Error processing method and device for long link transaction

Abstract

The present invention provides an error processing method and device for long-link transactions, which relates to the field of artificial intelligence technology and can be used in the financial field or other fields. The method includes: obtaining transaction messages of multiple transaction nodes according to the received transaction initiation request; analyzing and processing the transaction messages using a pre-established analysis model to obtain an analysis result; if the analysis result is a transaction error, the transaction is determined to be an abnormal transaction, and the abnormal transaction is automatically repaired. The present invention collects and analyzes the transaction messages on each node of the long-link transaction to determine the transaction message status, thereby providing a rapid positioning capability for transaction errors, analyzing the errors, providing repair rules for the error types, and realizing automatic repair.

Description

Technical Field

[0001] The present invention relates to the technical field of long link transactions, and in particular to an error processing method and device for long link transactions. Background Art

[0002] With the development of IT technology and the increasing complexity of business logic, business transaction processes are becoming longer and longer. A business often requires the cooperation of multiple systems to complete. The increased complexity of transaction processes and system interactions has also brought some problems in error location and analysis. For the stability of long-link transactions, existing technologies are relatively limited, mainly focusing on establishing a pre-monitoring network based on some relatively simple technical indicators. There is no method for rapid location, error analysis and self-repair of real business transaction failures. Summary of the invention

[0003] In view of the problems existing in the prior art, the main purpose of the embodiments of the present invention is to provide a method and device for error processing of long-link transactions, so as to realize real-time judgment of transaction message status, locate transaction errors and automatically repair them.

[0004] In order to achieve the above object, an embodiment of the present invention provides an error processing method for long link transactions, the method comprising:

[0005] According to the received transaction initiation request, the transaction messages of multiple transaction nodes are obtained;

[0006] Analyze and process the transaction message using a pre-established analysis model to obtain an analysis result;

[0007] If it is learned that the analysis result is a transaction error, the transaction is determined to be an abnormal transaction, and the abnormal transaction is automatically repaired.

[0008] Optionally, in an embodiment of the present invention, the analysis model is pre-established in the following manner:

[0009] Acquire historical messages whose analysis results show that the transaction is normal, extract a first historical message feature from the historical messages of the normal transaction, and use the first historical message feature as a first training sample;

[0010] The first training sample is input into a first initial neural network model, and the first initial neural network model is trained to obtain the analysis model.

[0011] Optionally, in an embodiment of the present invention, the analyzing and processing the transaction message using a pre-established analysis model to obtain the analysis result includes:

[0012] Extracting transaction message features from the transaction message; wherein the transaction message features include: message type, message length and transaction node sequence number;

[0013] The transaction message features are input into a pre-established analysis model to obtain an analysis result; wherein the analysis result includes a transaction node name.

[0014] Optionally, in an embodiment of the present invention, acquiring transaction messages of multiple transaction nodes according to the received transaction initiation request includes:

[0015] According to the received transaction initiation request, a unique serial number of the transaction is determined, and according to the unique serial number, transaction messages of multiple transaction nodes are obtained.

[0016] Optionally, in an embodiment of the present invention, obtaining transaction messages of multiple transaction nodes according to the unique serial number includes:

[0017] According to the unique serial number, a transaction message corresponding to the unique serial number is obtained from multiple transaction nodes in real time.

[0018] Optionally, in an embodiment of the present invention, if it is learned that the analysis result is a transaction error, determining the transaction as an abnormal transaction, and automatically repairing the abnormal transaction includes:

[0019] If it is learned that the analysis result is a transaction error, the transaction is determined to be an abnormal transaction, and the error type of the abnormal transaction is determined according to a pre-established classification model;

[0020] A corresponding preset repair rule is determined according to the error type, and the abnormal transaction is automatically repaired according to the preset repair rule.

[0021] Optionally, in one embodiment of the present invention, the classification model is pre-established in the following manner:

[0022] Acquire historical messages whose analysis results are transaction errors and corresponding error types, extract second historical message features from the historical messages of transaction errors, and use the second historical message features and the corresponding error types as second training samples;

[0023] The second training sample is input into a second initial neural network model, and the second initial neural network model is trained to obtain the classification model.

[0024] The embodiment of the present invention further provides an error processing device for long link transactions, the device comprising:

[0025] A transaction message module is used to obtain transaction messages of multiple transaction nodes according to the received transaction initiation request;

[0026] An analysis result module is used to analyze and process the transaction message using a pre-established analysis model to obtain an analysis result;

[0027] The automatic repair module is used to determine the transaction as an abnormal transaction and automatically repair the abnormal transaction if it is learned that the analysis result is a transaction error.

[0028] Optionally, in one embodiment of the present invention, the device also includes an analysis model module, which is used to obtain historical messages whose analysis results are normal transactions, and extract a first historical message feature from the historical messages with normal transactions, and use the first historical message feature as a first training sample; input the first training sample into a first initial neural network model, train the first initial neural network model, and obtain the analysis model.

[0029] Optionally, in an embodiment of the present invention, the analysis result module includes:

[0030] A feature extraction unit, used to extract transaction message features from the transaction message; wherein the transaction message features include: message type, message length and transaction node sequence number;

[0031] The analysis result unit is used to input the transaction message characteristics into a pre-established analysis model to obtain an analysis result; wherein the analysis result includes a transaction node name.

[0032] Optionally, in one embodiment of the present invention, the transaction message module is further used to determine a unique serial number of the transaction based on the received transaction initiation request, and obtain transaction messages of multiple transaction nodes based on the unique serial number.

[0033] Optionally, in an embodiment of the present invention, the transaction message module is further used to obtain transaction messages corresponding to the unique serial number from multiple transaction nodes in real time according to the unique serial number.

[0034] Optionally, in an embodiment of the present invention, the automatic repair module includes:

[0035] an error type unit, configured to determine the transaction as an abnormal transaction if it is learned that the analysis result is a transaction error, and determine the error type of the abnormal transaction according to a pre-established classification model;

[0036] The automatic repair unit is used to determine a corresponding preset repair rule according to the error type, and automatically repair the abnormal transaction according to the preset repair rule.

[0037] Optionally, in one embodiment of the present invention, the device also includes a classification model module, which is used to obtain historical messages whose analysis results are transaction errors and corresponding error types, extract second historical message features from the historical messages of transaction errors, and use the second historical message features and the corresponding error types as second training samples; input the second training samples into a second initial neural network model, train the second initial neural network model, and obtain the classification model.

[0038] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the program.

[0039] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for executing the above method.

[0040] The present invention collects transaction messages on each node of a long-link transaction, processes and analyzes the collected messages, and determines the status of the transaction messages, thereby providing the ability to quickly locate transaction errors during and after the event, analyzes the errors, and provides repair rules for the error types to achieve automatic repair. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 This is a flow chart of an error processing method for a long link transaction according to an embodiment of the present invention;

[0043] Figure 2 A flowchart of establishing an analysis model in an embodiment of the present invention;

[0044] Figure 3 A flowchart of obtaining analysis results in an embodiment of the present invention;

[0045] Figure 4 A flow chart of establishing a classification model in an embodiment of the present invention;

[0046] Figure 5 This is a flow chart of automatic error repair in an embodiment of the present invention;

[0047] Figure 6 This is a schematic diagram of the interaction of a system using the error processing method for long link transactions in an embodiment of the present invention;

[0048] Figure 7 It is a message analysis flow chart of a system in which an error processing method for long link transactions is applied in an embodiment of the present invention;

[0049] Figure 8 It is a flowchart of error processing in a system of an error processing method for applying a long link transaction in an embodiment of the present invention;

[0050] Fig. 9 This is a schematic diagram of the structure of an error processing device for long link transactions according to an embodiment of the present invention;

[0051] Fig.10 It is a structural schematic diagram of an error processing device for long link transactions in a specific embodiment of the present invention;

[0052] Fig.11 This is a schematic diagram of the structure of the analysis result module in an embodiment of the present invention;

[0053] Fig.12 It is a structural schematic diagram of an error processing device for long link transactions in another specific embodiment of the present invention;

[0054] Fig.13 This is a schematic diagram of the structure of an automatic repair module in an embodiment of the present invention;

[0055] Fig.14 The figure is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The embodiments of the present invention provide an error processing method and device for long link transactions, which relate to the field of artificial intelligence technology and can be used in the financial field or other technical fields. It should be noted that the error processing method and device for long link transactions of the present invention can be used in the financial field, and can also be used in any field other than the financial field. The application field of the error processing method and device for long link transactions of the present invention is not limited.

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] like Figure 1 The flowchart of the error processing method of a long link transaction according to the embodiment of the present invention is shown. The execution subject of the error processing method of a long link transaction provided by the embodiment of the present invention includes but is not limited to a computer. The method shown in the figure includes:

[0059] Step S1, acquiring transaction messages of multiple transaction nodes according to the received transaction initiation request.

[0060] Among them, a transaction initiation request sent by a customer is received, and a unique serial number is assigned to the transaction. Specifically, the unique serial number can be set by random generation or other methods. Each transaction has a unique serial number. According to the unique serial number, the transaction message corresponding to the unique serial number can be obtained from multiple transaction nodes in the long link transaction. Specifically, the transaction message includes the transaction request message and the transaction return message on each transaction node.

[0061] Step S2: Analyze and process the transaction message using the pre-established analysis model to obtain an analysis result.

[0062] Among them, a certain number of messages with normal transaction analysis results can be collected first, and message features can be extracted as training samples, such as message type, message length, json data dictionary, transaction node sequence number, etc., and then a three-layer neural network can be established using the BP algorithm. The neural network can be trained and optimized using the training samples to obtain an analysis model.

[0063] Furthermore, the transaction message features are extracted from the transaction message, and the transaction message features are input into the analysis model for analysis and processing. The obtained analysis results include whether the transaction status is normal. If the transaction is normal, the analysis result is a normal transaction. If the transaction is abnormal, the analysis result is a transaction error. In addition, the analysis result also includes the transaction node name. If the transaction is an error, the abnormal transaction node can be located according to the transaction node name.

[0064] Step S3: If it is learned that the analysis result is a transaction error, the transaction is determined to be an abnormal transaction, and the abnormal transaction is automatically repaired.

[0065] If the analysis result is a transaction error, it means that the transaction is an abnormal transaction. The error type of the abnormal transaction, such as the timeout type, is determined by using a pre-established classification model.

[0066] Furthermore, the error type can be manually pre-set and the corresponding data set can be prepared. The historical messages with the analysis result of transaction error can be used as the data set. The message features of the historical messages are extracted, and a three-layer neural network is built using the k-means algorithm with the message features as input and the preset error type and unknown exception as output. The neural network is trained and adjusted to obtain a classification model.

[0067] Different error types correspond to different preset repair rules, and the preset repair rules may be manual modification, etc. For example, if the error type is a timeout type, the preset repair rule may be transaction re-submission. If the error type is a complex type, the preset repair rule may be manual processing, thereby completing the automatic repair of abnormal transactions.

[0068] As an embodiment of the present invention, Figure 2 As shown, the analysis model is pre-established in the following manner:

[0069] Step S21, obtaining historical messages whose analysis results show that the transaction is normal, extracting a first historical message feature from the historical messages of the normal transaction, and using the first historical message feature as a first training sample;

[0070] Step S22: input the first training sample into a first initial neural network model, train the first initial neural network model, and obtain the analysis model.

[0071] Among them, a certain number of historical messages with normal transaction analysis results can be collected first, and the first historical message features can be extracted as the first training sample, such as message type, message length, json data dictionary, transaction node sequence number, etc., and then a first initial neural network model can be established using the BP algorithm, which can be a three-layer neural network model. The first training sample is used to train and optimize the first neural network, thereby obtaining an analysis model.

[0072] As an embodiment of the present invention, Figure 3 As shown, the transaction message is analyzed and processed using a pre-established analysis model, and the analysis results obtained include:

[0073] Step S31, extracting transaction message features from the transaction message; wherein the transaction message features include: message type, message length and transaction node sequence number;

[0074] Step S32, inputting the transaction message features into a pre-established analysis model to obtain an analysis result; wherein the analysis result includes a transaction node name.

[0075] Among them, the transaction message features are extracted from the transaction message, such as message type, message length, json data dictionary, transaction node sequence number, etc. The transaction message features are input into the analysis model for analysis and processing. The analysis results obtained include whether the transaction status is normal. If the transaction is normal, the analysis result is a normal transaction. If the transaction is abnormal, the analysis result is a transaction error. In addition, the analysis result also includes the transaction node name. If the transaction is an error, the abnormal transaction node can be located according to the transaction node name.

[0076] As an embodiment of the present invention, Figure 4 As shown, the classification model is pre-established in the following way:

[0077] Step S41, obtaining historical messages with analysis results of transaction errors and corresponding error types, extracting second historical message features from the historical messages of transaction errors, and using the second historical message features and the corresponding error types as second training samples;

[0078] Step S42: input the second training sample into a second initial neural network model, train the second initial neural network model, and obtain the classification model.

[0079] Among them, the error type can be manually pre-set and the corresponding data set can be prepared. The historical message with the analysis result of transaction error can be used as the data set. The second historical message feature of the historical message of transaction error is extracted, and the second training sample is obtained by using the k-means algorithm according to the second historical message feature as input and the preset error type and unknown exception as output. A three-layer neural network is pre-built as the second initial neural network model, and the second initial neural network is trained and adjusted to obtain a classification model.

[0080] As an embodiment of the present invention, the step of obtaining transaction messages of multiple transaction nodes according to the received transaction initiation request includes:

[0081] According to the received transaction initiation request, a unique serial number of the transaction is determined, and according to the unique serial number, transaction messages of multiple transaction nodes are obtained.

[0082] Among them, a transaction initiation request sent by a customer is received, and a unique serial number is assigned to the transaction. Specifically, the unique serial number can be set by random generation or other methods. Each transaction has a unique serial number. According to the unique serial number, the transaction message corresponding to the unique serial number can be obtained from multiple transaction nodes in the long link transaction.

[0083] In this embodiment, acquiring transaction messages of multiple transaction nodes according to the unique serial number includes: acquiring transaction messages corresponding to the unique serial number from multiple transaction nodes in real time according to the unique serial number.

[0084] Among them, since each transaction has a unique serial number, based on the unique serial number, the transaction message can be obtained in real time in multiple transaction nodes of the long link transaction, ensuring the real-time nature of the message analysis. Specifically, the transaction message corresponding to the unique serial number is obtained.

[0085] As an embodiment of the present invention, Figure 5As shown, if it is known that the analysis result is a transaction error, the transaction is determined to be an abnormal transaction, and the abnormal transaction is automatically repaired, including:

[0086] Step S51: if it is learned that the analysis result is a transaction error, the transaction is determined to be an abnormal transaction, and the error type of the abnormal transaction is determined according to a pre-established classification model.

[0087] Among them, the error type can be manually pre-set and the corresponding data set can be prepared. The historical messages with the analysis result of transaction error can be used as the data set. The message features of the historical messages are extracted, and a three-layer neural network is built using the k-means algorithm according to the message features as input, the preset error type and the unknown exception as output. The neural network is trained and adjusted to obtain a classification model.

[0088] Step S52, determining a corresponding preset repair rule according to the error type, and automatically repairing the abnormal transaction according to the preset repair rule.

[0089] Different error types correspond to different preset repair rules, and the preset repair rules may be manual modification, etc. For example, if the error type is a timeout type, the preset repair rule may be transaction re-submission. If the error type is a complex type, the preset repair rule may be manual processing, thereby completing the automatic repair of abnormal transactions.

[0090] In a specific embodiment of the present invention, Figure 6 The figure shows an interactive schematic diagram of a system that applies an error processing method for long-link transactions in an embodiment of the present invention. The system shown in the figure includes: a client, multiple transaction nodes (transaction systems), a message analysis module (including an analysis model), an error processing module (including a classification model) and a console.

[0091] The specific error handling process is as follows: according to the transaction initiation request sent by the client, a unique serial number is set for the transaction to facilitate the location and tracking of each transaction. The messages collected at each node on the transaction link are sent to the message analysis module together with the unique serial number, and the analysis model determines whether the message is abnormal. If there is no abnormality, the relevant message content and the corresponding node name are stored as a data set for analysis model training and optimization. If an abnormality occurs, the relevant information will be pushed to the error handling module, the abnormality will be diagnosed and classified, and the corresponding processing plan will be intelligently generated. According to the previous system settings, error repair will be automatically performed for errors that do not require manual intervention. The means include but are not limited to automatic re-transaction, error data change repair, etc. The risky type of repair plan is pushed to the system console for manual compounding, and the manual compound plan is archived and saved as an exception handling data set for self-training and optimization of the classification model.

[0092] In this embodiment, if Figure 7 The figure shows the workflow of the message analysis module, i.e., the message analysis process, which specifically includes:

[0093] 1. Model training (preprocessing): First collect a certain number of normal transaction messages as the basic data set, extract message features, such as message type, message length, JSON data dictionary, transaction node sequence number, etc., and then use the BP algorithm to establish a three-layer neural network, and use the basic data set to train and optimize the neural network.

[0094] 2. Implementation analysis: Messages are collected in real time during the transaction process, and the features are extracted and input into the neural network. Abnormal messages are sent to the error processing module. Because a unique transaction serial number is set and message collection and analysis are performed for each transaction node, the specific abnormal node can be quickly located for transaction anomalies, and the transaction status of each subsystem during a single transaction can be traced back.

[0095] In this embodiment, if Figure 8 The following is a flowchart of the error handling module, i.e., the error handling process, which specifically includes:

[0096] 1. Model training: First preset the error type and prepare the corresponding data set. Use k-means to build a three-layer neural network with message features as input and preset error types and unknown exceptions as output. Use the prepared data set to train and adjust the network.

[0097] 2. Implementation analysis: The message analysis module classifies abnormal messages using the k-means clustering algorithm to determine the error type. Based on the error type and the pre-set knowledge base, transactions can be re-proposed for timeout errors, and unknown and high-risk errors can be pushed to the console for manual processing.

[0098] The present invention uses machine learning to provide a method for quickly locating and repairing errors in long-link transaction scenarios. By setting up a proxy service to capture packets and collecting logs of each node in the transaction link, the transaction request and return message on each node of the long-link transaction are collected and stored, and a machine learning algorithm is selected to process and analyze the collected messages, and the transaction message status is judged in real time, thereby providing the ability to quickly locate transaction errors during and after the event. Machine learning methods such as supervised learning, unsupervised learning, and reinforcement learning are used to analyze errors, provide solutions for simple error types, and implement self-repair. It is possible to quickly locate errors at the granularity of a single transaction and self-repair of preset error types, solving the problem of difficult location and untimely resolution of transaction errors in long-link transaction scenarios.

[0099] like Fig. 9FIG. 1 is a schematic diagram of a structure of an error processing device for a long link transaction according to an embodiment of the present invention. The device shown in the figure includes:

[0100] The transaction message module 10 is used to obtain transaction messages of multiple transaction nodes according to the received transaction initiation request.

[0101] Among them, a transaction initiation request sent by a customer is received, and a unique serial number is assigned to the transaction. Specifically, the unique serial number can be set by random generation or other methods. Each transaction has a unique serial number. According to the unique serial number, the transaction message corresponding to the unique serial number can be obtained from multiple transaction nodes in the long link transaction. Specifically, the transaction message includes the transaction request message and the transaction return message on each transaction node.

[0102] The analysis result module 20 is used to analyze and process the transaction message using a pre-established analysis model to obtain an analysis result.

[0103] Among them, a certain number of messages with normal transaction analysis results can be collected first, and message features can be extracted as training samples, such as message type, message length, json data dictionary, transaction node sequence number, etc., and then a three-layer neural network can be established using the BP algorithm. The neural network can be trained and optimized using the training samples to obtain an analysis model.

[0104] Furthermore, the transaction message features are extracted from the transaction message, and the transaction message features are input into the analysis model for analysis and processing. The obtained analysis results include whether the transaction status is normal. If the transaction is normal, the analysis result is a normal transaction. If the transaction is abnormal, the analysis result is a transaction error. In addition, the analysis result also includes the transaction node name. If the transaction is an error, the abnormal transaction node can be located according to the transaction node name.

[0105] The automatic repair module 30 is used to determine the transaction as an abnormal transaction and automatically repair the abnormal transaction if it is learned that the analysis result is a transaction error.

[0106] If the analysis result is a transaction error, it means that the transaction is an abnormal transaction. The error type of the abnormal transaction, such as the timeout type, is determined by using a pre-established classification model.

[0107] Furthermore, the error type can be manually pre-set and the corresponding data set can be prepared. The historical messages with the analysis result of transaction error can be used as the data set. The message features of the historical messages are extracted, and a three-layer neural network is built using the k-means algorithm with the message features as input and the preset error type and unknown exception as output. The neural network is trained and adjusted to obtain a classification model.

[0108] Different error types correspond to different preset repair rules, and the preset repair rules may be manual modification, etc. For example, if the error type is a timeout type, the preset repair rule may be transaction re-submission. If the error type is a complex type, the preset repair rule may be manual processing, thereby completing the automatic repair of abnormal transactions.

[0109] As an embodiment of the present invention, Fig.10 As shown, the device also includes an analysis model module 50, which is used to obtain historical messages whose analysis results are normal transactions, and extract a first historical message feature from the historical messages with normal transactions, and use the first historical message feature as a first training sample; input the first training sample into a first initial neural network model, train the first initial neural network model, and obtain the analysis model.

[0110] As an embodiment of the present invention, Fig.11 As shown, the analysis result module 20 includes:

[0111] The feature extraction unit 21 is used to extract transaction message features from the transaction message; wherein the transaction message features include: message type, message length and transaction node sequence number;

[0112] The analysis result unit 22 is used to input the transaction message characteristics into a pre-established analysis model to obtain an analysis result; wherein the analysis result includes a transaction node name.

[0113] As an embodiment of the present invention, Fig.12 As shown, the device also includes a classification model module 40, which is used to obtain historical messages with analysis results of transaction errors and corresponding error types, and use the historical messages of transaction errors and corresponding error types as second training samples; input the second training samples into the second initial neural network model, train the second initial neural network model, and obtain the classification model.

[0114] As an embodiment of the present invention, the transaction message module is further used to determine a unique serial number of the transaction according to the received transaction initiation request, and obtain transaction messages of multiple transaction nodes according to the unique serial number.

[0115] In this embodiment, the transaction message module is further used to obtain the transaction message corresponding to the unique serial number from multiple transaction nodes in real time according to the unique serial number.

[0116] As an embodiment of the present invention, Fig.13 As shown, the automatic repair module 30 includes:

[0117] The error type unit 31 is used to determine the transaction as an abnormal transaction if the analysis result is a transaction error, and determine the error type of the abnormal transaction according to a pre-established classification model;

[0118] The automatic repair unit 32 is used to determine the corresponding preset repair rule according to the error type, and automatically repair the abnormal transaction according to the preset repair rule.

[0119] Based on the same application concept as the above-mentioned error processing method for a long link transaction, the present invention also provides the above-mentioned error processing device for a long link transaction. Since the principle of solving the problem by the error processing device for a long link transaction is similar to that of the error processing method for a long link transaction, the implementation of the error processing device for a long link transaction can refer to the implementation of the error processing method for a long link transaction, and the repeated parts will not be repeated.

[0120] The present invention collects transaction messages on each node of a long-link transaction, processes and analyzes the collected messages, and realizes real-time judgment of the transaction message status, thereby providing the ability to quickly locate transaction errors during and after the event, analyzes the errors, and provides repair rules for the error types to achieve automatic repair.

[0121] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the program.

[0122] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for executing the above method.

[0123] like Fig.14 As shown, the electronic device 600 may further include: a communication module 110, an input unit 120, an audio processor 130, a display 160, and a power supply 170. It is worth noting that the electronic device 600 does not necessarily have to include Fig.14 In addition, the electronic device 600 may also include Fig.14 For components not shown, reference may be made to the prior art.

[0124] like Fig.14 As shown, the central processor 100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor devices and / or logic devices. The central processor 100 receives inputs and controls the operations of various components of the electronic device 600.

[0125] The memory 140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 100 may execute the program stored in the memory 140 to implement information storage or processing.

[0126] The input unit 120 provides input to the CPU 100. The input unit 120 is, for example, a key or a touch input device. The power supply 170 is used to provide power to the electronic device 600. The display 160 is used to display display objects such as images and text. The display may be, for example, an LCD display, but is not limited thereto.

[0127] The memory 140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 140 may also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application / function storage unit 142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 600 through the central processor 100.

[0128] The memory 140 may also include a data storage unit 143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 144 of the memory 140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0129] The communication module 110 is a transmitter / receiver 110 that transmits and receives signals via an antenna 111. The communication module (transmitter / receiver) 110 is coupled to the central processor 100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0130] Based on different communication technologies, multiple communication modules 110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module. The communication module (transmitter / receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide an audio output via the speaker 131 and receive an audio input from the microphone 132, thereby realizing a common telecommunication function. The audio processor 130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 130 is also coupled to the central processor 100, so that the sound can be recorded on the local machine through the microphone 132, and the sound stored on the local machine can be played through the speaker 131.

[0131] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0132] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0133] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0134] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0135] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for handling errors in long-link transactions, characterized in that: The method comprises: According to the received transaction initiation request, a unique serial number is set for the transaction initiation request, and according to the unique serial number, a transaction message corresponding to the unique serial number is obtained in real time from multiple transaction nodes of the long link transaction; wherein the unique serial number is used to locate and track each transaction in the long link transaction; Analyze and process the transaction message using a pre-established analysis model to obtain an analysis result; If it is learned that the analysis result is a transaction error, the transaction is determined to be an abnormal transaction, and the abnormal transaction is automatically repaired; Wherein, if it is known that the analysis result is a transaction error, determining the transaction as an abnormal transaction, and automatically repairing the abnormal transaction includes: If it is learned that the analysis result is a transaction error, the transaction is determined to be an abnormal transaction, and the error type of the abnormal transaction is determined according to a pre-established classification model; Determine a corresponding preset repair rule according to the error type, and automatically repair the abnormal transaction according to the preset repair rule; The classification model is pre-established in the following manner: Acquire historical messages whose analysis results are transaction errors and corresponding error types, extract second historical message features from the historical messages of transaction errors, and use the second historical message features and the corresponding error types as second training samples; The second training sample is input into a second initial neural network model, and the second initial neural network model is trained to obtain the classification model.

2. The method according to claim 1, characterized in that The analysis model is pre-established in the following manner: Acquire historical messages whose analysis results show that the transaction is normal, extract a first historical message feature from the historical messages of the normal transaction, and use the first historical message feature as a first training sample; The first training sample is input into a first initial neural network model, and the first initial neural network model is trained to obtain the analysis model.

3. The method according to claim 1, characterized in that The transaction message is analyzed and processed by using the pre-established analysis model to obtain the analysis results including: Extracting transaction message features from the transaction message; wherein the transaction message features include: message type, message length and transaction node sequence number; The transaction message features are input into a pre-established analysis model to obtain an analysis result; wherein the analysis result includes a transaction node name.

4. A long link transaction error processing device, characterized in that: The device comprises: A transaction message module, used to set a unique serial number for the transaction initiation request according to the received transaction initiation request, and obtain a transaction message corresponding to the unique serial number in real time from multiple transaction nodes of the long link transaction according to the unique serial number; wherein the unique serial number is used to locate and track each transaction in the long link transaction; An analysis result module is used to analyze and process the transaction message using a pre-established analysis model to obtain an analysis result; An automatic repair module, for determining the transaction as an abnormal transaction and automatically repairing the abnormal transaction if it is learned that the analysis result is a transaction error; The automatic repair module is further used to determine the transaction as an abnormal transaction if it is known that the analysis result is a transaction error, and determine the error type of the abnormal transaction according to a pre-established classification model; determine a corresponding preset repair rule according to the error type, and automatically repair the abnormal transaction according to the preset repair rule; The classification model is pre-established in the following manner: Acquire historical messages whose analysis results are transaction errors and corresponding error types, extract second historical message features from the historical messages of transaction errors, and use the second historical message features and the corresponding error types as second training samples; The second training sample is input into a second initial neural network model, and the second initial neural network model is trained to obtain the classification model.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 3 is implemented.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program for executing the method according to any one of claims 1 to 3.

Images