Intelligent processing method, device and equipment for transaction abnormity and storage medium

Abstract

The invention provides an intelligent processing method and device for transaction abnormity, equipment and a storage medium, and the method comprises the steps: detecting whether a transaction state code generated in the processing process of a transaction request is an abnormal transaction state code or not, and if yes, determining whether to carry out the abnormal information reporting or not based on a reporting-free white list; if the abnormal information is reported, carrying out data processing on the abnormal information before reporting, generating abnormal information after desensitization and encryption processing, inputting the desensitization abnormal information into a transaction abnormal identification model, and identifying a first processing decision corresponding to the abnormal information after desensitization and encryption processing; repairing the desensitized and encrypted abnormal information based on the first processing decision, and determining an abnormal processing result; and on the basis of the type of the exception processing result, determining whether a work order is generated for the exception information subjected to desensitization and encryption processing, so that operation and maintenance personnel repair the exception information subjected to desensitization and encryption processing after receiving the work order, thereby improving the exception information processing efficiency.

Description

Technical Field

[0001] The present application relates to the field of computer data processing technology, and in particular to an intelligent processing method, device, equipment and storage medium for transaction anomalies. Background Art

[0002] In modern trading systems, the occurrence of abnormal transactions usually causes the system to be unable to process user requests normally, affecting user experience and trading efficiency. Currently, most trading systems use simple rules to determine abnormal situations and report abnormal information to the backend system for processing. However, these traditional methods usually rely on fixed error codes and predefined rules, and lack the ability to automatically identify new or complex exceptions. Existing technologies have also failed to make full use of intelligent learning (such as machine learning and deep learning) to optimize exception handling, resulting in the inability to respond and handle unknown exceptions in a timely manner. Therefore, how to improve the efficiency of transaction exception handling has become a technical issue that cannot be underestimated. Summary of the invention

[0003] In view of this, the purpose of this application is to provide an intelligent processing method, device, equipment and storage medium for transaction anomalies, perform data processing on abnormal information, ensure that sensitive data will not be leaked, and automatically determine the processing decision of abnormal information through a transaction anomaly identification model, thereby improving the efficiency of transaction anomaly processing. When the abnormal information is not resolved, a work order is automatically generated and assigned to the corresponding operation and maintenance personnel to ensure that the abnormal information can be processed in a timely manner.

[0004] The present application provides an intelligent processing method for transaction anomalies, and the intelligent processing method includes:

[0005] Detecting whether the transaction status code generated during the processing of the transaction request is an abnormal transaction status code, and if so, determining whether to report abnormal information for the abnormal transaction status code based on a preset whitelist of exemptions from reporting;

[0006] If the abnormal information is reported, data processing is performed on the abnormal information to generate desensitized and encrypted abnormal information, the desensitized and encrypted abnormal information is reported and then decrypted to obtain desensitized abnormal information, the desensitized abnormal information is input into a pre-trained transaction abnormality recognition model, and a first processing decision corresponding to the desensitized abnormal information is identified;

[0007] Performing repair processing on the desensitized exception information based on the first processing decision to determine an exception processing result; wherein the types of the exception processing result include unresolved and resolved;

[0008] Based on the type of the exception processing result, it is determined whether to generate a work order for the desensitized exception information, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order.

[0009] In a possible implementation manner, the performing data processing on the exception information to generate the exception information after desensitization and encryption processing includes:

[0010] Detecting whether there is sensitive information in the abnormal information based on regular expressions;

[0011] If not, the abnormal information will not be desensitized;

[0012] If so, data desensitization is performed on the abnormal information to generate desensitized abnormal information, and the desensitized abnormal information is encrypted based on an encryption algorithm to determine the abnormal information after desensitization and encryption.

[0013] In a possible implementation manner, inputting the desensitized abnormal information into a pre-trained transaction abnormality recognition model to identify a first processing decision corresponding to the desensitized abnormal information includes:

[0014] Inputting the desensitized abnormal information into the transaction abnormality identification model, performing information analysis on the desensitized abnormal information based on natural language processing technology, and determining keywords of the desensitized abnormal information;

[0015] Determining the abnormality type of the desensitized abnormal information based on the keyword, and matching the abnormality type with a plurality of reference abnormality types in the transaction abnormality identification model;

[0016] If the match is successful, the reference processing decision of the reference exception type that matches the exception type is used as the first processing decision.

[0017] In a possible implementation manner, determining whether to generate a work order for the desensitized exception information based on the type of the exception processing result, so that the operation and maintenance personnel repair the desensitized exception information after receiving the work order, includes:

[0018] If the type of the exception handling result is unresolved, the desensitized exception information is parsed and processed, and a work order is generated based on the error information in the parsed desensitized exception information;

[0019] The work order is assigned to the corresponding operation and maintenance team based on the error type corresponding to the error information, so that the operation and maintenance personnel in the operation and maintenance team can repair the desensitized exception information after receiving the work order; wherein the error type includes program error type and non-program error type.

[0020] In a possible implementation manner, after determining whether to generate a work order for the desensitized exception information based on the type of the exception processing result so that the operation and maintenance personnel repair the desensitized exception information after receiving the work order, the intelligent processing method further includes:

[0021] When a change in the processing progress of the work order is detected, the processing progress of the work order is pushed in real time based on the WebSocket connection, so that the processing progress of the work order is displayed to the user end in real time.

[0022] In a possible implementation manner, after determining whether to generate a work order for the desensitized exception information based on the type of the exception processing result so that the operation and maintenance personnel repair the desensitized exception information after receiving the work order, the intelligent processing method further includes:

[0023] When it is detected that the processing progress of the work order is in a completed state, a second processing decision of the operation and maintenance personnel to repair the desensitized abnormal information is obtained;

[0024] The desensitized exception information and the second processing decision are input into a transaction anomaly identification model to continuously optimize the transaction anomaly identification model.

[0025] In a possible implementation manner, the transaction anomaly recognition model is trained through the following steps:

[0026] Inputting historical work orders into the long short-term memory model, performing data analysis on the historical work orders, and determining the prediction error type corresponding to the historical work orders;

[0027] The transaction anomaly identification model is iteratively trained based on the loss value between the predicted error type and the actual error type of the historical work order to generate the transaction anomaly identification model.

[0028] The embodiment of the present application also provides an intelligent processing device for transaction anomalies, the intelligent processing device comprising:

[0029] The transaction exception capture and reporting module is used to detect whether the transaction status code generated during the processing of the transaction request is an abnormal transaction status code. If so, it determines whether to report the abnormal information of the abnormal transaction status code based on the preset white list of exemptions from reporting;

[0030] An abnormality type identification module is used for, if the abnormal information is reported, performing data processing on the abnormal information to generate abnormal information after desensitization and encryption, decrypting the abnormal information after desensitization and encryption after reporting to obtain desensitized abnormal information, inputting the desensitized abnormal information into a pre-trained transaction abnormality identification model, and identifying a first processing decision corresponding to the desensitized abnormal information;

[0031] A determination module, configured to perform repair processing on the desensitized exception information based on the first processing decision, and determine an exception processing result; wherein the types of the exception processing result include unresolved and resolved;

[0032] The work order generation module is used to determine whether to generate a work order for the desensitized exception information based on the type of the exception processing result, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order.

[0033] An embodiment of the present application also provides an electronic device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the intelligent processing method for transaction anomalies as described above are performed.

[0034] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the intelligent method for processing transaction anomalies as described above are executed.

[0035] The embodiments of the present application provide an intelligent processing method, device, equipment and storage medium for transaction anomalies. The intelligent processing method includes: detecting whether a transaction status code generated during the processing of a transaction request is an abnormal transaction status code, and if so, determining whether to report abnormal information for the abnormal transaction status code based on a preset whitelist of exemptions from reporting; if the abnormal information is reported, performing data processing on the abnormal information to generate abnormal information after desensitizing and encryption processing, decrypting the abnormal information after desensitizing and encryption processing to obtain desensitized abnormal information after reporting, inputting the desensitized abnormal information into a pre-trained transaction anomaly recognition model, and identifying a first processing decision corresponding to the desensitized abnormal information; repairing the desensitized abnormal information based on the first processing decision to determine an abnormal processing result; wherein the types of the abnormal processing results include unresolved and resolved; and determining whether to generate a work order for the desensitized abnormal information based on the type of the abnormal processing result, so that the operation and maintenance personnel can repair the desensitized abnormal information after receiving the work order. Data processing is performed on abnormal information to ensure that sensitive data will not be leaked. The transaction anomaly identification model can automatically determine the processing decision of abnormal information, which improves the efficiency of transaction anomaly processing. When the abnormal information is not resolved, a work order is automatically generated and assigned to the corresponding operation and maintenance personnel to ensure that the abnormal information can be processed in a timely manner.

[0036] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 A flowchart of an intelligent method for processing transaction anomalies provided in an embodiment of the present application;

[0039] Figure 2 A schematic diagram of an intelligent method for processing transaction anomalies provided in an embodiment of the present application;

[0040] Figure 3 One of the structural schematic diagrams of an intelligent processing device for transaction anomalies provided in an embodiment of the present application;

[0041] Figure 4The second structural diagram of an intelligent processing device for transaction anomalies provided in an embodiment of the present application;

[0042] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application usually described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, each other embodiment obtained by those skilled in the art without making creative work belongs to the scope of protection of the present application.

[0044] First, the application scenarios to which the present application is applicable are introduced. The present application can be applied in the field of computer data processing technology.

[0045] Research has found that in modern trading systems, the occurrence of abnormal transactions usually causes the system to be unable to process user requests normally, affecting user experience and trading efficiency. Currently, most trading systems use simple rules to determine abnormal situations and report abnormal information to the backend system for processing. However, these traditional methods usually rely on fixed error codes and predefined rules, and lack the ability to automatically identify new or complex exceptions. Existing technologies have also failed to make full use of intelligent learning (such as machine learning and deep learning) to optimize exception handling, resulting in the inability to respond and handle unknown exceptions in a timely manner. Therefore, how to improve the efficiency of transaction exception handling has become a technical issue that cannot be underestimated.

[0046] Based on this, the embodiment of the present application provides an intelligent processing method for transaction anomalies, which performs data processing on the abnormal information to ensure that sensitive data will not be leaked. The transaction anomaly identification model can automatically determine the processing decision of the abnormal information, thereby improving the efficiency of transaction anomaly processing. When the abnormal information is not resolved, a work order is automatically generated and assigned to the corresponding operation and maintenance personnel to ensure that the abnormal information can be processed in a timely manner.

[0047] See also Figure 1 , Figure 1 This is a flow chart of an intelligent method for processing transaction anomalies provided by an embodiment of the present application. Figure 1 As shown in , the intelligent processing method provided by the embodiment of the present application includes:

[0048] S101: Detect whether a transaction status code generated during the processing of a transaction request is an abnormal transaction status code. If so, determine whether to report abnormal information for the abnormal transaction status code based on a preset whitelist of exemptions from reporting.

[0049] In this step, it is detected whether the transaction status code generated during the processing of the transaction request is an abnormal transaction status code. If it is an abnormal transaction status code, it is determined whether to report the abnormal transaction status code abnormal information to the back-end system based on the preset whitelist of exemptions from reporting.

[0050] Among them, the back-end system will return a transaction status code to the front-end system when processing a transaction request. The normal transaction code is 'AAAAAAA', and any other code that is not all "A" (such as 'BBBBBBB' or 'AAAAAEE') indicates an exception. If it is an abnormal transaction status code, it will automatically determine whether the current status code needs to be reported based on the pre-configured whitelist of exemptions from reporting. If it is a known exception or an exception code that the user configures not to report, the reporting process is skipped; otherwise, the reporting process is entered.

[0051] Here, the system supports administrators or users to enable or disable the exception reporting function of specific menu items through permission management. For example, administrators can configure errors in certain modules not to be reported. The configuration content is stored in the database or configuration file and supports dynamic updates. Administrators can modify the error code whitelist in real time through the background. When the front end receives the error code, the front end will determine whether to report the exception information based on the current configuration. If the configuration allows, the exception information reporting process will be entered; if the configuration does not allow, the reporting process will be skipped.

[0052] S102: If the abnormal information is reported, data processing is performed on the abnormal information to generate desensitized and encrypted abnormal information, the desensitized and encrypted abnormal information is reported and then decrypted to obtain the desensitized abnormal information, the desensitized abnormal information is input into a pre-trained transaction abnormality recognition model, and a first processing decision corresponding to the desensitized abnormal information is identified.

[0053] In this step, if the abnormal information is reported, data processing is performed on the abnormal information before reporting to generate desensitized and encrypted abnormal information, and the desensitized abnormal information is decrypted after being reported to obtain the desensitized abnormal information, and the desensitized abnormal information is input into the transaction abnormality identification model to identify the first processing decision corresponding to the desensitized abnormal information.

[0054] When reporting an abnormal transaction status code, the abnormal information is constructed in a defined format. The abnormal information includes but is not limited to: error code (such as 'BBBBBBB'), user information (such as user ID, device information, etc.), transaction information (such as transaction number, transaction amount, etc.) and error occurrence time.

[0055] Here, in order to improve the user experience, the system uses the WebSocket real-time push mechanism to ensure that users can receive instant feedback on whether the abnormal information is successfully submitted. For example, after the abnormal information is successfully reported, the user interface will immediately display the prompt "Error reporting successful"; if the submission fails, the corresponding error will be prompted and a retry option will be provided.

[0056] Before reporting abnormal information, sensitive data (such as user identity information, payment information, etc.) will be desensitized to ensure that user privacy is not leaked. Exception information (including desensitized error codes, user information, transaction information and other key fields) will be formatted. Desensitized abnormal information will be reported to the backend system through the REST API.

[0057] In a possible implementation manner, the performing data processing on the abnormal information to generate desensitized abnormal information includes:

[0058] A: Detect whether there is sensitive information in the abnormal information based on regular expressions.

[0059] Here, the received abnormal information may contain the user's personal data, transaction data, etc., and predefined rules (such as regular expressions or custom data checking algorithms) are used to identify whether it contains sensitive data. For example, the ID card number format (\d{18}) or the bank card number format ((\d{16}|\d{19})).

[0060] B: If not, then the abnormal information is not subjected to data desensitization processing; if so, then the abnormal information is subjected to data desensitization processing to generate desensitized abnormal information, and the desensitized abnormal information is encrypted based on an encryption algorithm to determine the desensitized abnormal information.

[0061] Here, if not, the abnormal information is not subjected to data desensitization processing; if so, the abnormal information is subjected to data desensitization processing to generate desensitized abnormal information, and the desensitized abnormal information is encrypted according to the encryption algorithm to determine the desensitized abnormal information.

[0062] Among them, the fields containing sensitive information are desensitized to avoid the leakage of sensitive data. Desensitization includes but is not limited to replacing, hiding or deforming sensitive fields. For example, desensitization is performed on information such as ID card number and bank card number. For example, only the last 4 digits of the ID card number are displayed, and the rest is replaced with "****", and only the last four digits of the bank card number are displayed. Alternatively, "character masking" or "value replacement" is performed on sensitive fields: ID card number: ******1234.

[0063] Here, in addition to desensitization, all sensitive data is also encrypted by encryption algorithms to prevent it from being intercepted during data transmission. Encryption ensures the security of data and prevents external malicious access. Use encryption algorithms (AES) to encrypt sensitive data, encrypt each transmission of sensitive data (even after desensitization), and ensure that it cannot be decrypted externally during transmission. Use a dedicated key management system to store and manage encryption keys, and the use of keys should be limited to authorized applications and users. The encrypted data will be sent to the back-end server through a secure channel (such as HTTPS) to ensure that the data is not leaked during transmission. After receiving the data, the back-end server decrypts the data through the key management system. The decrypted data will be used for anomaly analysis to identify transaction anomalies, system errors or other problems.

[0064] In a possible implementation manner, inputting the desensitized abnormal information into a pre-trained transaction abnormality recognition model to identify a first processing decision corresponding to the abnormal information includes:

[0065] a: Input the desensitized abnormal information into the transaction abnormality identification model, perform information analysis on the desensitized abnormal information based on natural language processing technology, and determine the keywords of the desensitized abnormal information.

[0066] Here, the desensitized abnormal information is input into the transaction abnormality identification model, and the desensitized abnormal information is analyzed according to the natural language processing technology to determine the keywords of the desensitized abnormal information.

[0067] Among them, NLP technology is introduced to perform text classification and analysis on exception information. NLP technology can extract keywords from the exception information, automatically identify the error type, and recommend corresponding solutions for this type of error. For example, if the error message contains "network connection failure", the NLP model can automatically identify the error as network-related and recommend corresponding solutions.

[0068] b: Determine the exception type of the exception information based on the keyword, match the exception type with multiple reference exception types in the transaction exception identification model, and if the match is successful, use the reference processing decision of the reference exception type that matches the exception type as the first processing decision.

[0069] Here, the abnormal type of the abnormal information is determined based on the keyword, and the abnormal type is matched with multiple reference abnormal types in the transaction abnormality identification model. If the match is successful, the reference processing decision of the reference abnormal type that matches the abnormal type is used as the first processing decision. If the match is unsuccessful, a work order is generated based on the desensitized abnormal information, so that the operation and maintenance personnel can repair the desensitized abnormal information after receiving the work order.

[0070] Among them, if the back-end system detects anomalies that are similar or identical to historical problems, it will automatically generate processing suggestions based on historical data. The back-end transmits these processing suggestions (such as page pop-up prompts) to the front-end interface through the REST API, and the front-end prompts the user with solutions in the form of pop-ups.

[0071] In a possible implementation manner, the transaction anomaly recognition model is trained through the following steps:

[0072] i: Input the historical work orders into the long short-term memory model, perform data analysis on the historical work orders, and determine the prediction error type corresponding to the historical work orders.

[0073] Here, historical work order data is extracted from the database regularly. The data of historical work orders includes work order ID, error code, error type, error description, solution, etc. The extraction process is automated and updated periodically to ensure the accuracy and timeliness of the data. Historical work order data is analyzed using machine learning algorithms (such as decision trees, cluster analysis, etc.). These data can identify high-frequency anomalies and error patterns. For example, certain error codes may appear frequently in specific operation scenarios, which can identify these error types and optimize the processing flow.

[0074] ii: Iteratively training the transaction anomaly recognition model based on the loss value between the predicted error type and the actual error type of the historical work order to generate the transaction anomaly recognition model.

[0075] Here, the transaction anomaly recognition model is iteratively trained according to the loss value between the predicted error type and the actual error type of the historical work order to generate a transaction anomaly recognition model.

[0076] For complex anomalies with time series characteristics, the LSTM (Long Short-Term Memory Network) deep learning method is used to perform more accurate anomaly detection. LSTM can improve the detection ability of complex anomalies in transaction data and logs by analyzing time series data.

[0077] S103: Perform repair processing on the desensitized exception information based on the first processing decision to determine an exception processing result; wherein the types of the exception processing result include unresolved and resolved.

[0078] In this step, the desensitized exception information is repaired according to the first processing decision to determine the exception processing result.

[0079] S104: Based on the type of the exception processing result, determine whether to generate a work order for the desensitized exception information, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order.

[0080] In this step, it is determined whether to generate a work order for the desensitized exception information based on the type of exception processing result, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order.

[0081] In a possible implementation manner, determining whether to generate a work order for the desensitized exception information based on the type of the exception processing result, so that the operation and maintenance personnel repair the desensitized exception information after receiving the work order, includes:

[0082] (1): If the type of the exception handling result is unresolved, the desensitized exception information is parsed and processed, and a work order is generated based on the error information in the parsed desensitized exception information.

[0083] Here, if the type of the exception handling result is unresolved, the anonymized exception information is parsed and processed, and a work order is generated based on the error information in the parsed anonymized exception information.

[0084] Among them, the anonymized abnormal information is parsed, and the parsed content includes error code, transaction information, user information, etc. After parsing, the backend will judge the type and severity of the error based on the content of the anonymized abnormal information. A work order will be automatically generated based on the error information of the anonymized abnormal information. Each work order will be assigned a unique work order number and record detailed information such as error code, user information, transaction number, amount, error description, etc. Each work order will be marked with the corresponding priority according to the type and severity of the error.

[0085] (2): Based on the error type corresponding to the error information, the work order is assigned to the corresponding operation and maintenance team, so that the operation and maintenance personnel in the operation and maintenance team can repair the desensitized exception information after receiving the work order; wherein the error type includes program error type and non-program error type.

[0086] Here, the work order is assigned to the corresponding operation and maintenance team according to the error type corresponding to the error information, so that the operation and maintenance personnel in the operation and maintenance team can repair the desensitized abnormal information after receiving the work order.

[0087] The backend automatically assigns tickets to different processing teams based on the error type. For example, if the error is a network problem, the ticket will be assigned to the network operation and maintenance team; if it is a database-related problem, the ticket will be assigned to the database operation and maintenance team; other types of errors are assigned to the default support team. Ticket assignment is automated and executed according to preset rules.

[0088] Here, when generating a work order, the operation and maintenance personnel only have access to the desensitized data, avoiding direct contact with sensitive information.

[0089] In a specific embodiment, after receiving the work order, the operation and maintenance team will analyze the error information to determine the error type. If the cause of the error is a non-program error type, it needs to be transferred to the development team for further processing. If it is a program error type, the work order will be forwarded to the development team for repair. After receiving the work order, the developer uses logs and debugging tools to locate the problem and repair it. After the repair process is completed, the developer will perform unit testing and system verification to ensure that the problem has been resolved and the repair will not introduce new errors.

[0090] In a possible implementation manner, after determining whether to generate a work order for the desensitized exception information based on the type of the exception processing result so that the operation and maintenance personnel repair the desensitized exception information after receiving the work order, the intelligent processing method further includes:

[0091] When a change in the processing progress of the work order is detected, the processing progress of the work order is pushed in real time based on the WebSocket connection, so that the processing progress of the work order is displayed to the user end in real time.

[0092] Here, when a change in the processing progress of the work order is detected, the processing progress of the work order is pushed in real time according to the WebSocket connection, so that the processing progress of the work order can be displayed to the user end in real time.

[0093] In a specific embodiment, when a work order is created, the backend establishes a two-way communication channel with the frontend through WebSocket. The backend can push the status update of the work order processing to the frontend in real time through the WebSocket connection. The backend will push the work order status update to the frontend in real time through WebSocket according to the processing progress of the work order (such as "processing", "pending confirmation", "resolved"). The frontend user can understand the latest status of the work order processing in real time through this push. After the frontend receives the work order status update through WebSocket, it will update the user interface according to the status information. For example, when the frontend receives the "processing" status, the user interface will display "the problem is being processed"; when the status is updated to "resolved", the interface will display "the problem has been resolved".

[0094] In this application, abnormal transaction information is captured automatically, and errors are processed and reported according to predefined rules, work orders are generated, and the work order status is updated in real time. This process is designed to improve the efficiency of abnormal transaction processing, reduce manual intervention, optimize user experience, and speed up the timeliness of problem solving through real-time feedback.

[0095] In a possible implementation manner, after determining whether to generate a work order for the desensitized exception information based on the type of the exception processing result so that the operation and maintenance personnel repair the desensitized exception information after receiving the work order, the intelligent processing method further includes:

[0096] I: When it is detected that the processing progress of the work order is in a completed state, a second processing decision of the operation and maintenance personnel to repair the desensitized abnormal information is obtained.

[0097] Here, when it is detected that the processing progress of the order is in the completed state, a second processing decision is obtained by the operation and maintenance personnel to repair the desensitized abnormal information.

[0098] II: Inputting the desensitized abnormal information and the second processing decision into a transaction abnormality identification model to continuously optimize the transaction abnormality identification model.

[0099] Here, the desensitized abnormal information and the second processing decision are input into the transaction abnormality identification model to continuously optimize the transaction abnormality identification model.

[0100] Among them, every time the operation and maintenance personnel handle a new abnormal transaction, the system will ask them to feedback the solution. These solutions will be used as input data for the transaction anomaly identification model for system self-learning. The machine learning model is continuously optimized based on the feedback from the operation and maintenance personnel. For example, if certain errors occur frequently and the operation and maintenance personnel adopt a specific solution, the system will add this solution to the solution library and increase its priority level in similar scenarios. Through this feedback mechanism, the system can gradually improve its automatic processing capabilities for new anomalies. The system will establish standardized evaluation indicators (such as accuracy in solving problems, processing time, error rate, etc.) to measure the optimization effect. For newly added solutions, the system will perform performance verification to ensure that no new errors are introduced. Through continuous monitoring and data analysis, the accuracy and robustness of the optimized system are ensured.

[0101] In this application, the transaction anomaly identification model not only simply matches the existing solutions, but also continuously optimizes the solution library. If no solution can be matched, the operation and maintenance personnel will provide a new solution based on the actual situation and feed it back to the transaction anomaly identification model. The transaction anomaly identification model will self-learn based on the effect of the solution (whether the problem is solved, the timeliness of the processing, etc.), optimize the solution library, and improve the accuracy and efficiency of handling similar problems in the future.

[0102] For further information, see Figure 2 , Figure 2 A schematic diagram of an intelligent method for processing transaction anomalies provided in an embodiment of the present application. Figure 2 As shown, the transaction status code is received, and the transaction status code is detected to see if it is an abnormal transaction status code. If so, the normal transaction process is continued. If not, it is determined whether to report abnormal information for the abnormal transaction status code. If so, data processing is performed on the abnormal information to generate abnormal information after desensitization and encryption. The desensitized abnormal information is input into the transaction abnormality identification model to identify the first processing decision. It is detected whether the result of the repair process according to the first processing decision is a solution. If not, work order information is generated according to the desensitized abnormal information, and sent to the corresponding operation and maintenance personnel for repair processing according to the error type in the work order information, and the processing progress of the work order is pushed in real time. When the processing progress of the work order is completed, the second processing decision of the operation and maintenance personnel to repair the desensitized abnormal information, the desensitized abnormal information, and the second processing decision are input into the transaction abnormality identification model, so that the transaction abnormality identification model can be continuously optimized.

[0103] The beneficial effects of this application are as follows: Intelligent capture of anomalies: Automatically identify transaction anomalies through machine learning technology, and provide real-time processing and feedback. Abnormal transaction self-learning and optimization: Use historical transaction data to optimize anomaly identification rules and improve the ability to handle unknown anomalies. Intelligent work order classification and processing: Automatically classify and assign error information to the corresponding operation and maintenance team, and optimize the processing path through machine learning. Data desensitization and privacy protection: Ensure that sensitive information is not leaked during the exception handling process and enhance user privacy protection. Real-time feedback mechanism: Provide real-time feedback on the progress of exception handling to improve user experience.

[0104] An embodiment of the present application provides an intelligent processing method for transaction anomalies, the intelligent processing method comprising: detecting whether a transaction status code generated during the processing of a transaction request is an abnormal transaction status code, and if so, determining whether to report abnormal information for the abnormal transaction status code based on a preset whitelist of exemptions from reporting; if the abnormal information is reported, performing data processing on the abnormal information to generate abnormal information after desensitizing and encryption processing, decrypting the abnormal information after desensitizing and encryption processing to obtain desensitized abnormal information after reporting, inputting the desensitized abnormal information into a pre-trained transaction anomaly recognition model, and identifying a first processing decision corresponding to the desensitized abnormal information; repairing the desensitized abnormal information based on the first processing decision, and determining an abnormal processing result; wherein the types of the abnormal processing results include unresolved and resolved; based on the type of the abnormal processing result, determining whether to generate a work order for the desensitized abnormal information, so that the operation and maintenance personnel can repair the desensitized abnormal information after receiving the work order. Data processing is performed on abnormal information to ensure that sensitive data will not be leaked. The transaction anomaly identification model can automatically determine the processing decision of abnormal information, which improves the efficiency of transaction anomaly processing. When the abnormal information is not resolved, a work order is automatically generated and assigned to the corresponding operation and maintenance personnel to ensure that the abnormal information can be processed in a timely manner.

[0105] See also Figure 3 , Figure 4 , Figure 3 One of the structural schematic diagrams of an intelligent processing device for transaction anomalies provided in an embodiment of the present application; Figure 4 This is a second structural diagram of an intelligent processing device for transaction anomalies provided in an embodiment of the present application. Figure 3 As shown in , the intelligent processing device 300 includes:

[0106] The transaction exception capture and reporting module 310 is used to detect whether the transaction status code generated during the processing of the transaction request is an abnormal transaction status code. If so, it determines whether to report the abnormal information of the abnormal transaction status code based on a preset white list of exemptions from reporting;

[0107] The abnormal type identification module 320 is used for, if the abnormal information is reported, performing data processing on the abnormal information to generate abnormal information after desensitization and encryption, decrypting the abnormal information after desensitization and encryption to obtain desensitized abnormal information after reporting, inputting the desensitized abnormal information into a pre-trained transaction abnormality identification model, and identifying a first processing decision corresponding to the desensitized abnormal information;

[0108] A determination module 330 is used to perform repair processing on the desensitized exception information based on the first processing decision, and determine an exception processing result; wherein the types of the exception processing result include unresolved and resolved;

[0109] The work order generation module 340 is used to determine whether to generate a work order for the desensitized exception information based on the type of the exception processing result, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order.

[0110] Further, when the abnormality type identification module 320 is used to perform data processing on the abnormality information to generate the abnormality information after desensitization and encryption, the abnormality type identification module 320 is specifically used to:

[0111] Detecting whether there is sensitive information in the abnormal information based on regular expressions;

[0112] If not, the abnormal information will not be desensitized;

[0113] If so, data desensitization is performed on the abnormal information to generate desensitized abnormal information, and the desensitized abnormal information is encrypted based on an encryption algorithm to determine the abnormal information after desensitization and encryption.

[0114] Furthermore, when the abnormality type identification module 320 is used to input the desensitized abnormality information into the pre-trained transaction abnormality identification model and identify the first processing decision corresponding to the desensitized abnormality information, the abnormality type identification module 320 is specifically used to:

[0115] Inputting the desensitized abnormal information into the transaction abnormality identification model, performing information analysis on the desensitized abnormal information based on natural language processing technology, and determining keywords of the desensitized abnormal information;

[0116] Determining the abnormality type of the desensitized abnormal information based on the keyword, and matching the abnormality type with a plurality of reference abnormality types in the transaction abnormality identification model;

[0117] If the match is successful, the reference processing decision of the reference exception type that matches the exception type is used as the first processing decision.

[0118] Further, when the work order generation module 340 is used to determine whether to generate a work order for the desensitized exception information based on the type of the exception processing result, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order, the work order generation module 340 is specifically used to:

[0119] If the type of the exception handling result is unresolved, the desensitized exception information is parsed and processed, and a work order is generated based on the error information in the parsed desensitized exception information;

[0120] The work order is assigned to the corresponding operation and maintenance team based on the error type corresponding to the error information, so that the operation and maintenance personnel in the operation and maintenance team can repair the desensitized exception information after receiving the work order; wherein the error type includes program error type and non-program error type.

[0121] Further, such as Figure 4 As shown, the intelligent processing device 300 further includes a processing progress push module 350, and the processing progress push module 350 is used to:

[0122] When a change in the processing progress of the work order is detected, the processing progress of the work order is pushed in real time based on the WebSocket connection, so that the processing progress of the work order is displayed to the user end in real time.

[0123] Further, such as Figure 4 As shown, the intelligent processing device 300 further includes an intelligent learning module 360, and the intelligent learning module 360 ​​is used to:

[0124] When it is detected that the processing progress of the work order is in a completed state, a second processing decision of the operation and maintenance personnel to repair the desensitized abnormal information is obtained;

[0125] The desensitized exception information and the second processing decision are input into a transaction anomaly identification model to continuously optimize the transaction anomaly identification model.

[0126] Further, such as Figure 4 As shown, the intelligent processing device 300 further includes a model training module 370, and the model training module 370 is used to:

[0127] Inputting historical work orders into the long short-term memory model, performing data analysis on the historical work orders, and determining the prediction error type corresponding to the historical work orders;

[0128] The transaction anomaly identification model is iteratively trained based on the loss value between the predicted error type and the actual error type of the historical work order to generate the transaction anomaly identification model.

[0129] The embodiment of the present application provides an intelligent processing device for transaction anomalies, the intelligent processing device comprising: a transaction anomaly capture and reporting module, used to detect whether a transaction status code generated during the processing of a transaction request is an abnormal transaction status code, and if so, whether to report abnormal information for the abnormal transaction status code based on a preset whitelist of exemptions from reporting; an abnormal type identification module, used to perform data processing on the abnormal information if the abnormal information is reported, generate abnormal information after desensitizing and encryption processing, decrypt the abnormal information after desensitizing and encryption processing after reporting to obtain desensitized abnormal information, input the desensitized abnormal information into a pre-trained transaction anomaly identification model, and identify a first processing decision corresponding to the desensitized abnormal information; a determination module, used to perform repair processing on the desensitized abnormal information based on the first processing decision, and determine an abnormal processing result; wherein the types of the abnormal processing results include unresolved and resolved; a work order generation module, used to determine whether to generate a work order for the desensitized abnormal information based on the type of the abnormal processing result, so that the operation and maintenance personnel can repair the desensitized abnormal information after receiving the work order. Data processing is performed on abnormal information to ensure that sensitive data will not be leaked. The transaction anomaly identification model can automatically determine the processing decision of abnormal information, which improves the efficiency of transaction anomaly processing. When the abnormal information is not resolved, a work order is automatically generated and assigned to the corresponding operation and maintenance personnel to ensure that the abnormal information can be processed in a timely manner.

[0130] See also Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 5 As shown in , the electronic device 500 includes a processor 510 , a memory 520 and a bus 530 .

[0131] The memory 520 stores machine-readable instructions executable by the processor 510. When the electronic device 500 is running, the processor 510 communicates with the memory 520 via the bus 530. When the machine-readable instructions are executed by the processor 510, the above-mentioned Figure 1 as well as Figure 2 The steps of the intelligent method for handling transaction anomalies in the method embodiment shown in the figure, and the specific implementation method thereof can be referred to the method embodiment, which will not be described in detail here.

[0132] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the computer program can execute the above-mentioned Figure 1 as well as Figure 2 The steps of the intelligent method for handling transaction anomalies in the method embodiment shown in the figure, and the specific implementation method thereof can be referred to the method embodiment, which will not be described in detail here.

[0133] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0134] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0135] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0136] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0137] 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 non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application can essentially be embodied in the form of a software product, or in other words, the part that contributes to the prior art or the part of the technical solution. 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 method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0138] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application 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 in the present application, 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 application, and should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. An intelligent method for processing transaction anomalies, characterized in that: The intelligent processing method comprises: Detecting whether the transaction status code generated during the processing of the transaction request is an abnormal transaction status code, and if so, determining whether to report abnormal information for the abnormal transaction status code based on a preset whitelist of exemptions from reporting; If the abnormal information is reported, data processing is performed on the abnormal information to generate desensitized and encrypted abnormal information, the desensitized and encrypted abnormal information is reported and then decrypted to obtain desensitized abnormal information, the desensitized abnormal information is input into a pre-trained transaction abnormality recognition model, and a first processing decision corresponding to the desensitized abnormal information is identified; Performing repair processing on the desensitized exception information based on the first processing decision to determine an exception processing result; wherein the types of the exception processing result include unresolved and resolved; Based on the type of the exception processing result, it is determined whether to generate a work order for the desensitized exception information, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order.

2. The intelligent processing method according to claim 1, characterized in that: The performing data processing on the abnormal information to generate the abnormal information after desensitization and encryption processing includes: Detecting whether there is sensitive information in the abnormal information based on regular expressions; If not, the abnormal information will not be desensitized; If so, data desensitization is performed on the abnormal information to generate desensitized abnormal information, and the desensitized abnormal information is encrypted based on an encryption algorithm to determine the abnormal information after desensitization and encryption.

3. The intelligent processing method according to claim 1, characterized in that: The step of inputting the desensitized abnormal information into a pre-trained transaction abnormality recognition model to identify a first processing decision corresponding to the desensitized abnormal information includes: Inputting the desensitized abnormal information into the transaction abnormality identification model, performing information analysis on the desensitized abnormal information based on natural language processing technology, and determining keywords of the desensitized abnormal information; Determining the abnormality type of the desensitized abnormal information based on the keyword, and matching the abnormality type with a plurality of reference abnormality types in the transaction abnormality identification model; If the match is successful, the reference processing decision of the reference exception type that matches the exception type is used as the first processing decision.

4. The intelligent processing method according to claim 1, characterized in that: The determining whether to generate a work order for the desensitized exception information based on the type of the exception processing result, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order, includes: If the type of the exception handling result is unresolved, the desensitized exception information is parsed and processed, and a work order is generated based on the error information in the parsed desensitized exception information; The work order is assigned to the corresponding operation and maintenance team based on the error type corresponding to the error information, so that the operation and maintenance personnel in the operation and maintenance team can repair the desensitized exception information after receiving the work order; wherein the error type includes program error type and non-program error type.

5. The intelligent processing method according to claim 1, characterized in that: After determining whether to generate a work order for the desensitized exception information based on the type of the exception processing result, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order, the intelligent processing method further includes: When a change in the processing progress of the work order is detected, the processing progress of the work order is pushed in real time based on the WebSocket connection, so that the processing progress of the work order is displayed to the user end in real time.

6. The intelligent processing method according to claim 1, characterized in that: After determining whether to generate a work order for the desensitized exception information based on the type of the exception processing result, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order, the intelligent processing method further includes: When it is detected that the processing progress of the work order is in a completed state, a second processing decision of the operation and maintenance personnel to repair the desensitized abnormal information is obtained; The desensitized exception information and the second processing decision are input into a transaction anomaly identification model to continuously optimize the transaction anomaly identification model.

7. The intelligent processing method according to claim 1, characterized in that: The transaction anomaly recognition model is trained by the following steps: Inputting historical work orders into the long short-term memory model, performing data analysis on the historical work orders, and determining the prediction error type corresponding to the historical work orders; The transaction anomaly identification model is iteratively trained based on the loss value between the predicted error type and the actual error type of the historical work order to generate the transaction anomaly identification model.

8. An intelligent device for processing transaction anomalies, characterized in that: The intelligent processing device comprises: The transaction exception capture and reporting module is used to detect whether the transaction status code generated during the processing of the transaction request is an abnormal transaction status code. If so, it determines whether to report the abnormal information of the abnormal transaction status code based on the preset white list of exemptions from reporting; An abnormality type identification module is used for, if the abnormal information is reported, performing data processing on the abnormal information to generate abnormal information after desensitization and encryption, decrypting the abnormal information after desensitization and encryption after reporting to obtain desensitized abnormal information, inputting the desensitized abnormal information into a pre-trained transaction abnormality identification model, and identifying a first processing decision corresponding to the desensitized abnormal information; A determination module, configured to perform repair processing on the desensitized exception information based on the first processing decision, and determine an exception processing result; wherein the types of the exception processing result include unresolved and resolved; The work order generation module is used to determine whether to generate a work order for the desensitized exception information based on the type of the exception processing result, so that the operation and maintenance personnel can repair the desensitized exception information after receiving the work order.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the intelligent processing method for transaction anomalies as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the intelligent method for processing transaction anomalies as claimed in any one of claims 1 to 7 are executed.

Images