule of the target functional sub-module is determined. That is, the present invention can determine the fault repair rule from a finer-grained level of the target functional sub-module. Compared with the way of determining the fault repair rule manually according to experience, it can improve the accuracy of determining the fault repair rule. Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0053] Figure 1 Flowchart of a data processing method provided by an embodiment of the present invention;

[0054] Figure 2 Flowchart of another data processing method provided by an embodiment of the present invention;

[0055] Figure 3 Flowchart of yet another data processing method provided by an embodiment of the present invention;

[0056] Figure 4 Flowchart of still another data processing method provided by an embodiment of the present invention;

[0057] Figure 5 Flowchart of the fifth data processing method provided by an embodiment of the present invention;

[0058] Figure 6 Schematic structural diagram of a data processing device provided by an embodiment of the present invention. Detailed implementation manners

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0060] Currently, in order to avoid the situation where the system paralysis caused by the abnormality of the banking system cannot provide services to users, it is common to manually predict possible fault points of the banking system based on experience and determine fault correction rules, so as to perform fault repair in advance based on the fault correction rules and avoid the occurrence of faults.

[0061] However, this method has low accuracy, resulting in low accuracy of the predicted fault points of the banking system and low accuracy of the determined fault correction rules.

[0062] To solve this problem, the inventors found through research that if it is possible to automatically calculate possible fault points based on operation data and determine fault correction rules, the problem of low accuracy caused by manual determination based on experience can be avoided.

[0063] Furthermore, when automatically calculating fault points, for each functional module in the banking system, the corresponding data processing sub-model can be determined according to the data attributes of the functional module. Then, the accuracy of the fault prediction value of the functional module obtained based on the data processing sub-model is relatively high, and further, the accuracy of the fault prediction value of the banking system obtained based on the fault prediction value of the functional module is relatively high.

[0064] In addition, when determining the fault point and the fault correction rule, based on the fault prediction value situation of the above-mentioned functional sub-modules, the target functional sub-module that needs to be repaired can be determined, and the fault repair rule of the target functional sub-module can be determined. With a smaller granularity, the accuracy of the obtained fault point and the fault correction rule is higher.

[0065] Specifically, for each functional module, there is a corresponding data processing sub-model, and thus the fault prediction value of the functional module can be determined according to the corresponding data processing sub-model, which can avoid the situation where the accuracy of the predicted fault value is low due to manual prediction of the fault value based on experience, and improve the accuracy of the fault prediction value. In addition, in the present invention, for different functional modules, corresponding data processing sub-models are configured, making the data processing sub-model more adaptable to the functional module, and can further improve the accuracy of fault prediction. Moreover, in the present invention, the functional module includes at least one functional sub-module. When the fault prediction value is greater than a preset threshold, based on the fault prediction values of each functional sub-module, the target functional sub-module that needs to be repaired is determined, and the fault repair rule of the target functional sub-module is determined. That is, the present invention can determine the fault repair rule from a finer-grained level of the target functional sub-module, and compared with the way of manually determining the fault repair rule according to experience, can improve the accuracy of determining the fault repair rule.

[0066] More specifically, for each functional module, there is a corresponding data processing sub-model, and thus the fault prediction value of the functional module can be determined according to the corresponding data processing sub-model, which can avoid the situation where the accuracy of the predicted fault value is low due to manual prediction of the fault value based on experience, and improve the accuracy of the fault prediction value. In addition, in the present invention, for different functional modules, corresponding data processing sub-models are configured, making the data processing sub-model more adaptable to the functional module, and can also avoid the problem that the adaptability of the same model to different functional modules is poor and the predicted fault value is inaccurate, and improve the accuracy of the fault prediction value. Moreover, in the present invention, the functional module includes at least one functional sub-module. When the fault prediction value is greater than a preset threshold, based on the fault prediction values of each functional sub-module, the target functional sub-module that needs to be repaired is determined, and the fault repair rule of the target functional sub-module is determined. That is, the present invention can determine the fault repair rule from a finer-grained level of the target functional sub-module, and compared with the way of manually determining the fault repair rule according to experience, can improve the accuracy of determining the fault repair rule.

[0067] On the basis of the above content, another embodiment of the present invention provides a data processing method, which is applied to a controller in a banking system. The controller in this embodiment can be a background server.

[0068] The bank system includes multiple functional modules, and each functional module includes at least one functional sub-module. The functional modules in this embodiment may include a system data module, an interface module, a functional module, a business process module, etc. Each functional module can be used as an indicator.

[0069] The fault points in this embodiment may be software fault points in the above modules (such as incorrect number of test cases), or hardware fault points (such as hardware interface faults).

[0070] For each functional module, it includes at least one functional sub-module. Specifically, each indicator has multiple levels of monitoring and inspection. For example, the system data module has functional sub-modules such as data migration method, cleaning plan, audit process, test process, and exception handling process; the interface module has functional sub-modules such as interface implementation method, interface control plan, and interface exception handling; the functional module has functional sub-modules such as requirement management, development environment test process, and functional test process; the business process module has functional sub-modules such as high-risk process confirmation and defect control, and the functional risk module includes functional sub-modules such as page function, interface function, and non-requirement function.

[0071] Each level is independent of each other. The operation data of each functional module and the elements at each level of the functional module are stored in a knowledge base, and the relevant specifications of the corresponding level elements in the bank are extracted. Furthermore, based on this specification, it is possible to determine whether there is an abnormality.

[0072] The bank system is provided with a data processing model, and the data processing model includes multiple data processing sub-models, and the data processing sub-models correspond to the corresponding functional modules in the bank system. That is to say, each functional module corresponds to a corresponding data processing sub-model, and different functional modules correspond to different data processing sub-models.

[0073] For a functional module, the corresponding data processing sub-model is determined based on the data attributes of the functional module. For example, if the functional module is a system data module, and the system data module has functional sub-modules such as data migration method, cleaning plan, audit process, test process, and exception handling process, then the data processing sub-model corresponding to the system data module is used to determine whether the above several functional sub-modules are abnormal, such as judging whether the cleaning plan is reasonable, determining whether the test process is reasonable, and whether the test cases are reasonable, etc.

[0074] Refer to Figure 1 , the data processing method may include:

[0075] S11. Obtain the data processing model of the bank system.

[0076] The data processing model includes multiple data processing sub - models, and the data processing sub - models correspond to the corresponding functional modules in the banking system.

[0077] For the corresponding descriptions of the data processing sub - models and functional modules in this embodiment, please refer to the corresponding descriptions in the above - mentioned embodiment.

[0078] S12. Use the data processing sub - model to process the operation data of each functional sub - module of the corresponding functional module, obtain the fault prediction value of the functional sub - module, and calculate the fault prediction value of the functional module based on the fault prediction values of each functional sub - module of the functional module.

[0079] In this embodiment, the data output of the functional sub - modules of each functional module can be output to the data processing sub - model, and then the fault prediction value of the functional sub - module can be obtained.

[0080] In this embodiment, each functional module and each functional sub - module in the functional module perform monitoring, data collection, and fault prediction concurrently.

[0081] For each functional sub - module, the fault prediction result, that is, the risk monitoring result, adopts a discrete coding method of 0 and 1. When there is no abnormality in monitoring, the risk result is 0, and when weaknesses and security threats of this indicator are found, the risk result is set to 1.

[0082] It should be noted that in addition to outputting the risk result of 0 or 1, the data processing sub - model can also output the number of fault points of the functional sub - module, such as the number of test case errors in the test case, the number of abnormal processes in the abnormal handling process, etc. In this embodiment, the sum of the number of fault points of each functional sub - module of the functional module is the total number of corresponding fault points of the functional module.

[0083] In practical applications, step S12 may include:

[0084] 1) Obtain the operation data of each functional sub - module of the functional module.

[0085] 2) Use the data processing sub - model corresponding to the functional module to process the operation data of each functional sub - module of the functional module, and obtain the fault prediction value of the functional sub - module.

[0086] For the specific implementation processes of steps 1) and 2), please refer to the corresponding descriptions in the above - mentioned embodiment.

[0087] 3) Sum up the fault prediction values of each functional sub - module of the functional module to obtain the fault prediction value of the functional module.

[0088] For a functional module, after determining the fault prediction values of each functional sub-module of the functional module, sum up the fault prediction values of each functional sub-module, and the sum result is the fault prediction value of the functional module.

[0089] S13. Based on the association relationships between different functional modules, determine the weight values of each functional module, and calculate the fault prediction value of the banking system according to the weight values and fault prediction values of each functional module.

[0090] In this embodiment, after calculating the fault prediction values of each functional module in the banking system, the sum of the products of the weight values and fault prediction values of each functional module can be used as the fault prediction value of the banking system.

[0091] S14. When the fault prediction value of the banking system is greater than the preset prediction threshold, based on the fault prediction values of each functional sub-module, determine the target functional sub-module that needs to be repaired, and determine the fault repair rule of the target functional sub-module.

[0092] In this embodiment, a preset prediction threshold can be set, such as 0.8. If the fault prediction value of the banking system is greater than 0.8, it means that the current final fault prediction value, that is, the risk value is greater than the maximum risk threshold preset in the knowledge base. Then, start the risk post-processing module, classify the level of the scheduled security administrator in the knowledge base. For example, in the current duty schedule, A handles low-level security events and B handles high-level security events. When the final risk assessment result is less than the preset prediction threshold, it is in a low-risk state. Notify A to handle the event, and record the handling process and supervision event points of the event in the knowledge base. Once the event handling times out, promptly warn the corresponding handling personnel, so as to improve the security and stability of the system.

[0093] If the final risk assessment result is greater than the preset prediction threshold, it is in a high-risk state. At this time, notify B to handle the event, and perform corresponding monitoring operations as in the case of A handling.

[0094] In this embodiment, for each functional module, there is a corresponding data processing sub-model, so that the fault prediction value of the functional module can be determined according to the corresponding data processing sub-model, which can avoid the situation of low accuracy of the predicted fault value caused by manual prediction of the fault value based on experience and improve the accuracy of the fault prediction value. In addition, in the present invention, for different functional modules, corresponding data processing sub-models are configured, making the data processing sub-model more adaptable to the functional module, which can further improve the accuracy of fault prediction. Moreover, in the present invention, the functional module includes at least one functional sub-module. When the fault prediction value is greater than a preset threshold, based on the fault prediction values of each functional sub-module, the target functional sub-module that needs to be repaired is determined, and the fault repair rule of the target functional sub-module is determined. That is, the present invention can determine the fault repair rule from a finer-grained level of the target functional sub-module. Compared with the method of manually determining the fault repair rule based on experience, it can improve the accuracy of determining the fault repair rule.

[0095] On the basis of the above content, in another implementation manner of the present invention, the specific implementation process of step S13, "Based on the association relationship between different functional modules, determine the weight value of each functional module, and calculate the fault prediction value of the bank system according to the weight value and the fault prediction value of each functional module", is given. Refer to Figure 2 , and it may include:

[0096] S21. Obtain the total number of fault points of the functional module obtained based on the data processing sub-model corresponding to the functional module, and use the total number of fault points as the initial weight value of the functional module.

[0097] Specifically, the total number of fault points has been explained in the above embodiment. Please refer to the corresponding description above. In this embodiment, the total number of fault points is directly used as the initial weight value of the functional module.

[0098] S22. Based on the association relationship between different functional modules, correct the initial weight value of the functional module to obtain the weight value of the functional module.

[0099] In this embodiment, since there may be an association relationship between each functional module, after determining the initial weight value of the functional module, the weights of each functional module are optimized by the particle swarm algorithm to obtain the final weight value.

[0100] More specifically, when specifically using the particle swarm optimization algorithm, in the particle swarm optimization algorithm, the functional modules with an association relationship are regarded as a whole for weight calculation. After determining the weight value, the weight value is assigned to a certain functional module, and the weight value of another related functional module is zero. Among them, the assignment of the weight value to a certain functional module is obtained through training.

[0101] During the training process, assume that the number of functional modules with an association relationship is 2, namely functional module A and functional module B. First, the calculated weight value is assigned to functional module A, and the fault prediction value of the banking system is calculated based on the weight value of functional module A, and the risk level is determined based on the fault prediction value of the banking system, and then compared with the risk level marked manually. If they are the same, the calculated weight value is assigned to functional module A. The same process is carried out for functional module B.

[0102] Meanwhile, if the risk levels obtained based on both functional module A and functional module B are the same as the risk level marked manually, at this time, the functional module with the fault prediction value of the banking system closest to it is selected, and the calculated weight value is assigned to it.

[0103] After the training result, the weight value can be directly assigned according to the training result in the later stage. For example, during the training, the weight value is assigned to functional module A, then when using the particle swarm optimization algorithm hereafter, the weight value is assigned to functional module A.

[0104] S23. The sum of the products of the weight values of each of the functional modules and the fault prediction value is used as the fault prediction value of the banking system.

[0105] Specifically, the sum of the products of the weight values of each of the functional modules and the fault prediction value is the fault prediction value of the banking system.

[0106] In practical applications, since there may be cases of functional overlap and functional correlation among the functional sub-modules of each functional module, when performing fault prediction, the fault prediction value can be fine-tuned based on the overlap and correlation cases.

[0107] Refer to Figure 3 , step S23 may include:

[0108] S31. Based on the association relationship between different functional modules, determine the functional sub-modules with functional overlap and the functional sub-modules with functional correlation among different functional modules.

[0109] Specifically, the functional sub-modules of each functional module may overlap and be related. Among them, overlap means that one functional sub-module is included in the functional sub-module of another functional module, and related means that there is a related relationship between the functional sub-modules in two different functional modules.

[0110] For example, the interface function sub-module in the functional risk module and the exception handling sub-module of the interface in the interface module are functional sub-modules with overlapping functions.

[0111] The non-requirement function sub-modules in the functional risk module specifically include log processing, performance testing, exception monitoring, exception handling, system emergency handling, etc. Among them, exception handling is related to the exception handling process sub-module in the system data module. In addition, in the functional risk module, system exceptions and system emergency handling often need to be checked before the end of the non-requirement function sub-modules. At this time, it is also related to the exception handling process sub-module. Therefore, the non-requirement function sub-modules in the functional risk module and the exception handling process sub-module in the system data module are functional sub-modules with related functions.

[0112] S32. For functional sub-modules with overlapping functions, delete the failure prediction value of the functional sub-module with a smaller failure prediction value among the functional sub-modules with overlapping functions, and re-determine the failure prediction value of the functional module including the functional sub-module with a smaller failure prediction value.

[0113] Specifically, taking the above "the interface function sub-module in the functional risk module and the exception handling sub-module of the interface in the interface module are functional sub-modules with overlapping functions" as an example. Assume that the failure prediction value of the interface function sub-module is 3, and the failure prediction value of the exception handling sub-module of the interface is 5, that is, the failure prediction value 3 of the interface function sub-module is less than the failure prediction value 5 of the exception handling sub-module of the interface. At this time, delete the failure prediction value of the interface function sub-module in the functional risk module, then the failure prediction value of the functional risk module is updated to the failure prediction value calculated by the data processing sub-model corresponding to the functional risk module, such as 6, minus the above failure prediction value 3 of the interface function sub-module, and the final failure prediction value of the functional risk module is 3.

[0114] S33. For functional sub-modules with related functions, calculate the new failure prediction value of the functional module corresponding to the functional module corresponding to the functional sub-modules with related functions after deleting the failure prediction value of the functional sub-modules with related functions.

[0115] Specifically, the above "the non-requirement function sub-modules in the functional risk module and the exception handling process sub-module in the system data module are functional sub-modules with related functions" is used as an example for illustration.

[0116] Assume that the calculated failure prediction value of the non-requirement function sub-module in the functional risk module is 2, the failure prediction value of the functional risk module is 7, the failure prediction value of the exception handling process sub-module in the system data module is 1, and the failure prediction value of the system data module is 8.

[0117] Then the failure value of the functional risk module 7 - the failure prediction value of the non - required function sub - module 2 = 5, that is, the new failure prediction value of the functional risk module is 5.

[0118] Similarly, the failure prediction value of the system data module 8 - the failure prediction value of the exception handling process sub - module 1 = 7.

[0119] It should be noted that in this embodiment, calculating the new failure prediction value corresponding to the functional module is only to determine which functional module corresponding to the functional sub - module related to the function needs to be updated, rather than immediately updating the failure prediction value of the functional module corresponding to the functional sub - module related to the function.

[0120] S34. Screen out the functional modules with relatively smaller new failure prediction values, and use the new failure prediction value corresponding to the functional module as the failure prediction value of the functional module.

[0121] In this embodiment, still taking the above example for illustration. The functional module with a relatively smaller new failure prediction value is the functional risk module, then update the failure prediction value of the functional risk module to the corresponding new failure prediction value, that is, update it to 5.

[0122] At this time, the failure prediction value of the system data module remains unchanged, still being 8.

[0123] It should be noted that in this embodiment, it is illustrated with the number of related functional sub - modules being 2. If the number of related functional sub - modules is greater than 2, it is necessary to ensure that the failure prediction value of the functional module with a relatively larger new failure prediction value remains unchanged, and the failure prediction value of the functional module whose new failure prediction value is not the largest is updated to the corresponding new failure prediction value.

[0124] It should be noted that in practical applications, when determining the failure prediction value of the functional module, it can be implemented according to steps S32 - S34. In addition, to improve the effect, taking S32 as an example, the process of step S32 can be used as a training process, and then record the result of which failure prediction value of the functional sub - module should be deleted in S32. Subsequently, when using, directly delete the failure prediction value of the functional sub - module that needs to be deleted as recorded.

[0125] Steps S33 - S34 are similar.

[0126] S35. Take the sum of the products of the weight values of each functional module and the corresponding failure prediction values as the failure prediction value of the banking system.

[0127] In this embodiment, if the fault value of the functional module is updated, the new fault prediction value is used when calculating the sum of products; if not, the fault prediction value obtained from the original sub-model is still used for calculation.

[0128] In this embodiment, when calculating the fault prediction value of the banking system, the coincidence and correlation of the functional sub-modules are considered, so that the accuracy of the obtained fault prediction value is higher.

[0129] In another implementation manner of the present invention, referring to Figure 4 , the specific implementation process of step S14 "Based on the fault prediction values of each of the functional sub-modules, determine the target functional sub-module that needs to be repaired for faults, and determine the fault repair rule of the target functional sub-module" is as follows;

[0130] S41. Obtain the number of fault points of each functional sub-module of the functional module obtained based on the data processing sub-model corresponding to the functional module.

[0131] In this embodiment, for the specific implementation process of the number of fault points, please refer to the corresponding description in the above embodiment.

[0132] S42. Use the product of the number of fault points of the functional sub-module and the fault prediction value of the functional module corresponding to the functional sub-module as the fault repair parameter of the functional sub-module.

[0133] Specifically, assume that the above number of fault points is a, and the fault prediction value of the functional module corresponding to the functional sub-module is b, then the fault repair parameter c = ab.

[0134] S43. Screen out the functional sub-modules whose fault repair parameters are greater than the preset repair threshold, and use them as candidate functional sub-modules.

[0135] In this embodiment, the preset repair threshold can be set, such as 0.8, and then screen out the functional sub-modules whose fault repair parameters are greater than 0.8, and use them as candidate functional sub-modules that may need fault repair.

[0136] S44. According to the historical fault average value and the fault prediction value of the candidate functional sub-module, determine the target functional sub-module that needs to be repaired for faults, and determine the fault repair rule of the target functional sub-module.

[0137] In this embodiment, the historical fault average value refers to the average value of the actual fault values in the past few months. The fault prediction value is the value output by the above sub-model.

[0138] Specifically, screen out the candidate functional sub-modules that need to be repaired for faults from the candidate functional sub-modules that may need fault repair, use them as the target functional sub-modules, and determine the fault repair rule of the target functional sub-module.

[0139] Referring to Figure 5 , step S44 may include:

[0140] S51. Group the candidate functional sub - modules to obtain multiple combinations.

[0141] Wherein, each of the combinations includes a preset number of candidate functional sub - modules.

[0142] Specifically, for the candidate functional sub - modules, they can be grouped in pairs, or in combinations of other quantities, such as one candidate functional sub - module as a group, or three candidate functional sub - modules as a group. The specific grouping situation needs to be determined according to the number of candidate functional sub - modules. Among them, when combining, the combination method adopts a random combination method. Taking the pairwise combination as an example, which candidate functional sub - module is combined with which candidate functional sub - module can adopt the random combination method.

[0143] S52. For each of the combinations, calculate the deviation degree of the failure prediction value of the candidate functional sub - modules in the combination relative to the historical failure average value.

[0144] Specifically, the deviation degree = (failure prediction value - historical failure average value) / historical failure average value. The finally calculated deviation degree can be expressed in the form of a percentage, such as the deviation degree is 50%.

[0145] S53. Determine the failure repair rule for adjusting the deviation degree of the candidate functional sub - modules to the set deviation degree, and calculate the new failure prediction value of the bank system in the case where the candidate functional sub - modules in the combination are repaired.

[0146] In this embodiment, an attempt is made to adjust the content of the functional sub - modules. After the adjustment, the new failure prediction value of the bank system is recalculated. If it is less than the preset prediction threshold, it means that this adjustment will make the failure prediction value of the bank system return to normal, indicating that this adjustment is an effective adjustment, and the adjustment plan can be output so that technicians can make adjustments according to this adjustment plan. If the new failure prediction value of the bank system is still greater than the preset prediction threshold, it means that this adjustment has not made the failure prediction value of the bank system return to normal, and the adjustment plan is invalid. At this time, the adjustment plan needs to be adjusted again until the failure prediction value of the bank system returns to normal.

[0147] When making a specific adjustment, the deviation degree can be decreased step by step, such as decreasing by 10% each time. That is, the above - mentioned deviation degree of 50% is adjusted to a deviation degree of 40%. At this time, the set deviation degree is 40%.

[0148] If the original deviation degree is 30%, it should be adjusted to 20%. At this time, the set deviation degree is 20%.

[0149] Taking the candidate functional sub-module as an example of the test process, assume that in the test process, the number of errors in the test cases is 10. If the deviation needs to be adjusted from 50% to 40%, then the number of test cases needs to be reduced by 2, that is, reduced to 8. At this time, reducing the number of test cases by 2, and the final number of errors in the test cases is 8, which is the fault repair rule.

[0150] If taking the pairwise combination of the above-mentioned candidate functional sub-modules as an example, repair the test process according to the above-mentioned fault repair rule, and then repair another candidate functional sub-module according to its corresponding fault repair rule. After the repair is completed, calculate according to the above-mentioned method of calculating the fault prediction value of the banking system, and calculate the new fault prediction value of the banking system. That is to say, when making adjustments, adjust each candidate functional sub-module within the same combination, and then recalculate the fault prediction value of the banking system.

[0151] S54. In the case where the new fault prediction value of the banking system is less than the preset prediction threshold, take the candidate functional sub-module in the combination corresponding to the smallest new fault prediction value as the target functional sub-module, and output the fault repair rule corresponding to the target functional sub-module.

[0152] If there is a situation where the new fault prediction value of the banking system is less than the preset prediction threshold, it means that this adjustment can make the banking system return to normal. The lower the fault prediction value of the banking system, the lower the risk of the banking system. Then take the candidate functional sub-module in the combination corresponding to the smallest new fault prediction value as the target functional sub-module, and output the fault repair rule corresponding to the target functional sub-module.

[0153] When outputting the fault repair rule, use the combination method to output, that is, output the fault repair rules of the candidate functional sub-modules in this combination together. Technicians can control the risk of the system according to the combination to correct the corresponding items and prevent the risk.

[0154] In practical applications, in the case where there is no new fault prediction value of the bank system greater than the preset prediction threshold, it is necessary to reproduce and adjust the magnitude of the setting deviation. For example, after adjusting the above-mentioned deviation of 50% to a deviation of 40%, the fault prediction value of the bank system is non-compliant. At this time, when adjusting the deviation from 40% to 30% and then calculating the fault prediction value of the bank system, if it is still non-compliant, continue to reduce the deviation at a gradient of 10%, that is, adjust it to 20%, and so on until it is adjusted to 0%. If during the adjustment process, the new fault prediction value of the bank system obtained is less than the preset prediction threshold, stop the adjustment, take the candidate functional sub-module in the combination corresponding to the smallest new fault prediction value as the target functional sub-module, and output the fault repair rule corresponding to the target functional sub-module.

[0155] If when adjusted to 0%, the new fault prediction value of the bank system is still greater than the preset prediction threshold, at this time, the combination method can be adjusted and the above steps can be re-executed, or a warning message can be output to enable technicians to perform manual repair.

[0156] In this embodiment, when the risk of the bank system is relatively high, a reasonable correction strategy can be analyzed based on the data conditions of the functional sub-modules and provided to the technicians so that the technicians can perform actual repairs, reduce the risk of the bank system, and ensure the reliable operation of the bank system.

[0157] Optionally, based on the above embodiment of the data processing method, another embodiment of the present invention provides a data processing device applied to a controller in a bank system. The bank system includes multiple functional modules, and the functional module includes at least one functional sub-module; referring to Figure 6 , the data processing device includes:

[0158] A model acquisition module 11, configured to acquire the data processing model of the bank system. The data processing model includes multiple data processing sub-models, and the data processing sub-models correspond to the corresponding functional modules in the bank system;

[0159] A first fault prediction module 12, configured to use the data processing sub-model to perform data processing on the operation data of each functional sub-module of the corresponding functional module, obtain the fault prediction value of the functional sub-module, and calculate the fault prediction value of the functional module based on the fault prediction values of each functional sub-module of the functional module;

[0160] A second fault prediction module 13, configured to determine the weight value of each functional module based on the association relationship between different functional modules, and calculate the fault prediction value of the bank system according to the weight value and the fault prediction value of each functional module;

[0161] A fault repair module 14, configured to, when a fault prediction value of the bank system is greater than a preset prediction threshold, determine a target functional sub-module that needs to be repaired for fault based on the fault prediction values of the respective functional sub-modules, and determine a fault repair rule for the target functional sub-module.

[0162] Further, the first fault prediction module is specifically configured to:

[0163] Obtain the operation data of each functional sub-module of the functional module;

[0164] Use the data processing sub-model corresponding to the functional module to perform data processing on the operation data of each functional sub-module of the functional module to obtain the fault prediction value of the functional sub-module;

[0165] Sum up the fault prediction values of each functional sub-module of the functional module to obtain the fault prediction value of the functional module.

[0166] Further, the second fault prediction module 13 includes:

[0167] A first weight determination sub-module, configured to obtain the total number of fault points of the functional module obtained based on the data processing sub-model corresponding to the functional module, and use the total number of fault points as the initial weight value of the functional module;

[0168] A second weight determination sub-module, configured to correct the initial weight value of the functional module based on the association relationship between different functional modules to obtain the weight value of the functional module;

[0169] A prediction value calculation sub-module, configured to use the sum of the products of the weight values and the fault prediction values of each functional module as the fault prediction value of the bank system.

[0170] Further, the prediction value calculation sub-module may include:

[0171] A relationship determination unit, configured to determine, based on the association relationship between different functional modules, the functional sub-modules with overlapping functions and the functional sub-modules with related functions among different functional modules;

[0172] A first prediction value calculation unit, configured to, for the functional sub-modules with overlapping functions, delete the fault prediction value of the functional sub-module with a smaller fault prediction value among the functional sub-modules with overlapping functions, and re-determine the fault prediction value of the functional module including the functional sub-module with a smaller fault prediction value;

[0173] A second predicted value calculation unit, configured to calculate, for a functional sub-module with related functions, a new fault predicted value corresponding to the functional module after deleting the fault predicted value of the functional sub-module with related functions in the functional module corresponding to the functional sub-module with related functions;

[0174] A predicted value determination unit, configured to screen out functional modules with relatively small new fault predicted values, and use the new fault predicted values corresponding to the functional modules as the fault predicted values of the functional modules;

[0175] A third predicted value calculation unit, configured to use the sum of the products of the weight values of each functional module and the corresponding fault predicted values as the fault predicted value of the banking system.

[0176] Furthermore, the fault repair module 14 includes:

[0177] A point acquisition sub-module, configured to acquire the fault points of each functional sub-module of the functional module obtained based on the data processing sub-model corresponding to the functional module;

[0178] A parameter calculation sub-module, configured to use the product of the fault points of the functional sub-module and the fault predicted value of the functional module corresponding to the functional sub-module as the fault repair parameter of the functional sub-module;

[0179] A screening sub-module, configured to screen out functional sub-modules with fault repair parameters greater than a preset repair threshold and use them as candidate functional sub-modules;

[0180] A rule determination sub-module, configured to determine a target functional sub-module that needs to be repaired according to the historical fault average value and the fault predicted value of the candidate functional sub-module, and determine the fault repair rule of the target functional sub-module.

[0181] Furthermore, the rule determination sub-module includes:

[0182] A grouping unit, configured to group the candidate functional sub-modules to obtain a plurality of combinations; each combination includes a preset number of candidate functional sub-modules;

[0183] A deviation calculation unit, configured to calculate, for each combination, the deviation of the fault predicted value of the candidate functional sub-modules in the combination from the historical fault average value;

[0184] A fourth predicted value calculation unit, configured to determine a fault repair rule for adjusting the deviation of the candidate functional sub-module to a set deviation, and calculate the new fault predicted value of the banking system in the case of repairing the candidate functional sub-modules in the combination;

[0185] A data output unit, configured to, when there is a new fault prediction value of the bank system that is less than the preset prediction threshold, use the candidate functional sub-module in the combination corresponding to the smallest new fault prediction value as the target functional sub-module, and output the fault repair rule corresponding to the target functional sub-module.

[0186] Further, the rule determination sub-module further includes:

[0187] A deviation adjustment unit, configured to, when there is no new fault prediction value of the bank system that is greater than the preset prediction threshold, adjust the magnitude of the set deviation;

[0188] A fourth prediction value calculation unit is further configured to, after the deviation adjustment unit adjusts the magnitude of the set deviation, determine the fault repair rule for adjusting the deviation of the candidate functional sub-module to the set deviation;

[0189] In this embodiment, for each functional module, there is a corresponding data processing sub-model, and thus the fault prediction value of the functional module can be determined according to the corresponding data processing sub-model, which can avoid the situation where the fault value predicted manually based on experience has low accuracy, and improve the accuracy of the fault prediction value. In addition, in the present invention, for different functional modules, corresponding data processing sub-models are configured, making the data processing sub-model more adaptable to the functional module, and further improving the accuracy of fault prediction. Moreover, in the present invention, the functional module includes at least one functional sub-module. When the fault prediction value is greater than the preset threshold, based on the fault prediction values of the respective functional sub-modules, the target functional sub-module that needs to be repaired is determined, and the fault repair rule of the target functional sub-module is determined. That is, the present invention can determine the fault repair rule from a finer-grained level of the target functional sub-module, and can improve the accuracy of determining the fault repair rule compared with the method of determining the fault repair rule manually based on experience.

[0190] It should be noted that for the working processes of the various modules, sub-modules, and units in this embodiment, please refer to the corresponding descriptions in the above embodiments, and will not be elaborated here.

[0191] Optionally, based on the above embodiments of the data processing method and apparatus, another embodiment of the present invention provides an electronic device, including: a memory and a processor;

[0192] Wherein, the memory is used to store a program;

[0193] The processor calls the program and is used to execute the above data processing method.

[0194] In this embodiment, for each functional module, there is a corresponding data processing sub-model, so that the fault prediction value of the functional module can be determined according to the corresponding data processing sub-model, which can avoid the situation of low accuracy of the predicted fault value caused by manual prediction of the fault value according to experience and improve the accuracy of the fault prediction value. In addition, in the present invention, corresponding data processing sub-models are configured for different functional modules, making the data processing sub-model more adaptable to the functional module and further improving the accuracy of fault prediction. Furthermore, in the present invention, the functional module includes at least one functional sub-module. When the fault prediction value is greater than a preset threshold, based on the fault prediction values of the respective functional sub-modules, the target functional sub-module that needs to be repaired is determined, and the fault repair rule of the target functional sub-module is determined. That is, the present invention can determine the fault repair rule from a finer-grained level of the target functional sub-module. Compared with the method of manually determining the fault repair rule according to experience, it can improve the accuracy of determining the fault repair rule.

[0195] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A data processing method, characterized in that, it is applied to a controller in a banking system, the banking system includes a plurality of functional modules, and the functional modules include at least one functional sub-module; the data processing method includes: Obtain the data processing model of the banking system, the data processing model includes a plurality of data processing sub-models, and the data processing sub-models correspond to the corresponding functional modules in the banking system; Use the data processing sub-model to process the operation data of each functional sub-module of the corresponding functional module to obtain the fault prediction value of the functional sub-module, and calculate the fault prediction value of the functional module based on the fault prediction values of each functional sub-module of the functional module; Based on the association relationship between different functional modules, determine the weight value of each functional module, and calculate the fault prediction value of the banking system according to the weight value and fault prediction value of each functional module; When the fault prediction value of the banking system is greater than the preset prediction threshold, based on the fault prediction values of each functional sub-module, determine the target functional sub-module that needs to be repaired, and determine the fault repair rule of the target functional sub-module; Using the data processing sub-model to process the operation data of each functional sub-module of the corresponding functional module to obtain the fault prediction value of the functional sub-module, and calculate the fault prediction value of the functional module based on the fault prediction values of each functional sub-module of the functional module, includes: Obtain the operation data of each functional sub-module of the functional module; Use the data processing sub-model corresponding to the functional module to process the operation data of each functional sub-module of the functional module to obtain the fault prediction value of the functional sub-module; Sum the fault prediction values of each functional sub-module of the functional module to obtain the fault prediction value of the functional module; Based on the association relationship between different functional modules, determine the weight value of each functional module, and calculate the fault prediction value of the banking system according to the weight value and fault prediction value of each functional module, includes: Obtain the total number of fault points of the functional module obtained based on the data processing sub-model corresponding to the functional module, and use the total number of fault points as the initial weight value of the functional module; Based on the association relationship between different functional modules, correct the initial weight value of the functional module to obtain the weight value of the functional module; Take the sum of the products of the weight values and fault prediction values of each functional module as the fault prediction value of the banking system; Based on the fault prediction values of each functional sub-module, determine the target functional sub-module that needs to be repaired, and determine the fault repair rule of the target functional sub-module, includes: Obtain the number of fault points of each functional sub-module of the functional module obtained based on the data processing sub-model corresponding to the functional module; Take the product of the number of fault points of the functional sub-module and the fault prediction value of the functional module corresponding to the functional sub-module as the fault repair parameter of the functional sub-module; Screen out the functional sub - modules whose fault repair parameters are greater than the preset repair threshold, and use them as candidate functional sub - modules; Determine the target functional sub - modules that need fault repair according to the historical fault average value and the fault prediction value of the candidate functional sub - modules, and determine the fault repair rules for the target functional sub - modules.

2. The data processing method according to claim 1, characterized in that, Taking the sum of the products of the weight values of each of the functional modules and the fault prediction value as the fault prediction value of the banking system, includes: Based on the association relationship between different functional modules, determine the functional sub - modules with overlapping functions and the functional sub - modules with related functions among different functional modules; For the functional sub - modules with overlapping functions, delete the fault prediction value of the functional sub - module with a smaller fault prediction value among the functional sub - modules with overlapping functions, and re - determine the fault prediction value of the functional module including the functional sub - module with a smaller fault prediction value; For the functional sub - modules with related functions, calculate the new fault prediction value of the functional module corresponding to the functional sub - modules with related functions after deleting the fault prediction value of the functional sub - modules with related functions; Screen out the functional modules with smaller corresponding new fault prediction values, and use the new fault prediction value corresponding to the functional module as the fault prediction value of the functional module; Taking the sum of the products of the weight values of each of the functional modules and the corresponding fault prediction values as the fault prediction value of the banking system.

3. The data processing method according to claim 1, characterized in that, Determine the target functional sub - modules that need fault repair according to the historical fault average value and the fault prediction value of the candidate functional sub - modules, and determine the fault repair rules for the target functional sub - modules, includes: Group the candidate functional sub - modules to obtain multiple combinations; each combination includes a preset number of candidate functional sub - modules; For each combination, calculate the deviation degree of the fault prediction value of the candidate functional sub - modules in the combination relative to the historical fault average value; Determine the fault repair rules for adjusting the deviation degree of the candidate functional sub - modules to the set deviation degree, and calculate the new fault prediction value of the banking system in the case of repairing the candidate functional sub - modules in the combination; In the case where the new fault prediction value of the banking system is less than the preset prediction threshold, take the candidate functional sub - modules in the combination corresponding to the smallest new fault prediction value as the target functional sub - modules, and output the fault repair rules corresponding to the target functional sub - modules.

4. The data processing method according to claim 3, characterized in that, In the case where there is no new fault prediction value of the banking system greater than the preset prediction threshold, it further includes: Adjust the magnitude of the set deviation, and return to the step of determining the fault repair rule for adjusting the deviation of the candidate functional sub-module to the set deviation until the new fault prediction value of the bank system is less than the preset prediction threshold, then stop. Take the candidate functional sub-module in the combination corresponding to the smallest new fault prediction value as the target functional sub-module, and output the fault repair rule corresponding to the target functional sub-module.

5. A data processing device characterized in that it is applied to a controller in a bank system, the bank system includes a plurality of functional modules, and the functional modules include at least one functional sub-module; the data processing device includes: a model acquisition module, configured to acquire a data processing model of the bank system, the data processing model includes a plurality of data processing sub-models, and the data processing sub-models correspond to the corresponding functional modules in the bank system; a first fault prediction module, configured to use the data processing sub-model to perform data processing on the operation data of each functional sub-module of the corresponding functional module, obtain a fault prediction value of the functional sub-module, and calculate a fault prediction value of the functional module based on the fault prediction values of each functional sub-module of the functional module; a second fault prediction module, configured to determine a weight value of each functional module based on the association relationship between different functional modules, and calculate a fault prediction value of the bank system according to the weight values and fault prediction values of each functional module; a fault repair module, configured to, when the fault prediction value of the bank system is greater than a preset prediction threshold, determine a target functional sub-module that needs to be repaired based on the fault prediction values of each functional sub-module, and determine a fault repair rule for the target functional sub-module; The first fault prediction module is specifically configured to: acquire the operation data of each functional sub-module of the functional module; use the data processing sub-model corresponding to the functional module to perform data processing on the operation data of each functional sub-module of the functional module, and obtain a fault prediction value of the functional sub-module; sum the fault prediction values of each functional sub-module of the functional module to obtain a fault prediction value of the functional module; The second fault prediction module is specifically configured to: acquire the total number of fault points of the functional module obtained based on the data processing sub-model corresponding to the functional module, and use the total number of fault points as the initial weight value of the functional module; correct the initial weight value of the functional module based on the association relationship between different functional modules to obtain the weight value of the functional module; take the sum of the products of the weight values and fault prediction values of each functional module as the fault prediction value of the bank system; The fault repair module is specifically configured to: acquire the number of fault points of each functional sub-module of the functional module obtained based on the data processing sub-model corresponding to the functional module; take the product of the number of fault points of the functional sub-module and the fault prediction value of the functional module corresponding to the functional sub-module as the fault repair parameter of the functional sub-module; Screen out the functional sub-modules whose fault repair parameters are greater than the preset repair threshold, and use them as candidate functional sub-modules; Determine the target functional sub-module that needs to be repaired according to the historical fault average value and the fault prediction value of the candidate functional sub-module, and determine the fault repair rule of the target functional sub-module.

6. An electronic device, Characterized in that, Comprising: A memory and a processor; Wherein, the memory is used for storing programs; The processor calls the program and is used to execute the data processing method according to any one of claims 1-4.

Images