A transaction data verification method, device, equipment and storage medium

By performing type analysis and feature extraction on transaction data, target business scenarios are generated and correlation analysis is conducted, which solves the problem of transaction data identification errors, enables accurate processing and anomaly detection in multiple business scenarios, and improves the efficiency and accuracy of transaction data processing.

CN116843291BActive Publication Date: 2024-11-26广州淘通科技股份有限公司 +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310788458.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-29
Publication Date
2024-11-26
Estimated Expiration
2043-06-29

AI Technical Summary

Technical Problem

Existing technologies are prone to identification errors when processing transaction data with different configuration parameters, resulting in inaccurate transaction data processing, decreased work efficiency and accuracy, and can only be processed within limited business scenarios, failing to accurately reproduce the business scenario.

Method used

By intercepting transaction data from business systems, performing type analysis and feature extraction, generating target business scenarios, and using correlation analysis to determine transaction data anomalies, configuration parameters are modified to reduce the error rate.

Benefits of technology

It improves the accuracy and efficiency of transaction data processing, enabling accurate identification of transaction data and rapid identification of abnormal users in multiple business scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116843291B_ABST
    Figure CN116843291B_ABST
Patent Text Reader

Abstract

The present invention discloses a transaction data verification method, device, equipment and storage medium, the method comprising: intercepting transaction data in a business system, and performing type analysis on the transaction data to generate corresponding business types; according to the business types of the transaction data identified by the same user ID, obtaining corresponding transaction elements and generating target business scenarios; performing feature extraction on the transaction data identified by the same user ID to generate scenario feature parameters; according to the scenario feature parameters, matching the corresponding predicted business scenarios; judging whether the transaction data is abnormal according to the correlation value between the target business scenario and the predicted business scenario; if so, sending abnormal reminder information to the business system to notify the corresponding business personnel to modify the configuration parameters of the transaction data, so as to realize analysis and abnormal judgment of transaction data with different configuration parameters, and by modifying the abnormal transaction data, reducing the error rate of the transaction data and improving the work efficiency of transaction data processing.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

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

[0002] The existing transaction data processing technology obtains the user's transaction data for analysis to identify the user's business scenario. In the information age, users have various business scenarios, such as order payment, order settlement, after-sales refund, third-party service reconciliation, etc., among which payment includes business scenarios such as transfer, red envelope or consumption voucher deduction and lending.

[0003] In the process of obtaining user transaction data, since the user transaction data is generated in business systems located at different sites, the transaction data generated in different business scenarios of different business systems have different configuration parameters. Therefore, when the transaction data with different configuration parameters are collected and analyzed, it is easy to cause errors in transaction data recognition, which in turn generates abnormal data, which in turn leads to inaccurate transaction data processing, and thus reduces the efficiency of transaction data processing; in addition, the existing transaction data processing method only processes and identifies the acquired transaction data within a limited range of business scenarios, and cannot restore the accurate business scenario, which in turn generates abnormal business scenarios, and thus reduces the accuracy of transaction data processing. Summary of the invention

[0004] The present invention provides a transaction data verification method to analyze and judge abnormalities of transaction data with different configuration parameters, and to reduce the error rate of transaction data and improve the work efficiency of transaction data processing by modifying abnormal transaction data.

[0005] The present invention provides a transaction data verification method, comprising: intercepting first transaction data in a first business system, and performing type analysis on the first transaction data to generate a corresponding business type; the first transaction data carries a corresponding user ID; the first transaction data is generated by a user initiating a business request to the business system;

[0006] According to the business type of the second transaction data identified by the same user ID, the corresponding transaction elements are obtained, and a target business scenario is generated; the second transaction data identified by the same user ID is subjected to feature extraction to generate scenario feature parameters; according to the scenario feature parameters, the corresponding predicted business scenario is matched; and the predicted business scenario is stored in a scenario database;

[0007] According to the correlation value between the target business scenario and the predicted business scenario, it is determined whether the second transaction data is abnormal; if so, an abnormal reminder message is sent to the first business system to notify the corresponding business personnel to modify the configuration parameters of the second transaction data.

[0008] As a preferred solution, this application generates theoretically feasible target business scenarios by performing type analysis on transaction data, and generates practically feasible predicted business scenarios by extracting and matching transaction data features; by performing correlation analysis on the two business scenarios generated by theoretical and practical analysis of the same transaction data, it is possible to determine whether the transaction data is abnormal. In addition, this application identifies the transaction data of the same user ID, and realizes the construction and abnormal analysis of business scenarios for the same user, which can not only improve the accuracy of scenario construction, but also quickly lock abnormal users based on abnormal transaction data; when abnormal data is determined, by modifying the configuration parameters of the transaction data, the error rate of transaction data analysis with different configuration parameters can be reduced, which can improve the accuracy of transaction data and improve the work efficiency of transaction data processing.

[0009] Furthermore, the first transaction data in the first business system is intercepted, and the type of the first transaction data is analyzed to generate a corresponding business type, specifically:

[0010] When a business request event is received in the first business system within a preset time period, the user ID and the corresponding first transaction data in the business request event are collected, and the user ID is identified in the corresponding first transaction data; the second transaction data identified by the same user ID is classified, and a business type set is generated for the second transaction data identified by each user ID.

[0011] As a preferred solution, this application identifies the transaction data of the same user ID, and implements the construction and anomaly analysis of business scenarios for the same user, which not only improves the accuracy of scenario construction, but also can quickly lock abnormal users based on abnormal transaction data.

[0012] Furthermore, according to the business type of the second transaction data identified by the same user ID, the corresponding transaction elements are obtained, and a target business scenario is generated, specifically:

[0013] According to the business type set of the second transaction data identified by the same user ID, by searching for a mapping relationship between the business type of the transaction data and the transaction element, determining the transaction element corresponding to each business type in the second transaction data; the transaction element includes: one or more of transaction time, transaction cycle, transaction channel, transaction method and transaction subject;

[0014] The business type and the corresponding transaction elements are input into a scenario construction model; and at least one target business scenario that meets the conditions is generated according to each combination of the business type and the corresponding transaction elements.

[0015] Furthermore, according to each combination of the business type and the corresponding transaction elements, at least one target business scenario that meets the conditions is generated, specifically:

[0016] Determine the model parameters of the scenario construction model according to the business type and the corresponding transaction elements; cluster and fit the model parameters according to the historical transaction data and the preset transaction rules to generate combination information of each model parameter combination; generate scenario configuration data for each target business scenario based on each combination information; determine the configuration business parameters of each target business scenario based on the scenario configuration data, and construct the corresponding target business scenario.

[0017] Furthermore, feature extraction is performed on the second transaction data identified by the same user ID to generate scenario feature parameters, and the corresponding predicted business scenario is matched according to the scenario feature parameters, specifically:

[0018] Extracting a feature expression vector of second transaction data identified by the same user ID, calling a classification model to classify and identify the feature expression vector, generating a scene type label for the second transaction data, and using the scene type label and the business parameter value in the second transaction data as a scene feature parameter; the classification model is trained by a multi-label classifier on the feature expression vectors of historical transaction data and their corresponding scene type labels;

[0019] Calculate the similarity between the scenario feature parameters and the scenario features in the predicted business scenarios in the scenario database, and filter out the predicted business scenarios whose similarity values ​​meet the preset range.

[0020] As a preferred solution, the present application generates a theoretically feasible target business scenario by performing type analysis on transaction data and simulating the combination of transaction elements; and generates a practically feasible predicted business scenario with the most similar data features through transaction data feature extraction and matching; by comparing the two business scenarios generated by theoretical and practical analysis of the same transaction data, it is possible to judge the anomalies of transaction data, and by modifying the abnormal transaction data, reduce the error rate of transaction data and improve the work efficiency of transaction data processing.

[0021] Further, judging whether the second transaction data is abnormal according to the correlation value between the target business scenario and the predicted business scenario is as follows:

[0022] The business scenario correlation value calculation model is used to calculate the correlation value of each target business scenario and each predicted business scenario in turn through the joint probability density function; when the correlation value is greater than the first threshold, the calculation of the correlation value is stopped, and a normal result of the second transaction data is generated; when all correlation values ​​are not greater than the first threshold, an abnormal result of the second transaction data is generated.

[0023] As a preferred solution, the present application performs correlation analysis on two business scenarios generated by theoretical and actual analysis of the same transaction data through a joint probability density function, so as to determine whether the transaction data is abnormal, improve the accuracy of transaction data anomaly judgment, and improve the work efficiency of transaction data processing.

[0024] Accordingly, the present invention also provides a transaction data verification device, comprising: an interception module, a scenario construction module and an abnormality judgment module;

[0025] The interception module is used to intercept the first transaction data in the first business system, and perform type analysis on the first transaction data to generate a corresponding business type; the first transaction data carries a corresponding user ID; the first transaction data is generated by the user initiating a business request to the business system;

[0026] The scenario construction module is used to obtain corresponding transaction elements according to the business type of the second transaction data identified by the same user ID, and generate a target business scenario; extract features of the second transaction data identified by the same user ID to generate scenario feature parameters; match the corresponding predicted business scenario according to the scenario feature parameters; and store the predicted business scenario in the scenario database;

[0027] The abnormality judgment module is used to judge whether the second transaction data is abnormal based on the correlation value between the target business scenario and the predicted business scenario; if so, send abnormal reminder information to the first business system to notify the corresponding business personnel to modify the configuration parameters of the second transaction data.

[0028] Further, the scenario construction module includes: a target business scenario construction unit and a predicted business scenario construction unit;

[0029] The target business scenario construction unit is used to determine the transaction element corresponding to each business type in the second transaction data by searching for a mapping relationship between the business type of the transaction data and the transaction element according to the business type set of the second transaction data identified by the same user ID; the transaction element includes: one or more of transaction time, transaction cycle, transaction channel, transaction method and transaction subject;

[0030] Inputting the business type and the corresponding transaction elements into a scenario construction model; generating at least one target business scenario that meets the conditions according to each combination of the business type and the corresponding transaction elements;

[0031] The predicted business scenario construction unit is used to extract the feature expression vector of the second transaction data identified by the same user ID, call the classification model to classify and identify the feature expression vector, generate a scenario type label for the second transaction data, and use the scenario type label and the business parameter value in the second transaction data as the scenario feature parameter; the classification model is trained by a multi-label classifier on the feature expression vector of the historical transaction data and its corresponding scenario type label;

[0032] Calculate the similarity between the scenario feature parameters and the scenario features in the predicted business scenarios in the scenario database, and filter out the predicted business scenarios whose similarity values ​​meet the preset range.

[0033] As a preferred solution, this application intercepts transaction data in the business system through an interception module, and performs type analysis on the transaction data, generates theoretically feasible target business scenarios for the transaction data through a scenario construction module, and generates practically feasible predicted business scenarios through transaction data feature extraction and matching; the two business scenarios generated by the theoretical and actual analysis of the same transaction data are analyzed through an abnormal judgment module to determine whether the transaction data is abnormal. In addition, this application identifies the transaction data of the same user id, and realizes the construction and abnormal analysis of business scenarios for the same user, which can not only improve the accuracy of scenario construction, but also quickly lock abnormal users based on abnormal transaction data; when abnormal data is judged, by modifying the configuration parameters of the transaction data, the error rate of transaction data analysis with different configuration parameters can be reduced, which can improve the accuracy of transaction data and improve the work efficiency of transaction data processing.

[0034] Correspondingly, an embodiment of the present invention further provides a computer device, including a processor and a memory, wherein the memory is used to store a computer program, and the computer program implements the transaction data verification method when executed by the processor.

[0035] Correspondingly, the present invention also provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is running, it controls the device where the computer-readable storage medium is located to execute a transaction data verification method as described in the content of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of an embodiment of the transaction data verification method provided by the present invention;

[0037] Figure 2It is a structural schematic diagram of an embodiment of a transaction data verification device provided by the present invention. DETAILED DESCRIPTION

[0038] 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.

[0039] Embodiment 1

[0040] Please refer to Figure 1 , a transaction data verification method provided by an embodiment of the present invention, comprising steps S101-S103:

[0041] Step S101: intercepting first transaction data in a first business system, and performing type analysis on the first transaction data to generate a corresponding business type; the first transaction data carries a corresponding user ID; the first transaction data is generated when a user initiates a business request to the business system;

[0042] Furthermore, the first transaction data in the first business system is intercepted, and the type of the first transaction data is analyzed to generate a corresponding business type, specifically:

[0043] When a business request event is received in the first business system within a preset time period, the user ID and the corresponding first transaction data in the business request event are collected, and the user ID is identified in the corresponding first transaction data; the second transaction data identified by the same user ID is classified, and a business type set is generated for the second transaction data identified by each user ID.

[0044] In this embodiment, within a preset time period, when different users use different user IDs to initiate business requests, corresponding transaction data is generated in the business system; the transaction data includes transaction time, user ID, transaction content and transaction method, etc.; the transaction data within the preset time period is intercepted, and the transaction data is identified according to the user ID, and the transaction data identified by the same user ID is classified into one category for abnormal analysis and processing.

[0045] In this embodiment, the business type of the transaction data is determined according to the business request corresponding to each transaction data, and a business type set is generated for the second transaction data identified by each user ID, including at least one business type for the second transaction data identified by the same user ID.

[0046] Step S102: according to the business type of the second transaction data identified by the same user ID, the corresponding transaction elements are obtained, and a target business scenario is generated; the second transaction data identified by the same user ID is subjected to feature extraction to generate scenario feature parameters; according to the scenario feature parameters, the corresponding predicted business scenario is matched; and the predicted business scenario is stored in a scenario database;

[0047] Furthermore, according to the business type of the second transaction data identified by the same user ID, the corresponding transaction elements are obtained, and a target business scenario is generated, specifically:

[0048] According to the business type set of the second transaction data identified by the same user ID, by searching for a mapping relationship between the business type of the transaction data and the transaction element, determining the transaction element corresponding to each business type in the second transaction data; the transaction element includes: one or more of transaction time, transaction cycle, transaction channel, transaction method and transaction subject;

[0049] The business type and the corresponding transaction elements are input into a scenario construction model; and at least one target business scenario that meets the conditions is generated according to each combination of the business type and the corresponding transaction elements.

[0050] Before obtaining the corresponding transaction elements, the mapping rules between business types and transaction elements are defined in the database; the corresponding mapping rules are set for each business type data;

[0051] According to the business type set of the second transaction data identified by the same user ID, the business type set includes at least one business type. Each business type data in the business type set is matched with a corresponding transaction element according to a mapping rule, and each business type matches at least one transaction element to generate a transaction element set of the second transaction data identified by the same user ID.

[0052] Furthermore, according to each combination of the business type and the corresponding transaction elements, at least one target business scenario that meets the conditions is generated, specifically:

[0053] Determine the model parameters of the scenario construction model according to the business type and the corresponding transaction elements; cluster and fit the model parameters according to the historical transaction data and the preset transaction rules to generate combination information of each model parameter combination; generate scenario configuration data for each target business scenario based on each combination information; determine the configuration business parameters of each target business scenario based on the scenario configuration data, and construct the corresponding target business scenario.

[0054] In this embodiment, the model parameters of the scenario construction model are set according to the transaction element set of the second transaction data identified by the same user ID, and the scenario construction model is used to construct the transaction scenario of the second transaction data identified by the same user ID. The clustering algorithm is used to select different cluster centers in turn according to the historical transaction data and the preset transaction rules to cluster each model parameter combination, calculate the distance from each cluster sample to each cluster center, and update the cluster center according to the distance; the distance between the cluster sample and the cluster center is calculated by continuous iteration until the cluster center no longer changes or the change is lower than a certain threshold, and the output data is fitted and analyzed to obtain the combination information of each model parameter combination. In this process, according to the historical transaction data and the preset transaction rules, some model parameter combinations corresponding to the impossible transaction scenarios are eliminated, which reduces the workload of model calculation and improves the accuracy of scenario construction.

[0055] Furthermore, feature extraction is performed on the second transaction data identified by the same user ID to generate scenario feature parameters, and the corresponding predicted business scenario is matched according to the scenario feature parameters, specifically:

[0056] Extracting a feature expression vector of second transaction data identified by the same user ID, calling a classification model to classify and identify the feature expression vector, generating a scene type label for the second transaction data, and using the scene type label and the business parameter value in the second transaction data as a scene feature parameter; the classification model is trained by a multi-label classifier on the feature expression vectors of historical transaction data and their corresponding scene type labels;

[0057] Calculate the similarity between the scenario feature parameters and the scenario features in the predicted business scenarios in the scenario database, and filter out the predicted business scenarios whose similarity values ​​meet the preset range.

[0058] In this embodiment, the multi-label classifier is trained according to the feature expression vector of the historical transaction data and its corresponding scene type annotation: the historical transaction data is input into the initial multi-label classifier, the predicted annotation value of each annotation of the training sample is output, the error between the predicted annotation value and the true annotation value is calculated, and the parameters of each node of the initial multi-label classifier are modified until the error between the calculated predicted annotation value and the true annotation value is lower than the preset threshold, and the trained multi-label classifier is obtained. The multi-label classifier is used to input the feature expression vector into each node, and each path of the leaf node is a combination of different features. Each combination is predicted and analyzed to determine the scene type, and at least one scene type that conforms to the transaction data is generated; finally, the corresponding scene feature parameters are generated in combination with the business parameter value in the second transaction data;

[0059] Step S103: judging whether the second transaction data is abnormal according to the correlation value between the target business scenario and the predicted business scenario; if so, sending abnormal reminder information to the first business system to notify the corresponding business personnel to modify the configuration parameters of the second transaction data.

[0060] Further, judging whether the second transaction data is abnormal according to the correlation value between the target business scenario and the predicted business scenario is specifically as follows:

[0061] The business scenario correlation value calculation model is used to calculate the correlation value of each target business scenario and each predicted business scenario in turn through the joint probability density function; when the correlation value is greater than the first threshold, the calculation of the correlation value is stopped, and a normal result of the second transaction data is generated; when all correlation values ​​are not greater than the first threshold, an abnormal result of the second transaction data is generated.

[0062] Implementing the embodiments of the present invention has the following effects:

[0063] This application generates theoretically feasible target business scenarios by performing type analysis on transaction data, and generates practically feasible prediction business scenarios by extracting and matching transaction data features; by performing correlation analysis on the two business scenarios generated by theoretical and practical analysis of the same transaction data, it is possible to determine whether the transaction data is abnormal. In addition, this application identifies the transaction data of the same user ID, and realizes the construction and abnormal analysis of business scenarios for the same user, which can not only improve the accuracy of scenario construction, but also quickly lock abnormal users based on abnormal transaction data; when abnormal data is determined, by modifying the configuration parameters of the transaction data, the error rate of transaction data analysis with different configuration parameters can be reduced, which can improve the accuracy of transaction data and improve the work efficiency of transaction data processing.

[0064] Embodiment 2

[0065] Please refer to Figure 2 , a transaction data verification device provided by an embodiment of the present invention, comprising: an interception module 201, a scenario construction module 202 and an abnormality judgment module 203;

[0066] The interception module 201 is used to intercept the first transaction data in the first business system, and perform type analysis on the first transaction data to generate a corresponding business type; the first transaction data carries a corresponding user ID; the first transaction data is generated by the user initiating a business request to the business system;

[0067] The scenario construction module 202 is used to obtain corresponding transaction elements according to the business type of the second transaction data identified by the same user ID, and generate a target business scenario; extract features of the second transaction data identified by the same user ID to generate scenario feature parameters; match the corresponding predicted business scenario according to the scenario feature parameters; and store the predicted business scenario in the scenario database;

[0068] The abnormality judgment module 203 is used to judge whether the second transaction data is abnormal based on the correlation value between the target business scenario and the predicted business scenario; if so, an abnormal reminder message is sent to the first business system to notify the corresponding business personnel to modify the configuration parameters of the second transaction data.

[0069] The interception module 201 includes: a collection unit and a classification unit;

[0070] The collecting unit is used for collecting the user ID and the corresponding first transaction data in the business request event when receiving the business request event in the first business system within a preset time period, and marking the user ID in the corresponding first transaction data;

[0071] The classification unit is used to classify the second transaction data identified by the same user ID, and generate a business type set for the second transaction data identified by each user ID.

[0072] The scenario construction module 202 includes: a target business scenario construction unit and a predicted business scenario construction unit;

[0073] The target business scenario construction unit is used to determine the transaction element corresponding to each business type in the second transaction data by searching for a mapping relationship between the business type of the transaction data and the transaction element according to the business type set of the second transaction data identified by the same user ID; the transaction element includes: one or more of transaction time, transaction cycle, transaction channel, transaction method and transaction subject;

[0074] Inputting the business type and the corresponding transaction elements into a scenario construction model; generating at least one target business scenario that meets the conditions according to each combination of the business type and the corresponding transaction elements;

[0075] The predicted business scenario construction unit is used to extract the feature expression vector of the second transaction data identified by the same user ID, call the classification model to classify and identify the feature expression vector, generate a scenario type label for the second transaction data, and use the scenario type label and the business parameter value in the second transaction data as the scenario feature parameter; the classification model is trained by a multi-label classifier on the feature expression vector of the historical transaction data and its corresponding scenario type label;

[0076] Calculate the similarity between the scenario feature parameters and the scenario features in the predicted business scenarios in the scenario database, and filter out the predicted business scenarios whose similarity values ​​meet the preset range.

[0077] The abnormality judgment module 203 includes: a calculation unit and a judgment unit;

[0078] The calculation unit is used to use the business scenario relevance value calculation model to calculate the relevance value of each target business scenario and each predicted business scenario in turn through a joint probability density function;

[0079] The judgment unit is used to stop calculating the correlation value when the correlation value is greater than a first threshold, thereby generating a normal result of the second transaction data; when all correlation values ​​are not greater than the first threshold, then generating an abnormal result of the second transaction data.

[0080] The transaction data verification device described above can implement the transaction data verification method of the method embodiment described above. The options in the method embodiment described above are also applicable to this embodiment and will not be described in detail here. The rest of the contents of the embodiment of this application can refer to the contents of the method embodiment described above and will not be described in detail in this embodiment.

[0081] In addition, an embodiment of the present application further provides a computer device, which includes a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, the steps in any of the above method embodiments are implemented.

[0082] Implementing the embodiments of the present invention has the following effects:

[0083] This application intercepts transaction data in the business system through an interception module, and performs type analysis on the transaction data. It generates theoretically feasible target business scenarios for the transaction data through a scenario construction module, and generates practically feasible prediction business scenarios through transaction data feature extraction and matching; the two business scenarios generated by the theoretical and actual analysis of the same transaction data are analyzed through an abnormal judgment module to determine whether the transaction data is abnormal. In addition, this application identifies the transaction data of the same user id, and realizes the construction and abnormal analysis of business scenarios for the same user, which can not only improve the accuracy of scenario construction, but also quickly lock abnormal users based on abnormal transaction data; when abnormal data is judged, by modifying the configuration parameters of the transaction data, the error rate of transaction data analysis with different configuration parameters can be reduced, which can improve the accuracy of transaction data and improve the work efficiency of transaction data processing.

[0084] Embodiment 3

[0085] Correspondingly, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the transaction data verification method described in any one of the above embodiments.

[0086] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the terminal device.

[0087] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0088] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0089] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the terminal device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc.; the data storage area can store data created according to the use of the mobile terminal, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0090] Wherein, if the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0091] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A transaction data verification method, It is characterized in that include: Intercepting first transaction data in a first business system, and performing type analysis on the first transaction data to generate a corresponding business type, specifically: When a service request event is received in the first service system within a preset time period, a user ID and corresponding first transaction data in the service request event are collected, and the user ID is identified in the corresponding first transaction data; Classifying the second transaction data identified by the same user ID, and generating a business type set for the second transaction data identified by each user ID; the first transaction data carries the corresponding user ID; the first transaction data is generated by the user initiating a business request to the business system; According to the business type of the second transaction data identified by the same user ID, the corresponding transaction elements are obtained, and the target business scenario is generated, specifically: According to the business type set of the second transaction data identified by the same user ID, by searching for a mapping relationship between the business type of the transaction data and the transaction element, determining the transaction element corresponding to each business type in the second transaction data; the transaction element includes: one or more of transaction time, transaction cycle, transaction channel, transaction method and transaction subject; Inputting the business type and corresponding transaction elements into the scenario building model; According to each combination of the business type and the corresponding transaction elements, at least one target business scenario that meets the conditions is generated, specifically: Determine the model parameters of the scenario construction model according to the business type and the corresponding transaction elements; cluster and fit the model parameters according to the historical transaction data and the preset transaction rules to generate the combination information of each model parameter combination; generate the scenario configuration data of each target business scenario according to each combination information; determine the configuration business parameters of each target business scenario according to the scenario configuration data, and construct the corresponding target business scenario; Feature extraction is performed on the second transaction data identified by the same user ID to generate scenario feature parameters; and the corresponding predicted business scenario is matched according to the scenario feature parameters, specifically: Extracting a feature expression vector of second transaction data identified by the same user ID, calling a classification model to classify and identify the feature expression vector, generating a scene type label for the second transaction data, and using the scene type label and the business parameter value in the second transaction data as a scene feature parameter; the classification model is trained by a multi-label classifier on the feature expression vectors of historical transaction data and their corresponding scene type labels; Calculating the similarity between the scenario feature parameters and the scenario features in the predicted business scenarios in the scenario database, and screening out the predicted business scenarios whose similarity values ​​meet a preset range; the predicted business scenarios are stored in the scenario database; According to the correlation value between the target business scenario and the predicted business scenario, it is determined whether the second transaction data is abnormal; if so, an abnormal reminder message is sent to the first business system to notify the corresponding business personnel to modify the configuration parameters of the second transaction data.

2. A transaction data verification method as claimed in claim 1, It is characterized in that The determining whether the second transaction data is abnormal according to the correlation value between the target business scenario and the predicted business scenario is specifically: Using the business scenario correlation value calculation model, the correlation value of each target business scenario and each predicted business scenario is calculated in turn through the joint probability density function; when the correlation value is greater than the first threshold, the calculation of the correlation value is stopped, and a normal result of the second transaction data is generated; When all the correlation values ​​are not greater than the first threshold, a second transaction data abnormal result is generated.

3. A transaction data verification device, It is characterized in that include: Interception module, scenario construction module and abnormality judgment module; The interception module is used to intercept the first transaction data in the first business system, and perform type analysis on the first transaction data to generate a corresponding business type, specifically: When a service request event is received in the first service system within a preset time period, a user ID and corresponding first transaction data in the service request event are collected, and the user ID is identified in the corresponding first transaction data; second transaction data identified by the same user ID are classified, and a service type set is generated for each second transaction data identified by the user ID; the first transaction data carries the corresponding user ID identification; the first transaction data is generated when the user initiates a service request to the service system; The scenario construction module is used to obtain corresponding transaction elements according to the business type of the second transaction data identified by the same user ID, and generate a target business scenario; extract features of the second transaction data identified by the same user ID to generate scenario feature parameters; match the corresponding predicted business scenario according to the scenario feature parameters; and store the predicted business scenario in the scenario database; The scenario construction module includes: a target business scenario construction unit and a predicted business scenario construction unit; The target business scenario construction unit is used to determine the transaction element corresponding to each business type in the second transaction data by searching for a mapping relationship between the business type of the transaction data and the transaction element according to the business type set of the second transaction data identified by the same user ID; the transaction element includes: one or more of transaction time, transaction cycle, transaction channel, transaction method and transaction subject; The business type and the corresponding transaction elements are input into the scenario construction model; and according to each combination of the business type and the corresponding transaction elements, at least one target business scenario that meets the conditions is generated, specifically: Determine the model parameters of the scenario construction model according to the business type and the corresponding transaction elements; cluster and fit the model parameters according to the historical transaction data and the preset transaction rules to generate the combination information of each model parameter combination; generate the scenario configuration data of each target business scenario according to each combination information; determine the configuration business parameters of each target business scenario according to the scenario configuration data, and construct the corresponding target business scenario; The predicted business scenario construction unit is used to extract the feature expression vector of the second transaction data identified by the same user ID, call the classification model to classify and identify the feature expression vector, generate a scenario type label for the second transaction data, and use the scenario type label and the business parameter value in the second transaction data as the scenario feature parameter; the classification model is trained by a multi-label classifier on the feature expression vector of the historical transaction data and its corresponding scenario type label; Calculate the similarity between the scenario feature parameters and the scenario features in the predicted business scenarios in the scenario database, and select the predicted business scenarios whose similarity values ​​meet the preset range; The abnormality judgment module is used to judge whether the second transaction data is abnormal based on the correlation value between the target business scenario and the predicted business scenario; if so, send abnormal reminder information to the first business system to notify the corresponding business personnel to modify the configuration parameters of the second transaction data.

4. A transaction data verification device, It is characterized in that It comprises a processor and a memory, the memory is used to store a computer program, and when the computer program is executed by the processor, the transaction data verification method according to any one of claims 1 to 2 is implemented.

5. A computer-readable storage medium, It is characterized in that The computer-readable storage medium includes a stored computer program; wherein, when the computer program is run, it controls the device where the computer-readable storage medium is located to execute a transaction data verification method as described in any one of claims 1 to 2.