Transaction defect prediction method and device, equipment, medium and program product

Abstract

The invention provides a transaction defect prediction method which can be applied to the technical field of transaction risk management and can also be applied to the fields of artificial intelligence, financial science and technology or other fields. The transaction defect prediction method comprises the steps that target transaction features are acquired, the association degree of the target transaction features and transaction defects is larger than a preset threshold value, and the target transaction features comprise transaction system test environment features and transaction system development features; the transaction system test environment features are input into a first prediction model to obtain a first prediction result, and the first prediction result represents the occurrence probability of transaction defects by the transaction system test environment; the transaction system development features are input into a second prediction model to obtain a second prediction result, and the second prediction result represents the occurrence probability of transaction defects by the transaction system development data; and determining a target prediction result according to the first prediction result and the second prediction result. The invention further provides a transaction defect prediction device and equipment, a storage medium and a program product.

Description

Technical Field

[0001] This disclosure relates to the fields of transaction risk management and artificial intelligence, and specifically to a method, apparatus, equipment, medium, and program product for predicting transaction defects. Background Technology

[0002] With the rapid development of the financial industry, online bank counter transactions have become an important part of banking services. Since these transactions directly interact with customers, the accuracy of their processing is crucial. If a common bank counter transaction (such as an account opening) fails to complete successfully, it can lead to complaints and other negative consequences. However, during online bank counter transactions, various reasons (such as system malfunctions or human error) can cause defects to be exposed, resulting in losses for the bank.

[0003] Traditional over-the-counter transaction defect prediction methods mainly rely on transaction usage frequency, transaction complexity, and development version quality to build corresponding models to consider relevant factors before transaction delivery testing, but the accuracy of the prediction is low. Summary of the Invention

[0004] In view of the above problems, this disclosure provides a method, apparatus, device, medium and program product for predicting transaction defects.

[0005] According to a first aspect of this disclosure, a method for predicting transaction defects is provided. The method includes: acquiring target transaction features, wherein the correlation between the target transaction features and transaction defects is greater than a preset threshold, and the target transaction features include transaction system test environment features and transaction system development features; inputting the transaction system test environment features into a first prediction model to obtain a first prediction result, wherein the first prediction result characterizes the probability of transaction defects occurring in the transaction system test environment; inputting the transaction system development features into a second prediction model to obtain a second prediction result, wherein the second prediction result characterizes the probability of transaction defects occurring in the transaction system development data; and determining a target prediction result based on the first prediction result and the second prediction result.

[0006] According to an embodiment of this disclosure, inputting the characteristics of the trading system test environment into a first prediction model to obtain a first prediction result includes: obtaining a first weight value, wherein the first weight value characterizes the degree of influence of each trading system test environment characteristic on trading defects; and determining the first prediction result based on multiple trading system test environment characteristics and the first weight value.

[0007] According to an embodiment of this disclosure, inputting the development characteristics of the trading system into a second prediction model to obtain a second prediction result includes: determining the second prediction result based on the development characteristics of the trading system.

[0008] According to embodiments of this disclosure, determining a target prediction result based on a first prediction result and a second prediction result includes: obtaining a second weight value corresponding to the first prediction result and a third weight value corresponding to the second prediction result; and determining the target prediction result based on the first prediction result, the second weight value, the second prediction result, and the third weight value.

[0009] According to an embodiment of this disclosure, the first prediction model is obtained through the following operations: acquiring historical transaction features, the historical transaction data including historical system test environment features and the probability of occurrence of a first actual transaction defect when conducting transactions using a transaction system with historical system test environment features; inputting the historical system test environment features into the first prediction model to obtain a first predicted probability of occurrence of a transaction defect; adjusting the parameters of the first prediction model according to the difference between the first predicted probability of occurrence of a transaction defect and the probability of occurrence of a first actual transaction defect until the difference is minimized, thereby obtaining the first prediction model.

[0010] According to embodiments of this disclosure, the second prediction model is obtained through the following operations: acquiring historical transaction features, including historical transaction system development features and the probability of occurrence of actual transaction defects when using a transaction system with transaction system development features; inputting the historical transaction system development features into the second prediction model to obtain the second predicted probability of occurrence of transaction defects; adjusting the parameters of the first prediction model according to the difference between the first predicted probability of occurrence of transaction defects and the second actual probability of occurrence of transaction defects until the difference is minimized, thereby obtaining the second prediction model.

[0011] According to embodiments of this disclosure, the method further includes generating an early warning prompt when the probability of occurrence of a transaction defect represented by the target prediction result is greater than a target threshold.

[0012] According to embodiments of this disclosure, the test environment characteristics include: the last delivery time of the test environment program of the trading system, the cumulative number of test environment deliveries of the trading system, the time since the last successful transaction in the test environment of the trading system, the cumulative number of successful transactions of the trading system, the cumulative number of successful calls to production transactions of the trading system, the number of defects that occurred during production operation of the trading system, the number of defects discovered during testing of the trading system, the number of branch scenarios involved in the transaction, the trading currency, and the transaction type; the development characteristics of the trading system include: the development experience of the trading system developers.

[0013] A second aspect of this disclosure provides a transaction defect prediction device, comprising: an acquisition module for acquiring target transaction features, wherein the correlation between the target transaction features and transaction defects is greater than a preset threshold, and the target transaction features include transaction system test environment features and transaction system development features; a first prediction module for inputting the transaction system test environment features into a first prediction model to obtain a first prediction result, wherein the first prediction result characterizes the probability of transaction defects occurring in the transaction system test environment; a second prediction module for inputting transaction system development features into a second prediction model to obtain a second prediction result, wherein the second prediction result characterizes the probability of transaction defects occurring in the transaction system development data; and a determination module for determining a target prediction result based on the first prediction result and the second prediction result.

[0014] A third aspect of this disclosure provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0015] A fourth aspect of this disclosure also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0016] The fifth aspect of this disclosure also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method. Attached Figure Description

[0017] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0018] Figure 1 The illustration shows an application scenario of the transaction defect prediction method, apparatus, device, medium, and program product according to embodiments of the present disclosure;

[0019] Figure 2 A flowchart illustrating a transaction defect prediction method according to an embodiment of the present disclosure is shown schematically.

[0020] Figure 3 This schematically illustrates a flowchart of a method for inputting features of a trading system test environment into a first prediction model to obtain a first prediction result, according to an embodiment of the present disclosure.

[0021] Figure 4 A flowchart illustrating a method for determining a target prediction result based on a first prediction result and a second prediction result according to an embodiment of the present disclosure is shown schematically.

[0022] Figure 5A flowchart illustrating a first prediction model training method according to an embodiment of the present disclosure is shown schematically.

[0023] Figure 6 A flowchart illustrating a second prediction model training method according to an embodiment of the present disclosure is shown schematically.

[0024] Figure 7 A flowchart illustrating a transaction defect prediction method according to another embodiment of this disclosure is shown schematically;

[0025] Figure 8 A schematic diagram illustrating the structure of a transaction defect prediction apparatus according to an embodiment of the present disclosure is shown; and

[0026] Figure 9 A block diagram of an electronic device suitable for implementing a transaction defect prediction method according to an embodiment of the present disclosure is shown schematically. Detailed Implementation

[0027] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0028] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0029] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0030] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0031] In the technical solution disclosed herein, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation entry points are provided for users to choose to authorize or refuse.

[0032] In scenarios involving automated decision-making using personal information, the methods, devices, and systems provided in this disclosure all offer users corresponding entry points for choosing to agree to or reject the automated decision-making results. If the user chooses to reject, the process proceeds to the expert decision-making stage. Here, "automated decision-making" refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests, or economic, health, and credit status through computer programs, and then making a decision. Here, "expert decision-making" refers to the activity of making decisions by personnel who specialize in a particular field, possess specialized experience, knowledge, and skills, and have reached a certain level of professional expertise.

[0033] This disclosure provides a method, apparatus, device, medium, and program product for predicting transaction defects. Before introducing the technical solutions provided by this disclosure, the relevant technologies involved in this disclosure will be explained.

[0034] With the rapid development of the financial industry, online bank counter transactions have become an important part of banking services. Since these transactions directly interact with customers, the accuracy of their processing is crucial. If a common bank counter transaction (such as an account opening) fails to complete successfully, it can lead to complaints and other negative consequences. However, during online bank counter transactions, various reasons (such as system malfunctions or human error) can cause defects to be exposed, resulting in losses for the bank.

[0035] Traditional over-the-counter transaction defect prediction methods mainly rely on transaction usage frequency, transaction complexity, and development version quality to build corresponding models to consider relevant factors before transaction delivery testing, but the accuracy of the prediction is low.

[0036] This disclosure provides a method for predicting transaction defects. The method includes: acquiring target transaction features, wherein the correlation between the target transaction features and transaction defects is greater than a preset threshold; the target transaction features include transaction system test environment features and transaction system development features; inputting the transaction system test environment features into a first prediction model to obtain a first prediction result, the first prediction result representing the probability of transaction defects occurring in the transaction system test environment; inputting the transaction system development features into a second prediction model to obtain a second prediction result, the second prediction result representing the probability of transaction defects occurring in the transaction system development data; and determining a target prediction result based on the first and second prediction results. This method combines different dimensions of transaction system features and development features to predict the probability of transaction defects occurring. It enables real-time monitoring and early warning of transactions, helping financial institutions to promptly detect and address transaction defect leaks, thereby improving transaction security and efficiency.

[0037] Figure 1 The illustration shows an application scenario of the transaction defect prediction method, apparatus, device, medium, and program product according to embodiments of the present disclosure.

[0038] like Figure 1 As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0039] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0040] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0041] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0042] It should be noted that the transaction defect prediction method provided in this embodiment can generally be executed by server 105. Correspondingly, the transaction defect prediction device provided in this embodiment can generally be located in server 105. The transaction defect prediction method provided in this embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the transaction defect prediction device provided in this embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105.

[0043] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0044] The following will be based on Figure 1 The described scene, through Figures 2-7 The transaction defect prediction method of the disclosed embodiments is described in detail.

[0045] Figure 2 A flowchart illustrating a transaction defect prediction method according to an embodiment of the present disclosure is shown schematically.

[0046] like Figure 2 As shown, the transaction defect prediction method of this embodiment includes operations S210 to S240.

[0047] In operation S210, target transaction features are obtained. The correlation between the target transaction features and transaction defects is greater than a preset threshold. The target transaction features include transaction system test environment features and transaction system development features.

[0048] In operation S220, the characteristics of the trading system test environment are input into the first prediction model to obtain the first prediction result. The first prediction result represents the probability of the occurrence of trading defects in the trading system test environment.

[0049] In operation S230, the characteristics of the trading system development are input into the second prediction model to obtain the second prediction result, which represents the probability of the occurrence of trading defects due to the trading system development data.

[0050] In operation S240, the target prediction result is determined based on the first prediction result and the second prediction result.

[0051] For example, the transaction can be a banking transaction, a securities or investment transaction, an e-commerce transaction, etc. This disclosure embodiment uses online bank counter transactions as an example of banking transactions. Online bank counter transactions refer to digital financial transactions completed by customers at bank physical branches (counters) through teller operations, connecting to the bank's core system in real time.

[0052] A transaction system refers to an integrated hardware and software platform used to execute, manage, and monitor online bank counter transactions. When the transaction is an online bank counter transaction, the transaction system can be the bank's core system.

[0053] Target transaction features can be transaction characteristics that are significantly related to transaction defects and selected from raw transaction data. Raw data includes transaction records and user information, and further subdivisions can include historical transaction data, user behavior data, and transaction environment data. Target transaction features can be determined through evaluation by experts in the field.

[0054] The target trading characteristics include trading system testing environment characteristics and trading system development characteristics. Trading system testing environment characteristics can be data recorded in the testing environment related to the trading execution process. Trading system development characteristics can be data related to code quality during the development phase.

[0055] The first prediction model is used to predict the probability of subsequent trading defects caused by the current characteristics of the trading system test environment, based on the characteristics of the trading system test environment, and to obtain the first prediction result.

[0056] The second prediction model is used to predict the probability of subsequent trading defects caused by the trading system development characteristics, and obtains the second prediction result.

[0057] Based on the predictions of the probability of transaction defects occurring from multiple dimensions (first prediction result and second prediction result), the target prediction result for the probability of transaction defects occurring in the trading system is determined. The system is then used to confirm whether any anomalies exist. If anomalies are found, they are handled by bank staff. If no anomalies are found, normal trading operations are performed.

[0058] Understandably, combining the characteristics of different dimensions of the trading system to predict the probability of trading defects can enable real-time monitoring and early warning of transactions, helping financial institutions to promptly identify and address issues related to leaked trading defects, thereby improving the security and efficiency of transactions.

[0059] In some embodiments, the test environment characteristics include: the last delivery time of the test environment program of the trading system, the cumulative number of test environments delivered by the trading system, the time since the last successful transaction in the test environment of the trading system, the cumulative number of successful transactions of the trading system, the cumulative number of successful calls to production transactions of the trading system, the number of defects that occurred during production operation of the trading system, the number of defects discovered during testing of the trading system, the number of branch scenarios involved in the transaction, the currency of the transaction, and the transaction type.

[0060] For example, the impact of test and production call data analysis based on bank counter transactions on the probability of transaction defects.

[0061] Last delivery time (in days) of the test environment program for the over-the-counter online transaction system: In the bank's testing center, this refers to the time since the most recent patch or version delivery for this transaction. For example, using a 29-day version testing cycle as a dividing point, a delivery time of less than or equal to 29 days indicates a high defect leakage rate for this transaction, while a delivery time greater than 29 days indicates a low defect leakage rate.

[0062] Total number of times the counter online transaction system has been delivered to the test environment: This refers to the delivery records of the transaction program obtained from the bank's version delivery system, including the number of deliveries and the delivery time. A higher number of deliveries indicates stronger demand for the transaction, more frequent changes, a greater likelihood of errors, and a higher defect leakage rate.

[0063] Time since the last successful transaction in the system test environment (days): This is obtained from the bank's test environment transaction logs, including the number of transaction calls, completion time, completion status, request time, request frequency, request method, transaction amount, transaction time, transaction user ID, and success status. Compare this to the last delivery time of the online counter transaction system test environment program. If the time is greater than 0 days, it indicates that test coverage was performed after the patch version was delivered, resulting in a low degree of defect leakage. If the time is less than or equal to 0 days, it indicates that test coverage was not performed after the patch version was delivered, resulting in a high degree of defect leakage.

[0064] The cumulative number of successful transactions in the transaction system (times): This refers to the cumulative number of successful bank counter transactions in the test environment. Data includes completion time, completion status, request time, request frequency, request method, transaction amount, transaction time, transaction user ID, and whether the transaction was successful. A higher number of successful transactions in the test environment indicates more cumulative test coverage of the counter transaction, potentially covering more branch scenarios and reducing the probability of defects. Conversely, a lower number of successful transactions indicates a lower probability of defects.

[0065] Cumulative successful transaction calls (times) in the production system: In the production environment, this refers to the cumulative number of successful bank counter transactions, including completion time, completion status, request time, request frequency, request method, transaction amount, transaction time, transaction user ID, and whether it was successful. In the production environment, a higher number of successful transactions indicates that the counter transaction has been used more frequently in production, potentially covering more branch scenarios and reducing the probability of defects, and vice versa.

[0066] Number of defects encountered during production operation of the transaction system (times): In the production environment, this counter transaction has historically experienced transaction failures due to program defects, which can be found in the production problem database. The more times this counter transaction has historically experienced defects in the production environment, the more frequently the transaction is used, the more complex the transaction function, the lower the quality of the transaction program, and the greater the probability of defects. Conversely, the fewer defects encountered, the lower the probability of defects.

[0067] Number of defects found in the transaction system test (times): This refers to the number of times defects were found by testers in the test environment of this bank's counter transactions, including the time of discovery and user ID. The more defects testers find in the counter transaction program in the test environment before the transaction system version is delivered to production, the lower the version quality and the greater the probability of defects recurring.

[0068] The number of branch scenarios involved in a transaction in a transaction system is collected separately from transaction data, scenarios, and processes. If there are D different types of data, C types of scenarios, and S types of processes, then this transaction requires at least D*C*S types of business processing. Each process may use different program assemblies. A transaction with many fields, and each field's dropdown list data dictionary is complex. The more D*C*S scenarios a transaction has, the more fields it has, and the more dropdown list data dictionaries it has, the higher the transaction complexity and the greater the probability of defects.

[0069] Transaction currency (number of currencies). The number of currencies that this region or banking institution can offer in this region. The more currencies that this region or banking institution can offer in this region, the greater the probability of a transaction procedure defect.

[0070] Transaction type. This includes queries, adjustments, additions, and deletions. Different weights are assigned based on the transaction type (query, adjustment, addition, deletion, etc.), representing varying transaction complexity. Higher weights increase the probability of transaction defects. For example, query transactions have a weight of 20%, adjustment transactions 80%, addition transactions 100%, and deletion transactions 60%.

[0071] The various features are interconnected and influence each other. Each feature variable has a certain impact on the occurrence of transaction defects, and the impact varies under different circumstances. The relationship between the probability of transaction defects and the feature factors is non-linear, requiring the use of non-linear regression methods to resolve.

[0072] In some embodiments, the trading system development characteristics include: the number of years the trading system developers have worked.

[0073] For example, considering that the quality of the developed and delivered version of a bank counter transaction is a key factor determining the number of defects in that transaction, thus increasing prediction accuracy, we add factors related to defect leakage from the counter transaction development and delivery end. We select the length of time the developer responsible for the transaction has been in charge as a key factor for analysis, as this factor has a significantly greater impact on transaction quality than other factors. Generally speaking, the longer the developer has been in charge of the transaction, the lower the probability of defect leakage; conversely, the shorter the developer's length of charge, the higher the probability of defect leakage, showing a relatively clear linear relationship.

[0074] Figure 3 The flowchart illustrates a method for inputting characteristics of a trading system test environment into a first prediction model to obtain a first prediction result, according to an embodiment of the present disclosure.

[0075] As described above, in operation S220, the characteristics of the trading system test environment are input into the first prediction model to obtain the first prediction result. In one possible implementation, such as... Figure 3 As shown, this operation may further include operations S310 to S320.

[0076] In operation S310, the first weight value is obtained. The first weight value represents the degree of influence of each trading system test environment feature on trading defects.

[0077] In operation S320, the first prediction result is determined based on the characteristics of multiple trading system test environments and the first weight value.

[0078] In one example, for the nonlinear relationship between features of the trading system test environment, the first prediction model can employ a multiple logistic regression model. The current-time trading system test environment features are input into the multiple logistic regression model. The model first analyzes the test environment features to obtain the weights (first weight values) corresponding to each feature, and then determines the first prediction result based on the parameters and weights of each feature. See details for further information. Figure 5 The first prediction model training method in China.

[0079] As described above, in operation S230, the characteristics of the trading system are input into the second prediction model to obtain the second prediction result. In one possible implementation, such as... Figure 4 As shown, this operation may further include the operation of determining a second prediction result based on the characteristics of the trading system development.

[0080] In one example, since the development characteristics of the trading system are linearly related to the probability of trading defects, a second prediction model can be used. The development characteristics of the trading system at the current moment are input into the univariate linear regression model, which analyzes these characteristics to obtain the second prediction result. See details for further information. Figure 6 The first prediction model training method in China.

[0081] Figure 4 A flowchart illustrating a method for processing a first image to obtain a second image according to an embodiment of the present disclosure is shown.

[0082] As described above, in operation S240, the target prediction result is determined based on the first and second prediction results. In one possible implementation, such as... Figure 4 As shown, this operation may further include operations S410 to S420.

[0083] In operation S410, the second weight value corresponding to the first prediction result and the third weight value corresponding to the second prediction result are obtained.

[0084] In operation S420, the target prediction result is determined based on the first prediction result, the second weight value, the second prediction result, and the third weight value.

[0085] In one example, based on the importance principle of different types of features, the first prediction result of the multivariate logistic regression model is assigned a weight of 60%, and the second prediction result of the univariate linear regression model is assigned a weight of 40%. The following formula is constructed to finally comprehensively predict the leakage of transaction defects (target prediction result).

[0086] M = 60%Pt + 40%Yt

[0087] Wherein, Pt is the first prediction result, that is, the probability of the influence of the test environment characteristics of the trading system on the trading defects, and Yt is the second prediction result, that is, the probability of the influence of the open characteristics of the trading system on the trading defects.

[0088] It should be noted that the embodiments disclosed herein do not impose specific limitations on the values ​​of the second and third weight values, and the magnitudes of the second and third weight values ​​can be adjusted according to actual needs.

[0089] Figure 5A flowchart illustrating a first prediction model training method according to an embodiment of the present disclosure is shown.

[0090] As mentioned above, Figure 5 As shown, the first prediction model is obtained through the following operations, including operations S510 to S530.

[0091] When operating S510, historical transaction characteristics are acquired. Historical transaction data includes historical system test environment characteristics and the probability of occurrence of the first actual transaction defect when trading is conducted using a trading system with historical system test environment characteristics.

[0092] When operating S520, the characteristics of the historical system test environment are input into the first prediction model to obtain the first predicted probability of the occurrence of communication defects.

[0093] In operation S530, based on the difference between the probability of occurrence of the first predicted trading defect and the probability of occurrence of the first actual trading defect, the parameters of the first prediction model are adjusted until the difference is minimized, thus obtaining the first prediction model.

[0094] In one example, a bank has 2,000 SOCT counter online transactions. Due to new business needs or technical upgrades, a certain version is scheduled for system function testing. The version program delivery date is January 5, 2024, the version delivery and production launch date is February 3, and the testing period is 29 days.

[0095] For training the multivariate logistic regression model (testing and production), 2,000 over-the-counter transactions were acquired using model tools deployed on testing and production scenarios. This data, accumulated over nearly 20 years with 12 versions released annually (totaling 48,000 data points, of which 40,000 were used for training and 8,000 for testing), was automatically obtained from the system's raw data, as shown in Table 1.

[0096] Table 1

[0097]

[0098] The raw data is preprocessed and cleaned, including removing duplicate data, missing data, outliers, etc., to facilitate input into subsequent machine learning algorithms.

[0099] Features that influence the degree of defect leakage were extracted from the raw data, such as transaction amount, transaction time, trader, and transaction type, as shown in Table 2:

[0100]

[0101] Wherein, X1 is the last delivery time (days) of the test environment program for the over-the-counter online trading system; X2 is the cumulative number of test environment deliveries for the over-the-counter online trading system; X3 is the time since the last successful transaction in the test environment of the trading system (days); X4 is the cumulative number of successful transactions in the trading system (times); X5 is the cumulative number of successful calls to the production trading system (times); X6 is the number of defects that occurred during the production operation of the trading system (times); X7 is the number of defects discovered during the testing of the trading system (times); X8 is the number of branch scenarios involved in the trading of the trading system (types); X9 is the number of currencies traded (number); and X10 is the transaction type.

[0102] To obtain the mathematical relationship between the probability of program defects occurring in bank counter transactions and the characteristics of the testing environment, a mathematical model is established using a multiple logistic regression model, with the probability of defect occurrence as the output variable (dependent variable) and the input variables (independent variables). As shown in equation (1):

[0103]

[0104] Wherein, the dependent variable Y: Y=1: A defective transaction occurred. Y=0: No defective transaction occurred. Independent variable... , P represents the probability of a defect occurring (range (0,1)), and the coefficient is... : The parameters that need to be estimated, reflecting the impact of each independent variable (feature) on the defect probability.

[0105] To transform the nonlinear model into a linear form, a log-odds transformation is performed on the probability P:

[0106]

[0107] make The model simplifies to:

[0108]

[0109] Collect m sets of sample data (m > n) to form the following system of equations:

[0110]

[0111] Represent the above equation using a matrix:

[0112] Y=Xβ

[0113] Y is an m×1 column vector (actual value):

[0114]

[0115] X is an m×(n+1) design matrix (independent variable, including intercept term β0):

[0116]

[0117] β is a (n+1)×1 parameter vector:

[0118]

[0119] The parameters are estimated by minimizing the residual sum of squares (the difference between the actual y and the model prediction). The objective function is:

[0120]

[0121]

[0122] in, .

[0123] To find a β value that meets the conditions, take the partial derivative with respect to β and set the derivative to zero:

[0124]

[0125] When k=0 =1 (intercept term).

[0126] Simplifying, we get the normal equation: ,

[0127]

[0128] k = 0, 1, 2, ..., n.

[0129] Further results were obtained: .

[0130] Using 40,000 training data points, the parameter values ​​of the first prediction model were obtained; that is, the model's β was adjusted through an optimization algorithm. 8,000 test data points were used to validate the first prediction model against new data. Through multiple iterations using the training data, the first prediction model was continuously improved, resulting in a fully trained first prediction model.

[0131] Figure 6 A flowchart illustrating a second prediction model training method according to an embodiment of the present disclosure is shown.

[0132] As mentioned above, Figure 6 As shown, the second prediction model is obtained through the following operations, including operations S610 to S630.

[0133] When operating S610, historical transaction characteristics are obtained. Historical transaction data includes historical transaction system development characteristics and the probability of actual transaction defects occurring when using a transaction system with transaction system development characteristics.

[0134] When operating S620, the historical transaction system development features are input into the second prediction model to obtain the second predicted probability of the occurrence of communication defects.

[0135] In operation S630, based on the difference between the probability of occurrence of the first predicted trading defect and the probability of occurrence of the second actual trading defect, the parameters of the first prediction model are adjusted until the difference is minimized, thus obtaining the second prediction model.

[0136] In one example, the historical transaction system development characteristics are obtained: the number of years (in days) that several developers were responsible for several transactions over several years, and the number of program defects that occurred during the production of those transactions. See Table 3:

[0137] Define the objective function: In the formula, x t The value represents the developer's years of service, k is the regression coefficient, and b is the intercept.

[0138] By inputting the collected data, including the number of years (in days) the developers were responsible for developing the transaction and the number of program defects that occurred in the transaction's history, the regression coefficient k and intercept b were obtained using the least squares method.

[0139] Taking the above example, the target prediction result is determined based on the prediction results of the multiple logistic regression model and the univariate linear regression model, as shown in Table 4:

[0140]

[0141] Understandably, the machine learning algorithms used can quickly predict large amounts of data, significantly improving the efficiency of predicting the extent of defects in online bank counter transactions. Furthermore, combining multiple logistic regression and univariate linear regression models allows for accurate prediction of the degree of defect leakage in online bank counter transactions, helping banks to identify and resolve problems promptly. This method can be applied to various types of transactions, and model parameters can be updated and improved at any time to enhance predictive performance.

[0142] Figure 7 A flowchart illustrating a transaction defect prediction method according to another embodiment of this disclosure is shown schematically.

[0143] As mentioned above, Figure 7 As shown, the transaction defect prediction method of this embodiment may further include operations S710 to S730.

[0144] In operation S710, it is determined that the probability of the occurrence of the transaction defect represented by the target prediction result is greater than the target threshold.

[0145] When operating S720, if the probability of a transaction defect occurring as indicated by the target prediction result is greater than the target threshold, an early warning is generated.

[0146] When operating S730, if the probability of a transaction defect occurring as indicated by the target prediction result is greater than the target threshold, the target prediction result is displayed.

[0147] For example, the target prediction result is 0.7, and the target threshold is 0.5. Since the target prediction result is greater than the target threshold, it indicates a higher probability of a transaction defect occurring, so an early warning is generated, requiring staff intervention. If the target prediction result is 0.3, since the target prediction result is less than the target threshold, it indicates a lower probability of a transaction defect occurring, so no early warning is generated, and only the target prediction result is displayed.

[0148] Understandably, the transaction defect prediction function can promptly detect and address defects, thereby improving the stability of online transactions at bank counters.

[0149] Based on the above-described method for predicting transaction defects, this disclosure also provides a device for predicting transaction defects. The following will be combined with... Figure 8 The device is described in detail.

[0150] Figure 8 A schematic block diagram of a transaction defect prediction apparatus according to an embodiment of the present disclosure is shown.

[0151] like Figure 8 As shown, the transaction defect prediction device 800 of this embodiment includes an acquisition module 810, a first prediction module 820, a second prediction module 830, and a determination module 840.

[0152] The acquisition module 810 is used to acquire target transaction features. The correlation between the target transaction features and the transaction defects is greater than a preset threshold. The target transaction features include transaction system test environment features and transaction system development features. In one embodiment, the acquisition module 810 can be used to perform the operation S210 described above, which will not be repeated here.

[0153] The first prediction module 820 is used to input the characteristics of the trading system test environment into the first prediction model to obtain a first prediction result. The first prediction result characterizes the probability of the occurrence of trading defects in the trading system test environment. In one embodiment, the first prediction module 820 can be used to perform the operation S220 described above, which will not be repeated here.

[0154] The second prediction module 830 is used to input the development features of the trading system into the second prediction model to obtain a second prediction result, which characterizes the probability of the occurrence of trading defects due to the development data of the trading system. In one embodiment, the second prediction module 830 can be used to perform the operation S230 described above, which will not be repeated here.

[0155] The determining module 840 is used to determine the target prediction result based on the first prediction result and the second prediction result. In one embodiment, the determining module 840 can be used to perform the operation S240 described above, which will not be repeated here.

[0156] According to embodiments of this disclosure, any plurality of modules among the acquisition module 810, the first prediction module 820, the second prediction module 830, and the determination module 840 may be combined into one module, or any one of these modules may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in one module. According to embodiments of this disclosure, at least one of the acquisition module 810, the first prediction module 820, the second prediction module 830, and the determination module 840 may be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in any one of the three implementation methods of software, hardware, and firmware, or in a suitable combination of any of these. Alternatively, at least one of the acquisition module 810, the first prediction module 820, the second prediction module 830, and the determination module 840 may be implemented at least partially as a computer program module, which can perform corresponding functions when the computer program module is run.

[0157] Figure 9 A block diagram of an electronic device suitable for implementing a transaction defect prediction method according to an embodiment of the present disclosure is shown schematically.

[0158] like Figure 9As shown, an electronic device 900 according to an embodiment of the present disclosure includes a processor 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage portion 908 into a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0159] RAM 903 stores various programs and data required for the operation of electronic device 900. Processor 901, ROM 902, and RAM 903 are interconnected via bus 904. Processor 901 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 902 and / or RAM 903. It should be noted that programs may also be stored in one or more memories other than ROM 902 and RAM 903. Processor 901 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in one or more memories.

[0160] According to embodiments of this disclosure, the electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to a bus 904. The electronic device 900 may also include one or more of the following components connected to the input / output (I / O) interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output (I / O) interface 905 as needed. A removable medium 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 910 as needed so that computer programs read from it can be installed into the storage section 908 as needed.

[0161] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0162] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include ROM 902 and / or RAM 903 and / or one or more memories other than ROM 902 and RAM 903 described above.

[0163] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement the transaction defect prediction method provided in the embodiments of this disclosure.

[0164] When the computer program is executed by the processor 901, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0165] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 909, and / or installed from a removable medium 911. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0166] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 909, and / or installed from the removable medium 911. When the computer program is executed by the processor 901, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0167] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can execute entirely on a user's computing device, partially on a user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0168] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0169] Those skilled in the art will understand that the features described in the various embodiments of this disclosure can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments of this disclosure can be combined and / or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0170] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A transaction defect prediction method characterized by, The method comprises: obtaining target transaction features, the target transaction features having a degree of association with transaction defects greater than a preset threshold, the target transaction features including transaction system test environment features and transaction system development features; inputting the transaction system test environment features into a first prediction model to obtain a first prediction result, the first prediction result representing the probability of transaction system test environment causing transaction defects; inputting the transaction system development features into a second prediction model to obtain a second prediction result, the second prediction result representing the probability of transaction system development data causing transaction defects; determining a target prediction result according to the first prediction result and the second prediction result.

2. The method of claim 1, wherein, inputting the transaction system test environment features into a first prediction model to obtain a first prediction result, comprising: obtaining first weight values, the first weight values representing the influence of each transaction system test environment feature on transaction defects; determining a first prediction result according to the plurality of transaction system test environment features and the first weight values.

3. The method of claim 1, wherein, inputting the transaction system development features into a second prediction model to obtain a second prediction result, comprising: determining a second prediction result according to the transaction system development features.

4. The method of claim 1, wherein, determining a target prediction result according to the first prediction result and the second prediction result, comprising: obtaining second weight values corresponding to the first prediction result and third weight values corresponding to the second prediction result; determining a target prediction result according to the first prediction result, the second weight values, the second prediction result, and the third weight values.

5. The method according to claim 1 or 2, characterized in that, The first prediction model is obtained by: obtaining historical transaction features, the historical transaction data including historical system test environment features and the probability of first actual transaction defects occurring when transactions are performed using a transaction system having the historical system test environment features; inputting the historical system test environment features into a first prediction model to obtain a first predicted transaction defect occurrence probability; adjusting the parameters of the first prediction model until the difference is minimized according to the difference between the first predicted transaction defect occurrence probability and the first actual transaction defect occurrence probability, thereby obtaining the first prediction model.

6. The method according to claim 1 or 3, characterized in that, The second prediction model is obtained by: obtaining historical transaction features, the historical transaction data including historical transaction system development features and the probability of actual transaction defects occurring when transactions are performed using a transaction system having the transaction system development features; inputting the historical transaction system development features into a second prediction model to obtain a second predicted transaction defect occurrence probability; adjusting the parameters of the first prediction model until the difference is minimized according to the difference between the first predicted transaction defect occurrence probability and the second actual transaction defect occurrence probability, thereby obtaining the second prediction model.

7. The method of claim 1, wherein, The method further comprises: generating a warning prompt if the probability of transaction defects represented by the target prediction result is greater than a target threshold.

8. The method of claim 1, wherein, The test environment characteristics include: the last delivery time of the test environment program of the trading system, the cumulative number of test environment deliveries of the trading system, the time since the last successful transaction in the test environment of the trading system, the cumulative number of successful transactions of the trading system, the cumulative number of successful calls to production transactions of the trading system, the number of defects that occurred during production operation of the trading system, the number of defects discovered during testing of the trading system, the number of branch scenarios involved in the transaction, the currency of the transaction, and the transaction type. The development characteristics of the trading system include: the number of years the trading system developers have worked on it.

9. A transaction defect prediction apparatus characterized by comprising: The device includes: The acquisition module is used to acquire target transaction features, wherein the correlation between the target transaction features and transaction defects is greater than a preset threshold, and the target transaction features include transaction system test environment features and transaction system development features; The first prediction module is used to input the characteristics of the trading system test environment into the first prediction model to obtain a first prediction result, wherein the first prediction result characterizes the probability of the occurrence of trading defects in the trading system test environment. The second prediction module is used to input the development features of the trading system into the second prediction model to obtain a second prediction result, wherein the second prediction result characterizes the probability of the occurrence of trading defects due to the development data of the trading system. The determination module is used to determine the target prediction result based on the first prediction result and the second prediction result.

10. An electronic device, comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

11. A computer readable storage medium having stored thereon a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 8.

12. A computer program product comprising computer programs or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 8.

Images