Transaction data processing method and system, electronic equipment and storage medium

Abstract

The invention provides a transaction data processing method and system, electronic equipment and a storage medium. The method comprises the steps of extracting target transaction data corresponding to a target detection dimension set from to-be-detected transaction data; the target detection dimension set is determined according to the first historical transaction data and the corresponding actual transaction state; inputting the target transaction data into the transaction detection model to obtain a predicted transaction state of the to-be-detected transaction data; the transaction detection model is obtained by training based on second historical transaction data and a corresponding actual transaction state; and if the predicted transaction state is an abnormal transaction state, performing abnormal transaction early warning according to the to-be-detected transaction data and the predicted transaction state corresponding to the to-be-detected transaction data. Through the mode of the application, the identification accuracy of the abnormal transaction can be improved through the transaction detection model, and early warning can be timely performed on the abnormal transaction data.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a method, a system, an electronic device, and a storage medium for processing transaction data. Background Art

[0002] The traditional method for monitoring credit card transactions is to identify potential abnormal transactions by presetting a series of rules. When the transaction data triggers these rules, it will be marked as an abnormal transaction, and may trigger a further manual review process and give an early warning.

[0003] However, the traditional method for monitoring credit card transactions has low monitoring efficiency when dealing with large-scale transaction data, and due to the limitations of the rules and the subjectivity of manual review, it reduces the accuracy of identifying abnormal transactions, and thus cannot give an early warning based on abnormal transactions. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, a system, an electronic device, and a storage medium for processing transaction data, which can improve the accuracy of identifying abnormal transactions through a transaction detection model and give an early warning of abnormal transaction data in a timely manner.

[0005] In a first aspect, an embodiment of this application provides a method for processing transaction data, and the method for processing transaction data includes:

[0006] Extract target transaction data corresponding to a target detection dimension set from the transaction data to be detected; the target detection dimension set is determined according to first historical transaction data and the corresponding actual transaction status; the first historical transaction data is historical transaction data used to update the target detection dimension set;

[0007] Input the target transaction data into a transaction detection model to obtain the predicted transaction status of the transaction data to be detected; the transaction detection model is trained based on second historical transaction data and the corresponding actual transaction status; the second historical transaction data is transaction data used to train the transaction detection model;

[0008] If the predicted transaction status is an abnormal transaction status, give an early warning of abnormal transactions according to the transaction data to be detected and the predicted transaction status corresponding to the transaction data to be detected.

[0009] In a possible implementation manner, the target detection dimension set is updated through the following steps:

[0010] Obtain the first historical transaction data within the latest first preset duration and the actual transaction status corresponding to the first historical transaction data according to a preset target detection dimension update duration;

[0011] From each of the first historical transaction data, obtain the first target historical transaction data corresponding to each initial detection dimension set; the initial detection dimension set includes at least one initial detection dimension;

[0012] Input each first target historical transaction data into the transaction detection model to obtain the predicted transaction status corresponding to each first target historical transaction data;

[0013] According to the predicted transaction status and the actual transaction status corresponding to all the first target historical transaction data, determine the performance score corresponding to each initial detection dimension set;

[0014] Determine the initial detection dimension set with the maximum performance score as the target detection dimension set.

[0015] In a possible implementation manner, the method further includes:

[0016] If the non-update duration of the transaction detection model is greater than or equal to the preset model update duration, obtain the third historical transaction data within the latest second preset duration and the actual transaction status corresponding to the third historical transaction data;

[0017] From each of the third historical transaction data, obtain the second target historical transaction data corresponding to the target detection dimension set;

[0018] Update the transaction detection model according to all the second target historical transaction data and the actual transaction status corresponding to all the second target historical transaction data.

[0019] In a possible implementation manner, the updating the transaction detection model according to all the second target historical transaction data and the actual transaction status corresponding to all the second target historical transaction data includes:

[0020] Input each second target historical transaction data into the transaction detection model to obtain the predicted transaction status corresponding to each second target historical transaction data;

[0021] Count the number of second target historical transaction data with inconsistent actual transaction status and predicted transaction status;

[0022] If the number is greater than the preset number, use all the second target historical transaction data as samples and the actual transaction status corresponding to all the second target historical transaction data as labels to update the transaction detection model.

[0023] In a possible implementation manner, the performing abnormal transaction warning according to the transaction data to be detected and the predicted transaction status corresponding to the transaction data to be detected includes:

[0024] Display the to-be-detected transaction data and the predicted transaction status corresponding to the to-be-detected transaction data on the client of the monitoring user and / or the client of the transaction user corresponding to the to-be-detected transaction data.

[0025] In a possible implementation manner, the method further includes:

[0026] Obtain second historical transaction data and the corresponding actual transaction status;

[0027] Use the second historical transaction data as a sample and the actual transaction status corresponding to the second historical transaction data as a label to train the transaction detection model.

[0028] In a second aspect, an embodiment of the present application further provides a transaction data processing system, and the system includes:

[0029] An extraction module, configured to extract target transaction data corresponding to a target detection dimension set from the to-be-detected transaction data; the target detection dimension set is determined according to first historical transaction data and the corresponding actual transaction status; the first historical transaction data is historical transaction data for updating the target detection dimension set;

[0030] An input module, configured to input the target transaction data into a transaction detection model to obtain the predicted transaction status of the to-be-detected transaction data; the transaction detection model is trained based on second historical transaction data and the corresponding actual transaction status; the second historical transaction data is transaction data for training the transaction detection model;

[0031] An early warning module, configured to, if the predicted transaction status is an abnormal transaction status, perform an abnormal transaction early warning according to the to-be-detected transaction data and the predicted transaction status corresponding to the to-be-detected transaction data.

[0032] In a possible implementation manner, the extraction module is further configured to:

[0033] Obtain the first historical transaction data within the latest first preset duration and the actual transaction status corresponding to the first historical transaction data according to a preset target detection dimension update duration;

[0034] Obtain, from each of the first historical transaction data, first target historical transaction data corresponding to each initial detection dimension set; at least one initial detection dimension is included in the initial detection dimension set;

[0035] Input each first target historical transaction data into the transaction detection model to obtain the predicted transaction status corresponding to each first target historical transaction data;

[0036] Determine the performance score corresponding to each initial detection dimension set according to the predicted transaction status and the actual transaction status corresponding to all the first target historical transaction data;

[0037] Determine the initial detection dimension set with the maximum performance score as the target detection dimension set.

[0038] In a possible implementation manner, the input module is further configured to:

[0039] If the non-update duration of the transaction detection model is greater than or equal to the preset model update duration, obtain the third historical transaction data within the latest second preset duration and the actual transaction status corresponding to the third historical transaction data;

[0040] From each piece of the third historical transaction data, obtain the second target historical transaction data corresponding to the target detection dimension set;

[0041] Update the transaction detection model according to all the second target historical transaction data and the actual transaction status corresponding to all the second target historical transaction data.

[0042] In a possible implementation manner, the input module is specifically configured to input each piece of the second target historical transaction data into the transaction detection model to obtain the predicted transaction status corresponding to each piece of the second target historical transaction data; count the number of pieces of the second target historical transaction data where the actual transaction status and the predicted transaction status are inconsistent; if the number is greater than the preset number, use all the second target historical transaction data as samples and the actual transaction status corresponding to all the second target historical transaction data as labels to update the transaction detection model.

[0043] In a possible implementation manner, the warning module is specifically configured to display the to-be-detected transaction data and the predicted transaction status corresponding to the to-be-detected transaction data on the client of the monitoring user and / or the client of the transaction user corresponding to the to-be-detected transaction data.

[0044] In a possible implementation manner, the input module is further configured to:

[0045] Obtain the second historical transaction data and the corresponding actual transaction status;

[0046] Use the second historical transaction data as samples and the actual transaction status corresponding to the second historical transaction data as labels to train the transaction detection model.

[0047] In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of the transaction data processing method according to any one of the first aspects.

[0048] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is run by a processor, it executes the steps of the transaction data processing method according to any one of the first aspects.

[0049] An embodiment of the present application provides a transaction data processing method, system, electronic device, and storage medium. The method includes: extracting target transaction data corresponding to a target detection dimension set from the to-be-detected transaction data; the target detection dimension set is determined according to the first historical transaction data and the corresponding actual transaction status; inputting the target transaction data into a transaction detection model to obtain the predicted transaction status of the to-be-detected transaction data; the transaction detection model is trained based on the second historical transaction data and the corresponding actual transaction status; if the predicted transaction status is an abnormal transaction status, an abnormal transaction warning is given according to the to-be-detected transaction data and the predicted transaction status corresponding to the to-be-detected transaction data. Through the method of the present application, the recognition accuracy of abnormal transactions can be improved through the transaction detection model, and early warnings can be given to abnormal transaction data in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0051] Figure 1 Shows a flowchart of a transaction data processing method provided by an embodiment of the present application;

[0052] Figure 2 Shows a flowchart of updating the target detection dimension set provided by an embodiment of the present application;

[0053] Figure 3 Shows a schematic structural diagram of a transaction data processing system provided by an embodiment of the present application;

[0054] Figure 4 Shows a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It should be understood that the accompanying drawings in the present application are only for the purposes of illustration and description, and are not used to limit the protection scope of the present application. In addition, it should be understood that the schematic drawings are not drawn in actual proportions. The flowcharts used in the present application illustrate the operations implemented according to some embodiments of the present application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and the steps without logical context relationships may be reversed or implemented simultaneously. In addition, those skilled in the art may add one or more other operations to the flowchart or remove one or more operations from the flowchart under the guidance of the content of the present application.

[0056] In addition, the described embodiments are only some embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application usually described and illustrated in the accompanying 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 accompanying drawings is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the protection scope of the present application.

[0057] To enable those skilled in the art to use the content of the present application, the following implementation manners are given in combination with a specific application scenario, the "data processing technology field". For those skilled in the art, without departing from the spirit and scope of the present application, the general principles defined here can be applied to other embodiments and application scenarios. Although the present application is mainly described around the "data processing technology field", it should be understood that this is only an exemplary embodiment.

[0058] It should be noted that the term "including" will be used in the embodiments of the present application to indicate the existence of the features stated thereafter, but does not exclude the addition of other features.

[0059] The following provides a detailed description of a method for processing transaction data provided in the embodiments of the present application.

[0060] Refer to Figure 1 As shown, it is a schematic flowchart of a method for processing transaction data provided in the embodiments of the present application. The following describes the exemplary steps in the embodiments of the present application:

[0061] S101. Extract target transaction data corresponding to a target detection dimension set from the to-be-detected transaction data.

[0062] In the embodiment of the present application, the transaction data to be detected is the transaction data that needs to be processed after cleaning, deduplication, and formatting. The number of transaction data dimensions corresponding to the transaction data to be detected is greater than or equal to the number of target detection dimensions in the target detection dimension set. The target detection dimension is one or more transaction data dimensions selected from all transaction data dimensions corresponding to the transaction data in the credit card transaction log. Among them, the target detection dimension set is determined according to the first historical transaction data and the corresponding actual transaction status. The actual transaction status includes abnormal transaction status and normal transaction status.

[0063] For example, the transaction data dimensions corresponding to the transaction data to be detected of a certain credit card include transaction amount, transaction location, transaction time, transaction frequency, and transaction type (consumption type, transfer type, refund type, etc.). The target detection dimensions include transaction user name, transaction amount, transaction location, and transaction frequency. The transaction data corresponding to the transaction amount, the transaction data corresponding to the transaction location, and the transaction data corresponding to the transaction frequency are extracted from the transaction data to be detected to obtain the target transaction data.

[0064] Here, in the embodiment of the present application, the target detection dimension set determined in advance can reduce the data volume of the target transaction data, thereby improving the processing speed of the transaction data.

[0065] In addition, the target detection dimension set in the present application needs to be updated every once in a while to avoid reducing the processing accuracy of the transaction data due to changes or upgrades in abnormal transaction types (such as fraud).

[0066] Further, as shown in Figure 2 the flowchart for updating the target detection dimension set provided by the embodiment of the present application, the specific steps include:

[0067] S201. Obtain the latest first historical transaction data within the first preset duration and the actual transaction status corresponding to the first historical transaction data according to the preset target detection dimension update duration.

[0068] In the embodiment of the present application, the first historical transaction data is the transaction data that needs to be processed after cleaning, deduplication, and formatting. The latest first historical transaction data within the first preset duration and the actual transaction status corresponding to the first historical transaction data are obtained from the transaction log of the credit card. The first historical transaction data is the historical transaction data used to update the target detection dimension set.

[0069] Among them, the first historical transaction data needs to ensure that it is the latest and the transaction data within the first preset duration, so as to ensure that the updated target detection dimension set is more applicable to processing the latest abnormal transaction types, thereby improving the processing accuracy of the transaction data.

[0070] S202. Obtain the first target historical transaction data corresponding to each initial detection dimension set from each first historical transaction data.

[0071] In the embodiment of the present application, the initial detection dimension set includes at least one initial detection dimension. Generally, all transaction data dimensions corresponding to the transaction data in the transaction log are determined as the initial detection dimensions. The number of initial detection dimensions in each initial detection dimension set is greater than or equal to 1 and less than or equal to the total number of initial detection dimensions.

[0072] Example: A certain initial detection dimension set includes transaction amount, transaction location, and transaction time; the transaction data is that at 2 am, Zhang San transferred 500 yuan to Li Si in City X, then the first target historical transaction data is a transaction of 500 yuan in City X at 2 am. Here, transfer refers to the transaction type, and the transaction type is not included in the initial detection dimension set, so transfer in the first target historical transaction data should be replaced with transaction.

[0073] Among them, the transaction data dimensions may include dimensions such as transaction user name, transaction amount, transaction location, transaction time, transaction frequency, and transaction type. Among them, the transaction frequency in a certain transaction data of a certain transaction user refers to the proportion of transaction data similar to this transaction data among all transaction data of this transaction user.

[0074] Determine whether two transaction data are similar through the following steps: count the number of transaction data dimensions with the same values between the two transaction data; when the preset amount ranges where the transaction amounts of the two transaction data are located are the same, it is considered that the values corresponding to the transaction amounts between the two transaction data are the same; if the number of transaction data dimensions with the same values between the two transaction data is greater than the preset same number, then the two transaction data are similar.

[0075] Here, since it is difficult for transaction amounts to be exactly the same, the present application pre-sets multiple preset amount ranges. If the preset amount ranges where two transaction amounts are located are the same, then it is considered that the two transaction amounts are the same.

[0076] S203. Input each first target historical transaction data into the transaction detection model to obtain the predicted transaction status corresponding to each first target historical transaction data.

[0077] In the embodiment of the present application, the predicted transaction status includes an abnormal transaction status and a normal transaction status.

[0078] S204. Determine the performance score corresponding to each initial detection dimension set according to the predicted transaction status and the actual transaction status corresponding to all first target historical transaction data.

[0079] In the embodiment of the present application, for any initial detection dimension set, the number of target historical transaction data corresponding to the initial detection dimension set where the predicted transaction status is consistent with the actual transaction status is counted to obtain the target number; the ratio of the target number to the total number of the first target historical transaction data corresponding to the initial detection dimension set is determined as the performance score corresponding to the initial detection dimension set.

[0080] Here, the number of transaction data at different times may be different, and the number of the first historical transaction data obtained each time the target detection dimension set is updated will also be different. Therefore, it is inaccurate to determine the performance score corresponding to the initial detection dimension set solely based on the target number, and it is more accurate to use the ratio method to determine the performance score corresponding to the initial detection dimension set.

[0081] S205. Determine the initial detection dimension set with the maximum performance score as the target detection dimension set.

[0082] S102. Input the target transaction data into the transaction detection model to obtain the predicted transaction status of the transaction data to be detected.

[0083] In the embodiment of the present application, the transaction detection model is trained based on the second historical transaction data and the corresponding actual transaction status, and is used to determine the predicted transaction status of the transaction data. The second historical transaction data is the historical transaction data used to train the transaction detection model; the predicted transaction status includes an abnormal transaction status and a normal transaction status.

[0084] Here, the transaction detection mode is pre-trained. The specific training steps of the transaction detection model include:

[0085] Step 1. Obtain the third historical transaction data and the corresponding actual transaction status.

[0086] In the embodiment of the present application, the third historical transaction data is the transaction data to be processed after cleaning, de-duplication, and formatting. The third historical transaction data is used to train the transaction detection model.

[0087] Step 2. Use the third historical transaction data as a sample and the actual transaction status corresponding to the third historical transaction data as a label to perform model training on the transaction detection model.

[0088] In the embodiment of the present application, the third historical transaction data includes the transaction data corresponding to all transaction data dimensions. Input the third historical transaction data into the transaction detection model to obtain the predicted transaction status corresponding to the third historical transaction data; train the transaction detection model according to the predicted transaction status and the actual transaction status corresponding to the third historical transaction data.

[0089] It should be noted that when the samples used to train the transaction detection model contain transaction data corresponding to all transaction data dimensions, the accuracy of the transaction detection model obtained after training is not affected regardless of which transaction data dimensions the target detection dimension set includes. From the update process of the target detection dimensions, it can be seen that the accuracy of determining the target detection dimension set completely depends on the accuracy of the transaction detection model. Therefore, the training of the transaction detection model needs to be completed before the target detection dimension set is first determined.

[0090] In addition, in order to enable the transaction detection model to better adapt to changes or upgrades in abnormal transaction types, it is necessary to frequently update the transaction detection model with new transaction data. The transaction detection model is updated specifically through the following steps:

[0091] Step 1: If the non-update duration of the transaction detection model is greater than or equal to the preset model update duration, obtain the second historical transaction data within the latest second preset duration and the actual transaction status corresponding to the second historical transaction data.

[0092] In the embodiment of the present application, the second historical transaction data is the transaction data that needs to be processed after being cleaned, de-duplicated, and formatted. Obtain the second historical transaction data within the latest second preset duration and the actual transaction status corresponding to the second historical transaction data from the transaction log of the credit card. The non-update duration of the transaction detection model refers to the duration between the current time and the last update time corresponding to the transaction detection model. The second historical transaction data is used to update the transaction detection model.

[0093] Among them, the second historical transaction data needs to ensure that it is the latest and within the first preset duration to ensure that the updated transaction detection model is more suitable for processing the latest abnormal transaction types, thereby improving the processing accuracy of transaction data.

[0094] Step 2: Obtain the second target historical transaction data corresponding to the target detection dimension set from each second historical transaction data.

[0095] Step 3: Update the transaction detection model according to all the second target historical transaction data and the actual transaction status corresponding to all the second target historical transaction data.

[0096] Specifically, updating the transaction detection model includes:

[0097] i. Input each second target historical transaction data into the transaction detection model to obtain the predicted transaction status corresponding to each second target historical transaction data.

[0098] ii. Count the number of second target historical transaction data whose actual transaction status and predicted transaction status are inconsistent.

[0099] iii. If the quantity is greater than the preset quantity, all the second target historical transaction data will be used as samples, and the actual transaction status corresponding to all the second target historical transaction data will be used as labels to update the transaction detection model.

[0100] In the embodiment of the present application, the second target historical transaction data is input into the transaction detection model to obtain the predicted transaction status corresponding to the second target historical transaction data; the transaction detection model is updated according to the actual transaction status and the predicted transaction status corresponding to the second target historical transaction data.

[0101] It should be noted that when updating the transaction detection model, in addition to considering the unupdated duration, the quantity of the second target historical transaction data in the second historical transaction data also needs to be considered. In order to ensure that the transaction detection model can be updated in a timely manner, in the embodiment of the present application, the transaction detection model cannot be updated at a preset time interval.

[0102] S103. If the predicted transaction status is an abnormal transaction status, an abnormal transaction warning is given according to the transaction data to be detected and the predicted transaction status corresponding to the transaction data to be detected.

[0103] In the embodiment of the present application, the transaction data to be detected and the predicted transaction status corresponding to the transaction data to be detected are displayed on the client of the monitoring user and / or the client of the transaction user corresponding to the transaction data to be detected. Among them, the detection user can be a bank employee, and the transaction user can refer to the cardholder of the credit card.

[0104] The embodiment of the present application provides a method for processing transaction data. The method includes: extracting target transaction data corresponding to a target detection dimension set from the transaction data to be detected; the target detection dimension set is determined according to the first historical transaction data and the corresponding actual transaction status; inputting the target transaction data into the transaction detection model to obtain the predicted transaction status of the transaction data to be detected; the transaction detection model is trained based on the second historical transaction data and the corresponding actual transaction status; if the predicted transaction status is an abnormal transaction status, an abnormal transaction warning is given according to the transaction data to be detected and the predicted transaction status corresponding to the transaction data to be detected. Through the method of the present application, the recognition accuracy of abnormal transactions can be improved by the transaction detection model, and abnormal transaction data can be warned in a timely manner.

[0105] Based on the same inventive concept, the embodiment of the present application also provides a transaction data processing system corresponding to the transaction data processing method. Since the principle of solving problems by the system in the embodiment of the present application is similar to the above-mentioned transaction data processing method in the embodiment of the present application, the implementation of the system can refer to the implementation of the method, and the repeated parts will not be described again.

[0106] Refer to Figure 3As shown in the figure, it is a schematic diagram of a processing system for transaction data provided by an embodiment of the present application. The system includes:

[0107] An extraction module 301, configured to extract target transaction data corresponding to a target detection dimension set from the to-be-detected transaction data; the target detection dimension set is determined according to first historical transaction data and corresponding actual transaction statuses; the first historical transaction data is historical transaction data used to update the target detection dimension set;

[0108] An input module 302, configured to input the target transaction data into a transaction detection model to obtain a predicted transaction status of the to-be-detected transaction data; the transaction detection model is trained based on second historical transaction data and corresponding actual transaction statuses; the second historical transaction data is historical transaction data used to train the transaction detection model;

[0109] An early warning module 303, configured to, if the predicted transaction status is an abnormal transaction status, perform an abnormal transaction early warning according to the to-be-detected transaction data and the predicted transaction status corresponding to the to-be-detected transaction data.

[0110] In a possible implementation manner, the extraction module 301 is further configured to:

[0111] Obtain first historical transaction data within a latest first preset time period and the actual transaction status corresponding to the first historical transaction data according to a preset target detection dimension update duration;

[0112] From each of the first historical transaction data, obtain first target historical transaction data corresponding to each initial detection dimension set; at least one initial detection dimension is included in the initial detection dimension set;

[0113] Input each first target historical transaction data into the transaction detection model to obtain a predicted transaction status corresponding to each first target historical transaction data;

[0114] Determine a performance score corresponding to each initial detection dimension set according to the predicted transaction statuses and actual transaction statuses corresponding to all the first target historical transaction data;

[0115] Determine the initial detection dimension set with the maximum performance score as the target detection dimension set.

[0116] An embodiment of the present application provides a transaction data processing system, which includes: an extraction module 301 for extracting target transaction data corresponding to a target detection dimension set from the to-be-detected transaction data; the target detection dimension set is determined according to the first historical transaction data and the corresponding actual transaction status; an input module 302 for inputting the target transaction data into a transaction detection model to obtain a predicted transaction status of the to-be-detected transaction data; the transaction detection model is trained based on the second historical transaction data and the corresponding actual transaction status; an early warning module 303 for, if the predicted transaction status is an abnormal transaction status, performing an abnormal transaction early warning according to the to-be-detected transaction data and the predicted transaction status corresponding to the to-be-detected transaction data. By the method of the present application, the recognition accuracy of abnormal transactions can be improved through the transaction detection model, and early warning of abnormal transaction data can be performed in a timely manner.

[0117] As Figure 4 shown, an electronic device 400 provided by an embodiment of the present application includes: a processor 401, a memory 402, and a bus. The memory 402 stores machine-readable instructions executable by the processor 401. When the electronic device runs, communication between the processor 401 and the memory 402 is carried out through the bus, and the processor 401 executes the machine-readable instructions to execute the steps of the transaction data processing method as described above.

[0118] Specifically, the above-mentioned memory 402 and processor 401 can be general-purpose memory and processor, which are not specifically limited here. When the processor 401 runs the computer program stored in the memory 402, the transaction data processing method as described above can be executed.

[0119] Corresponding to the above-mentioned transaction data processing method, an embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the steps of the above-mentioned transaction data processing method are executed.

[0120] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the method embodiments, and will not be elaborated herein. In 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 illustrative. For example, the division of the modules is only a logical function division, and there can be other division methods in actual implementation. For another example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of the devices or modules can be in electrical, mechanical, or other forms.

[0121] The modules described as separate components may or may not be physically separated. The components displayed as modules may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] In addition, in each embodiment of the present application, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0123] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the information processing method described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0124] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for processing transaction data, characterized in that: The method comprises: Extracting target transaction data corresponding to the target detection dimension set from the transaction data to be detected; the target detection dimension set is determined based on the first historical transaction data and the corresponding actual transaction status; the first historical transaction data is the historical transaction data used to update the target detection dimension set; The target transaction data is input into a transaction detection model to obtain a predicted transaction status of the transaction data to be detected; the transaction detection model is trained based on the second historical transaction data and the corresponding actual transaction status; the second historical transaction data is the historical transaction data used to train the transaction detection model; If the predicted transaction status is an abnormal transaction status, an abnormal transaction warning is issued according to the transaction data to be detected and the predicted transaction status corresponding to the transaction data to be detected.

2. The method for processing transaction data according to claim 1, characterized in that: Update the target detection dimension set by the following steps: According to the preset target detection dimension update duration, the first historical transaction data within the latest first preset duration and the actual transaction status corresponding to the first historical transaction data are obtained; From each of the first historical transaction data, obtain first target historical transaction data corresponding to each initial detection dimension set; the initial detection dimension set includes at least one initial detection dimension; Inputting each first target historical transaction data into the transaction detection model to obtain a predicted transaction state corresponding to each first target historical transaction data; Determine a performance score corresponding to each initial detection dimension set according to the predicted transaction status and the actual transaction status corresponding to all first target historical transaction data; The initial detection dimension set with the largest performance score is determined as the target detection dimension set.

3. The method for processing transaction data according to claim 2, characterized in that: The method further comprises: If the non-updated time length of the transaction detection model is greater than or equal to the preset model update time length, obtaining the latest third historical transaction data within the second preset time length and the actual transaction status corresponding to the third historical transaction data; From each third historical transaction data, obtain second target historical transaction data corresponding to the target detection dimension set; The transaction detection model is updated according to all the second target historical transaction data and the actual transaction status corresponding to all the second target historical transaction data.

4. The method for processing transaction data according to claim 3, characterized in that: The updating of the transaction detection model according to all the second target historical transaction data and the actual transaction status corresponding to all the second target historical transaction data includes: Inputting each second target historical transaction data into the transaction detection model to obtain a predicted transaction state corresponding to each second target historical transaction data; Counting the number of second target historical transaction data for which the actual transaction status is inconsistent with the predicted transaction status; If the number is greater than a preset number, all second target historical transaction data are used as samples, and the actual transaction states corresponding to all second target historical transaction data are used as labels to update the transaction detection model.

5. The method for processing transaction data according to claim 1, characterized in that: The abnormal transaction warning is performed according to the transaction data to be detected and the predicted transaction status corresponding to the transaction data to be detected, including: The transaction data to be detected and the predicted transaction status corresponding to the transaction data to be detected are displayed to a client of a monitoring user and / or a client of a transaction user corresponding to the transaction data to be detected.

6. The method for processing transaction data according to any one of claims 1 to 5, characterized in that: The method further comprises: Obtaining second historical transaction data and corresponding actual transaction status; The second historical transaction data is used as a sample, and the actual transaction status corresponding to the second historical transaction data is used as a label to perform model training on the transaction detection model.

7. A transaction data processing system, characterized in that: The system comprises: An extraction module, used to extract target transaction data corresponding to a target detection dimension set from the transaction data to be detected; the target detection dimension set is determined based on the first historical transaction data and the corresponding actual transaction status; An input module, used to input the target transaction data into a transaction detection model to obtain a predicted transaction status of the transaction data to be detected; the transaction detection model is trained based on the second historical transaction data and the corresponding actual transaction status; The early warning module is used for issuing an abnormal transaction early warning according to the transaction data to be detected and the predicted transaction status corresponding to the transaction data to be detected if the predicted transaction status is an abnormal transaction status.

8. The transaction data processing system according to claim 7, characterized in that: Extraction module, also used for: According to the preset target detection dimension update duration, the first historical transaction data within the latest first preset duration and the actual transaction status corresponding to the first historical transaction data are obtained; From each of the first historical transaction data, obtain first target historical transaction data corresponding to each initial detection dimension set; the initial detection dimension set includes at least one initial detection dimension; Inputting each first target historical transaction data into the transaction detection model to obtain a predicted transaction state corresponding to each first target historical transaction data; Determine a performance score corresponding to each initial detection dimension set according to the predicted transaction status and the actual transaction status corresponding to all first target historical transaction data; The initial detection dimension set with the largest performance score is determined as the target detection dimension set.

9. An electronic device, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform the steps of the transaction data processing method as described in any one of claims 1 to 6.

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 transaction data processing method according to any one of claims 1 to 6 are executed.

Images