Transaction data processing method and device, equipment, storage medium and program product

Abstract

The embodiment of the invention provides a transaction data processing method and device, equipment, a storage medium and a program product, and is applied to the technical field of big data. The method comprises the steps of determining target transaction data of a current time period; the target transaction data comprises a transaction characteristic quantity and a time period characteristic quantity; the current time period is a sliding time window changing according to a preset time interval; determining a transaction volume prediction value corresponding to the current time period according to the time period characteristic quantity and a target regression model; determining a first deviation value corresponding to the current time period according to the transaction characteristic quantity and the transaction quantity predicted value; and when the first deviation value meets a first preset early warning condition, outputting first early warning information. The electronic equipment can realize real-time dynamic detection of the transaction volume anomaly in the transaction data, predicts the transaction volume predicted value based on the target regression model, does not need to manually set a detection threshold value, and can improve the accuracy of transaction volume anomaly detection in the transaction data.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to a transaction data processing method, device, equipment, storage medium and program product. Background Art

[0002] With the rapid development of computer and Internet technology, the transaction volume of financial systems is also growing rapidly. In order to ensure the stability and security of the transaction systems of banks and other financial institutions, the detection and processing of abnormal transaction volume has become an important function that the transaction system needs to have.

[0003] In related technologies, staff usually use manually set thresholds to detect abnormal transaction volumes. This transaction data detection and processing method relies too much on manual experience, has low accuracy, and the manually set thresholds are relatively simple, which cannot achieve real-time and dynamic detection of transaction data. Summary of the invention

[0004] The embodiments of the present application provide a transaction data processing method, apparatus, device, storage medium and program product, which can improve the accuracy of transaction volume anomaly detection and realize dynamic and real-time detection of transaction volume anomalies.

[0005] In a first aspect, an embodiment of the present application provides a transaction data processing method, including:

[0006] Determine target transaction data for the current time period; the target transaction data includes transaction feature quantities and time period feature quantities; the current time period is a sliding time window that changes according to a preset time interval;

[0007] Determine the transaction volume forecast value corresponding to the current time period according to the time period characteristic quantity and the target regression model;

[0008] Determine a first deviation value corresponding to a current time period according to the transaction characteristic quantity and the transaction volume prediction value;

[0009] When the first deviation value satisfies a first preset warning condition, a first warning message is output.

[0010] In a possible implementation manner, determining the target transaction data for the current time period includes:

[0011] Obtaining transaction log data of at least one transaction system in the current time period;

[0012] For each transaction system, the transaction log data is parsed and processed, the transaction feature quantity of the current time period is calculated, and the time period feature quantity corresponding to the current time period is determined.

[0013] In a possible implementation, the method further includes:

[0014] Acquiring historical transaction data, wherein the historical transaction data includes historical time period feature quantities and historical transaction feature quantities corresponding to a plurality of historical time periods;

[0015] Performing weighted processing on the historical transaction data to obtain sample data;

[0016] Randomly dividing the sample data according to a preset ratio to obtain training sample data and decision sample data corresponding to the sample data;

[0017] The preset regression model is iteratively trained according to the training sample data and the decision sample data to obtain the target regression model.

[0018] In a possible implementation, determining the transaction volume forecast value corresponding to the current time period according to the time period feature value and the target regression model includes:

[0019] If the current time period corresponds to a first time period and a second time period, determining a first proportion of the current time period in the first time period, and determining a second proportion of the current time period in the second time period;

[0020] Determine a first transaction volume prediction value corresponding to the first time period according to the time period feature quantity corresponding to the first time period and the target regression model;

[0021] Determine a second transaction volume prediction value corresponding to the second time period according to the time period feature value corresponding to the second time period and the target regression model;

[0022] The transaction volume prediction value corresponding to the current time period is determined according to the first proportion, the second proportion, the first transaction volume prediction value, and the second transaction volume prediction value.

[0023] In a possible implementation, the method further includes:

[0024] In each preset period, based on the target regression model, the transaction volume forecast value corresponding to each historical time period is updated.

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

[0026] Determine target total transaction data within a target period; the target total transaction data includes the total transaction feature amount corresponding to the target period and the target period feature amount corresponding to the target period;

[0027] Determine a predicted value of the total transaction volume corresponding to the target period according to the target period characteristic quantity and the target outlier detection model;

[0028] Determining a second deviation value between the transaction feature total amount and the transaction total amount prediction value;

[0029] When the second deviation value satisfies a second preset warning condition, a second warning message is output.

[0030] In a second aspect, an embodiment of the present application provides a transaction data processing device, including:

[0031] A first determination module is used to determine target transaction data for a current time period; the target transaction data includes transaction feature quantities and time period feature quantities; the current time period is a sliding time window that changes according to a preset time interval;

[0032] A second determination module is used to determine the transaction volume prediction value corresponding to the current time period according to the time period characteristic quantity and the target regression model;

[0033] A third determination module, configured to determine a first deviation value corresponding to a current time period according to the transaction characteristic quantity and the transaction volume prediction value;

[0034] The output module is used to output a first warning message when the first deviation value meets a first preset warning condition.

[0035] In a possible implementation manner, the first determining module is specifically configured to:

[0036] Obtaining transaction log data of at least one transaction system in the current time period;

[0037] For each transaction system, the transaction log data is parsed and processed, the transaction feature quantity of the current time period is calculated, and the time period feature quantity corresponding to the current time period is determined.

[0038] In a possible implementation manner, the device is further used for:

[0039] Acquire historical transaction data, wherein the historical transaction data includes historical time period feature quantities and historical transaction feature quantities corresponding to multiple historical time periods;

[0040] Performing weighted processing on the historical transaction data to obtain sample data;

[0041] Randomly dividing the sample data according to a preset ratio to obtain training sample data and decision sample data corresponding to the sample data;

[0042] The preset regression model is iteratively trained according to the training sample data and the decision sample data to obtain the target regression model.

[0043] In a possible implementation manner, the second determining module is specifically configured to:

[0044] If the current time period corresponds to a first time period and a second time period, determining a first proportion of the current time period in the first time period, and determining a second proportion of the current time period in the second time period;

[0045] Determine a first transaction volume prediction value corresponding to the first time period according to the time period feature quantity corresponding to the first time period and the target regression model;

[0046] Determine a second transaction volume prediction value corresponding to the second time period according to the time period feature value corresponding to the second time period and the target regression model;

[0047] The transaction volume prediction value corresponding to the current time period is determined according to the first proportion, the second proportion, the first transaction volume prediction value, and the second transaction volume prediction value.

[0048] In a possible implementation manner, the device is further used for:

[0049] In each preset period, based on the target regression model, the transaction volume forecast value corresponding to each historical time period is updated.

[0050] In a possible implementation manner, the device is further used for:

[0051] Determine target total transaction data within a target period; the target total transaction data includes the total transaction feature amount corresponding to the target period and the target period feature amount corresponding to the target period;

[0052] Determine a predicted value of the total transaction volume corresponding to the target period according to the target period characteristic quantity and the target outlier detection model;

[0053] Determining a second deviation value between the transaction feature total amount and the transaction total amount prediction value;

[0054] When the second deviation value satisfies a second preset warning condition, a second warning message is output.

[0055] In a third aspect, an embodiment of the present application provides a transaction data processing device, including: a memory, a processor;

[0056] The memory stores computer-executable instructions;

[0057] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the transaction data processing method as described in any one of the first aspects above.

[0058] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer execution instructions, and when the computer execution instructions are executed by a processor, they are used to implement the transaction data processing method as described in any one of the first aspects above.

[0059] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the transaction data processing method as described in any one of the first aspects above.

[0060] The transaction data processing method, device, equipment, storage medium and program product provided by the embodiment of the present application determine the target transaction data of the current time period; the target transaction data includes transaction feature quantity and time period feature quantity; the current time period is a sliding time window that changes according to a preset time interval; according to the time period feature quantity and the target regression model, the transaction volume prediction value corresponding to the current time period is determined; according to the transaction feature quantity and the transaction volume prediction value, the first deviation value corresponding to the current time period is determined; when the first deviation value meets the first preset warning condition, the first warning information is output. In the present application, the electronic device determines the transaction feature quantity and the time period feature quantity of the current time period, predicts the transaction volume prediction value corresponding to the current time period based on the time period feature quantity and the target regression model, and determines whether the first deviation value between the transaction feature quantity and the transaction volume prediction value meets the first preset warning condition, and outputs the first warning information when it meets. The current time period can slide according to the preset time interval, so that the electronic device can realize the real-time dynamic detection of the abnormal transaction volume in the transaction data, and predict the transaction volume prediction value based on the target regression model, without manually setting the detection threshold, which can improve the accuracy of the abnormal transaction volume detection in the transaction data. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0062] Figure 1 A flowchart of a transaction data processing method provided in this application;

[0063] Figure 2 A flowchart of another transaction data processing method provided by this application;

[0064] Figure 3 A schematic diagram of abnormal transaction volume detection in the current time period provided by this application;

[0065] Figure 4 A schematic diagram of a target period transaction volume anomaly detection provided for this application;

[0066] Figure 5 A module diagram of transaction data processing provided by this application;

[0067] Figure 6 A schematic diagram of the structure of a transaction data processing device provided by the present application;

[0068] Figure 7 A schematic diagram of the structure of a transaction data processing device provided for this application.

[0069] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0070] In order to enable those skilled in the art to better understand the technical solution of the present application, the present application is further described in detail below in conjunction with the drawings and embodiments. It should be understood that the specific embodiments and drawings described herein are only used to explain the present application, and are not intended to limit the present application.

[0071] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0072] In addition, this application involves big data analysis of user information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.), and the use of artificial intelligence technology for automated decision-making, and a technical solution for making decisions that have a significant impact on personal rights and interests based on the results of automated decision-making. The application provides users with corresponding operation entrances for them to choose to agree or reject the results of automated decision-making; if the user chooses to reject, the expert decision-making process will be entered.

[0073] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0074] It should be noted that the transaction data processing method, device, equipment, storage medium and program product of the present application can be used in the field of big data technology, and can also be used in any field outside the field of big data technology. The present application does not limit the specific application field of the transaction data processing method, device, equipment, storage medium and program product.

[0075] With the rapid development of computer and Internet technologies, the transaction volume of financial systems has increased dramatically, and the transaction volume of financial institutions such as banks has also increased rapidly. Since the availability and security of trading systems are of great importance to financial institutions, monitoring the normal operation of transactions, detecting abnormal transaction volumes, and discovering and eliminating problems as early as possible is an important function of financial trading systems. The rapid growth of transaction volumes has further increased the difficulty of detecting abnormal transaction volumes.

[0076] In the related art, the existing financial transaction system usually uses the method of manually setting thresholds when performing transaction volume anomaly detection on transaction data. This method of manually setting thresholds for transaction volume anomaly detection relies on the actual work experience of the staff, has low flexibility and accuracy, and cannot achieve real-time detection and processing of transaction volume anomalies.

[0077] In order to solve the above problems, the present application provides a transaction data processing method, device, equipment, storage medium and program product, wherein the electronic device determines the target transaction data of the current time period, the target transaction data includes a transaction feature quantity and a time period feature quantity, and the current time period is a sliding time window that changes according to a preset time interval; based on the time period feature quantity and the target regression model, the transaction volume prediction value corresponding to the current time period is predicted, and it is determined whether the first deviation value between the transaction feature quantity and the transaction volume prediction value meets the first preset warning condition, and outputs the first warning information when it meets the condition. The current time period can slide according to the preset time interval, so that the electronic device can realize real-time dynamic detection of transaction volume anomalies in transaction data, and at the same time predict the transaction volume prediction value based on the target regression model, without manually setting the detection threshold, which can improve the accuracy of transaction volume anomaly detection in transaction data.

[0078] The scheme shown in the present application is described in detail below through specific embodiments. It should be noted that the following embodiments can exist independently or in combination with each other, and the same or similar contents will not be described repeatedly in different embodiments.

[0079] Figure 1 A flowchart of a transaction data processing method provided for this application. Figure 1 , the transaction data processing method may include:

[0080] S101, determining target transaction data for the current time period; the target transaction data includes transaction feature quantities and time period feature quantities; the current time period is a sliding time window that changes according to a preset time interval.

[0081] The execution subject of the embodiment of the present application may be an electronic device, such as a mobile phone, a computer or a wearable device, etc., and may specifically refer to a financial institution transaction system in an electronic device, etc., or may be a transaction data processing device set in the electronic device. The transaction data processing device may be implemented by software, or by a combination of software and hardware. For ease of understanding, the following description is made by taking the execution subject as an electronic device as an example.

[0082] In an embodiment of the present application, the current time period may refer to the time period of the current moment, and the current time period may refer to a sliding time window that changes according to a preset time interval, wherein the preset time interval may refer to a preset time window sliding interval, and the sliding time window may refer to a time window of fixed length. The sliding time window slides at every preset time interval, and the electronic device may use the sliding time window after sliding as the current time period to obtain the target transaction data corresponding to the current time period. For example, the sliding time window may be 10 minutes, and the preset time interval is 10 seconds, that is, every 10 seconds, the sliding time window of 10 minutes in length is slid back 10 seconds, and the electronic device recalculates the target transaction data corresponding to the sliding time window of the current time period.

[0083] For example, if the current time is 10 o'clock, the electronic device can use the ten minutes from 9:50 to 10 o'clock as the current time period; then at 10:00:10, the sliding time window slides back 10 seconds, and the electronic device can use the ten minutes from 9:50:10 to 10:00:10 as the current time period. Of course, the preset time interval can be other time intervals, such as 20 seconds, 30 seconds, 60 seconds, etc., and the sliding time window can also be a time window of other lengths, such as 5 minutes, 20 minutes, 30 minutes, etc., which can be set based on actual needs, and the embodiments of the present application are not limited to this.

[0084] The target transaction data may refer to the transaction data of each transaction system in the current time period, which may include transaction feature quantities and time period feature quantities. The transaction feature quantity may refer to the feature quantity related to the transaction in the target transaction data, such as the transaction system identification, transaction type, number of transactions (or transaction quantity) and transaction amount, etc. The time period feature quantity may refer to the time characteristics of the current time period, such as the transaction system identification, the day of the week, the hour of the day, the ten minutes of the hour, whether it is a holiday, whether it is a working day, whether it is a system upgrade update day, etc. Of course, the transaction feature quantity and time period feature may also include other data, which is not limited in the embodiments of the present application.

[0085] In this step, for each transaction system, the electronic device can determine the target transaction data of the current time period, and the target transaction data includes transaction feature quantities and time period feature quantities. Subsequently, the abnormal transaction volume can be detected based on the target transaction data. In addition, the electronic device can obtain the target transaction data corresponding to the current time period at a preset time interval, which can realize dynamic and real-time detection of abnormal transaction volume.

[0086] S102: Determine the transaction volume forecast value corresponding to the current time period according to the time period characteristic quantity and the target regression model.

[0087] In the embodiment of the present application, the target regression model may refer to a pre-trained regression model, which can be used to predict the transaction volume prediction value corresponding to each time period. The target regression model may specifically refer to a random forest regression model, a decision tree regression model, or a gradient boosting regression model, etc., and of course, it may also be other regression models, which are not limited in the embodiment of the present application. The transaction volume prediction value may refer to the warning value of the transaction volume corresponding to the current time period, wherein the transaction volume may refer to the number of transactions and / or the transaction amount, etc.

[0088] In this step, the electronic device includes a pre-trained target regression model, the input of which is the time period feature of the current time period, and the output is the transaction volume prediction value corresponding to the current time period, such as the prediction value of the number of transactions and / or the transaction amount prediction value corresponding to the current time period. After determining the time period feature of the current time period, the electronic device can input the time period feature into the target regression model to obtain the transaction volume prediction value corresponding to the current time period. Subsequently, the electronic device can determine whether to issue an early warning based on the deviation between the transaction feature in the target transaction data and the transaction volume prediction value.

[0089] S103: Determine a first deviation value corresponding to the current time period according to the transaction feature quantity and the transaction volume prediction value.

[0090] In the embodiment of the present application, the first deviation value may refer to the absolute value of the difference between the transaction feature value and the transaction volume prediction value in the current time period. Specifically, after determining the transaction volume prediction value, the electronic device may calculate the difference between the transaction feature value and the transaction volume prediction value, and calculate the absolute value of the difference to obtain the first deviation value between the two.

[0091] S104: When the first deviation value satisfies a first preset warning condition, output a first warning message.

[0092] In the embodiment of the present application, the first preset warning condition may refer to a pre-set warning identification condition, which includes specific alarm indicator data, such as a first deviation value in a certain trading system is greater than or equal to a preset fluctuation range, etc. The first warning information may refer to abnormal trading volume prompt information, which may include forecast prompt text and time information, etc.

[0093] Specifically, after determining the first deviation value, the electronic device can further determine whether the first deviation value meets the first preset warning condition. If it does, the electronic device can output the first warning information. The specific output method of the first warning information can be to display it in a pop-up window on the display screen of the electronic device, or to send the first warning information to the mobile terminal of the relevant staff in the form of a notification message. It can also be combined with a voice output method to play the audio corresponding to the first warning information or output it by controlling the flashing of an external indicator light. The embodiment of the present application does not limit the specific output method of the first warning information. In this way, the electronic device can output the first warning information when the first deviation value between the transaction feature quantity and the transaction volume prediction value meets the first preset warning condition, thereby realizing a real-time warning of abnormal transaction volume.

[0094] The transaction data processing method provided by the embodiment of the present application determines the target transaction data of the current time period; the target transaction data includes a transaction feature quantity and a time period feature quantity; the current time period is a sliding time window that changes according to a preset time interval; according to the time period feature quantity and the target regression model, the transaction volume prediction value corresponding to the current time period is determined; according to the transaction feature quantity and the transaction volume prediction value, the first deviation value corresponding to the current time period is determined; when the first deviation value meets the first preset warning condition, the first warning information is output. In the present application, the electronic device determines the transaction feature quantity and the time period feature quantity of the current time period, predicts the transaction volume prediction value corresponding to the current time period based on the time period feature quantity and the target regression model, and determines whether the first deviation value between the transaction feature quantity and the transaction volume prediction value meets the first preset warning condition, and outputs the first warning information when it meets. The current time period can slide according to the preset time interval, so that the electronic device can realize the real-time dynamic detection of the abnormal transaction volume in the transaction data, and predict the transaction volume prediction value based on the target regression model, without manually setting the detection threshold, which can improve the accuracy of the abnormal transaction volume detection in the transaction data.

[0095] Based on the above embodiments, Figure 2 A flowchart of another transaction data processing method provided for this application. Figure 2 , the transaction data processing method may include:

[0096] S201. Obtain transaction log data of at least one transaction system in a current time period; for each transaction system, parse and process the transaction log data, calculate the transaction feature quantity of the current time period, and determine the time period feature quantity corresponding to the current time period.

[0097] In the embodiment of the present application, the transaction system may refer to a system used in a financial institution for online transactions. The transaction log data may refer to the original transaction data of the transaction system, such as a transaction log. Specifically, the electronic device may obtain the transaction log data of each transaction system in the current time period, extract the key fields therein and perform calculations to obtain the transaction feature quantity of the current time period, wherein the key fields may include the transaction system identifier, transaction time, transaction type, transaction account number, number of transactions, and transaction amount, etc. At the same time, the electronic device may determine the time period feature quantity corresponding to the current time period, such as determining the number of hours in a day of the current time period, whether the current time period is a working day, etc.

[0098] In an embodiment of the present application, the electronic device obtains the transaction log data of the current time period and performs analysis and processing to obtain the transaction feature quantity and time period feature quantity of the current time period, thereby ensuring the comprehensiveness and accuracy of the transaction data detection and processing. At the same time, the current time period can change with time, thereby realizing real-time and dynamic detection of transaction anomalies.

[0099] S202: Determine the transaction volume forecast value corresponding to the current time period according to the time period feature quantity and the target regression model.

[0100] In the embodiment of the present application, after determining the time period characteristic quantity of the current time period, the electronic device can input the time period characteristic quantity into the target regression model to obtain the transaction volume prediction value corresponding to the current time period.

[0101] In a possible implementation, the target regression model in the electronic device can be trained through the following steps (1) to (4):

[0102] (1) Obtain historical transaction data, where the historical transaction data includes historical time period feature quantities and historical transaction feature quantities corresponding to multiple historical time periods.

[0103] In the embodiment of the present application, historical transaction data may refer to transaction data of each historical time period of each transaction system within the historical time. The time length of each historical time period may be consistent with the length of the sliding time window, for example, each hour is divided into 6 historical time periods, and the time length of each historical time period is 10 minutes. The historical time may refer to the past 3 months, 6 months, or 12 months, etc. The historical time period feature quantity may refer to the time characteristics of each historical time period. The historical transaction feature quantity may refer to transaction-related features within the historical time period, such as the transaction system identifier, the number of historical transactions, and the amount of historical transactions.

[0104] Specifically, the electronic device may obtain historical transaction log data, and parse and process the historical transaction log data to obtain historical time period feature quantities and historical transaction feature quantities corresponding to each historical time period.

[0105] (2) Perform weighted processing on historical transaction data to obtain sample data.

[0106] In the embodiment of the present application, the sample data may refer to the data used to finally train the target regression model. Specifically, since the transaction data at different historical times may be different, the electronic device may determine the weight of the historical transaction data according to the time interval between the historical transaction data and the model training time, and then perform weighted processing on the historical transaction data based on the weight to obtain the sample data corresponding to the historical transaction data.

[0107] When the electronic device determines the weight of the historical transaction data, the time interval from the model training moment and the weight may be in inverse proportion, that is, the greater the time interval from the model training moment, the smaller the weight corresponding to the historical transaction data. For example, the electronic device may determine that at the time of model training, the weight of the historical transaction data in the past two weeks is set to A, the weight of the historical transaction data in the past two weeks to the past month is set to B, and the weight of the historical transaction data before the past month is set to C, where A is greater than B, and B is greater than C. Of course, the electronic device may also use other methods to set the weight of the historical transaction data, which is not limited in the embodiments of the present application.

[0108] (3) Randomly divide the sample data according to a preset ratio to obtain training sample data and decision sample data corresponding to the sample data.

[0109] In the embodiment of the present application, the preset ratio may refer to a preset sample division ratio, for example, 7:3 or 6:4, etc. The training sample data may be used for iterative training of the model, and the decision sample data may also be referred to as test sample data, etc., which may be used for validation and optimization of the model.

[0110] Specifically, after the electronic device performs weighted processing on the historical transaction data, it obtains sample data corresponding to the historical transaction data, and then randomly divides the sample data according to a preset ratio to obtain training sample data and decision sample data. For example, taking the target regression model as a random forest regression model, the electronic device can randomly screen the sample data to form multiple decision trees, and the sample data in each decision tree is randomly divided into training sample data and decision sample data according to a preset ratio (for example, 7:3, etc.). In the subsequent training process, the electronic device can use the average of the prediction results of multiple decision trees as the final prediction value.

[0111] (4) According to the training sample data and the decision sample data, the preset regression model is iteratively trained to obtain the target regression model.

[0112] In the embodiment of the present application, the preset regression model may refer to a regression model in an initial state. Specifically, after determining the training sample data and the decision sample data, the electronic device may iteratively train the preset regression model to obtain a target regression model. Taking the target regression model as a random forest regression model as an example, the random forest regression prediction model is composed of a plurality of decision trees composed of historical transaction data of each historical time period of each day in a large amount of historical transaction data. The mean square error (MSE) can be used to predict the result value and calculate the average value of multiple decision trees. Finally, the predicted value of historical transaction volume in the historical time period can be trained, and the iterative training can be continuously optimized based on the predicted value of historical transaction volume and the actual historical transaction feature quantity, and finally the target regression model can be obtained.

[0113] When training the preset regression model, the electronic device can minimize the residual sum of squares (RSS) of each decision tree during the training process. The RSS can be expressed by the following formula:

[0114]

[0115] Wherein, n is the number of sample data corresponding to the decision tree, which may specifically refer to the number of training sample data corresponding to the decision tree or the number of decision sample data corresponding to the decision tree. i Refers to the actual value of the i-th sample data. It may refer to the predicted value of the i-th sample data. Of course, when the target model is of other types, the electronic device may be trained in other ways, which is not limited in the embodiments of the present application.

[0116] In an embodiment of the present application, the electronic device obtains sample data by weighted processing of historical transaction data, and divides the sample data to obtain training sample data and decision sample data, and then iteratively trains the preset regression model based on the training sample data and the decision sample data, and finally obtains the target regression model, which can improve the accuracy and reliability of model training.

[0117] Since the historical time period of the target regression model during the training process is targeted at a specific time period, for example, one hour is divided into six time periods. However, in actual application, the current time period for the electronic device to collect target transaction data may span two time periods. For example, the historical time period may be 9:40:00 (hour: minute: second) -9:49:59, 9:50:00-9:59:59, and the current time period is 9:45:00-9:54:59. At this time, the electronic device only makes predictions for one of the historical time periods, which obviously does not correspond to the current time period. The electronic device can determine the transaction volume prediction values ​​of the two historical time periods, and calculate the transaction volume prediction value corresponding to the current time period based on the two transaction volume prediction values.

[0118] In a possible implementation, the transaction volume prediction value in step S202 may be calculated as follows:

[0119] (5) If the current time period corresponds to the first time period and the second time period, a first proportion of the current time period in the first time period is determined, and a second proportion of the current time period in the second time period is determined.

[0120] In the embodiment of the present application, the first time period and the second time period may refer to two time periods spanned by the current time period. The first proportion may refer to the proportion of the overlapping part of the current time period and the first time period in the first time period. The second proportion may refer to the proportion of the overlapping part of the current time period and the second time period in the second time period.

[0121] (6) Determine a first transaction volume prediction value corresponding to the first time period based on the time period characteristic quantity corresponding to the first time period and the target regression model; determine a second transaction volume prediction value corresponding to the second time period based on the time period characteristic quantity corresponding to the second time period and the target regression model.

[0122] (7) Determine a transaction volume forecast value corresponding to the current time period based on the first proportion, the second proportion, the first transaction volume forecast value, and the second transaction volume forecast value.

[0123] In the embodiment of the present application, the first transaction volume prediction value may refer to the transaction volume prediction value corresponding to the first time period, and the second transaction volume prediction value may refer to the transaction volume prediction value corresponding to the second time period. The electronic device determines the time period characteristic quantity of the first time period and the time period characteristic quantity of the second time period, and then respectively inputs them into the target regression model to obtain the first transaction volume prediction value corresponding to the first time period and the second transaction volume prediction value corresponding to the second time period. Afterwards, the electronic device can determine the transaction volume prediction value corresponding to the current time period by weighted summation based on the first proportion, the first transaction volume prediction value, the second proportion, and the second transaction volume prediction value.

[0124] In an embodiment of the present application, when the current time period spans the first time period and the second time period, the electronic device can determine a first transaction volume prediction value for the first time period and a second transaction volume prediction value for the second time period, and then calculate the transaction volume prediction value corresponding to the current time period by weighted summation, which can improve the accuracy and rationality of the transaction volume prediction value, and thus improve the accuracy of transaction volume anomaly detection.

[0125] In addition, the transaction volume forecast value of each historical time period can be dynamically updated, which can ensure the real-time and dynamic nature of transaction volume anomaly detection. In a possible implementation, the transaction data processing method further includes the following steps:

[0126] In each preset period, the transaction volume forecast value corresponding to each historical time period is updated based on the target regression model.

[0127] In the embodiment of the present application, the preset period may refer to a preset data update period. The preset period may be equal to the length of the sliding time window, or may be greater than the length of the sliding time window. For example, if the sliding time window is 10 minutes, the preset period may be 10 minutes, 15 minutes, 20 minutes, etc. Of course, the preset period may also be less than the length of the sliding time window, but in this case, if the target regression model is not optimized by parameter adjustment, the predicted transaction volume values ​​predicted within a sliding time window may be the same. The embodiment of the present application does not limit the specific duration of the preset period.

[0128] In an embodiment of the present application, the electronic device can update the transaction volume prediction value of each historical time period based on the target regression model at every preset period, so as to ensure the real-time and accuracy of the transaction volume anomaly detection. It should be noted that the target regression model in the electronic device can also be dynamically updated. For example, the electronic device can obtain updated historical transaction data, perform optimization training based on the updated historical transaction data, obtain an updated target regression model, and then dynamically update the transaction volume prediction value based on the updated target regression model, so as to further ensure the real-time and accuracy of the transaction volume prediction value.

[0129] S203: Determine a first deviation value corresponding to the current time period according to the transaction feature quantity and the transaction volume prediction value.

[0130] S204: When the first deviation value satisfies a first preset warning condition, output a first warning message.

[0131] In an embodiment of the present application, after the electronic device determines the transaction volume prediction value for the current time period, when the first deviation value between the transaction feature quantity and the transaction volume prediction value meets the first preset warning condition, the electronic device can output a first warning information to provide real-time reminders to relevant personnel, facilitating timely disposal.

[0132] For example, Figure 3 This is a schematic diagram of anomaly detection of transaction volume in the current time period provided by this application. Figure 3 As shown, the horizontal axis is time, i.e., each current time period on September 2; the vertical axis is the transaction feature quantity, which can specifically refer to the transaction volume (number of transactions or transaction amount, etc.). The broken line is the predicted transaction volume value predicted by the target regression model, and the black dots represent the actual transaction feature quantity of each current time period. Figure 3 It can be seen that between 17:00 and 18:00 on September 2, the first deviation value between the transaction feature quantity and the transaction volume prediction value was large and exceeded the fluctuation range. The electronic device can determine that the first deviation values ​​of these current time periods meet the first preset warning condition. The electronic device can output the first warning information to achieve real-time and dynamic detection of abnormal transaction volume.

[0133] S205. Determine the target total transaction data within the target period; the target total transaction data includes the total transaction feature amount corresponding to the target period and the target period feature amount corresponding to the target period.

[0134] In the embodiments of the present application, the target period may refer to the time period during which the total transaction volume anomaly detection is required, such as one day, three days, five days, etc. The target total transaction data may refer to the overall transaction data corresponding to the target period. The total transaction feature amount may refer to the sum of the transaction feature amounts corresponding to each time period within the target period. The target period feature amount may refer to the time feature amount corresponding to the target period. For example, when the target period is one day, the target period feature amount may include the transaction system identifier, the day of the week, whether it is a holiday, whether it is a working day, whether it is a system upgrade and update day, etc.

[0135] In this step, in addition to detecting transaction anomalies in the current time period, the electronic device can also perform anomaly detection on the target total transaction data within the target period. For example, at the end of each trading day or when the market is closed, or in a time period with less trading volume, the electronic device can perform transaction volume anomaly detection on the target total transaction data within the target period.

[0136] S206: Determine the predicted value of the total transaction volume corresponding to the target period according to the target period characteristic quantity and the target outlier detection model.

[0137] In the embodiments of the present application, the target outlier detection model may refer to a pre-trained regression prediction model, specifically a support vector machine regression model, or an outlier detection model based on proximity, density or clustering, etc. The embodiments of the present application do not limit the specific type of the target outlier detection model. The total transaction volume prediction value may refer to the prediction value of the total transaction volume corresponding to the target period.

[0138] Specifically, the electronic device may be based on the historical total transaction data of multiple historical periods, including the total historical transaction feature amount of each historical period and the historical period feature amount corresponding to the historical period. Afterwards, the electronic device may iteratively train the preset outlier detection model based on the historical total transaction data, and finally obtain the target outlier detection model. Then, the electronic device may input the target period feature amount corresponding to the target period into the target outlier detection model to obtain the predicted value of the total transaction amount corresponding to the target period.

[0139] Taking the target outlier detection model as a support vector machine regression model as an example, electronic equipment can use different kernel functions for prediction calculations based on historical total transaction data through cross-validation, and select the kernel function with the smallest error as the final target kernel function, thereby obtaining a target outlier detection model based on the target kernel function. The purpose of the kernel function is to find the optimal linear segmentation plane of the data in the high-dimensional space, map the nonlinear separable to a higher dimension, turn it into a linear separable, and better handle nonlinear transaction data.

[0140] It should be noted that the target outlier detection model can be continuously iteratively optimized as the historical total transaction data is continuously updated, which can improve the accuracy and rationality of outlier detection.

[0141] S207, determining a second deviation value between the total transaction feature amount and the predicted transaction total amount; when the second deviation value satisfies a second preset warning condition, outputting a second warning information.

[0142] In the embodiment of the present application, the second deviation value may refer to the absolute value of the difference between the target total transaction data of the target period and the predicted value of the total transaction volume. The second preset warning condition may refer to the pre-set judgment and identification condition of the abnormal transaction volume of the target period. The second warning information may be the prompt information corresponding to the abnormal transaction volume of the target period.

[0143] Specifically, after determining the total transaction volume forecast value corresponding to the target period, the electronic device can calculate the second deviation value between the target total transaction data and the total transaction volume forecast value, and output the second warning information when the second deviation value meets the second preset warning condition, so as to realize real-time reminder for relevant personnel.

[0144] In the embodiment of the present application, the electronic device obtains the target total transaction data corresponding to the target period, and then calculates the predicted value of the total transaction volume corresponding to the target period based on the target period characteristic quantity and the target outlier detection model, and then calculates the second deviation value between the total transaction characteristic quantity and the predicted value of the total transaction volume, and outputs the second warning information when the second deviation value meets the second preset warning condition. In this way, the electronic device can realize comprehensive and accurate detection and identification of abnormal transaction volume by performing abnormal transaction volume detection on the target period.

[0145] For example, Figure 4 A schematic diagram of a target period transaction volume anomaly detection provided in this application. Figure 4As shown, the horizontal axis is time, where each day within six months is a target period; the vertical axis is the total amount of transaction features, which can specifically refer to the total amount of transactions within the target period (total number of transactions or total transaction amount, etc.). The broken line is the predicted value of the total transaction volume predicted by the target outlier detection model, and the black dots represent the actual total amount of transaction features within each target period. Figure 4 It can be seen that the second deviation value between the total transaction feature amount and the predicted transaction amount in the target period on August 31 is large and exceeds the fluctuation range. The electronic device can determine that the second deviation value meets the second preset warning condition. The electronic device can output the second warning information and promptly notify relevant personnel to handle it, which can achieve accurate detection of abnormal transaction volume in the target period and ensure the stability and security of the trading system.

[0146] Based on the above embodiments, Figure 5 A module diagram of transaction data processing provided by this application. Figure 5 As shown, the electronic device may include a transaction data acquisition module, a real-time regression prediction module, and a target period outlier detection module. Among them, the transaction data acquisition module is used to collect and calculate the target transaction data of the current time period and the target total transaction data of the target period. The real-time regression prediction module is used to calculate the transaction volume prediction value corresponding to the current time period according to the target regression model, and determine whether the first preset warning condition is met according to the first deviation value between the transaction feature quantity in the target transaction data and the transaction volume prediction value, and output the first warning information when it is met, so as to realize dynamic and real-time detection of abnormal transaction volume in each current time period.

[0147] The target cycle outlier detection module is used to calculate the total transaction volume forecast value corresponding to the target cycle according to the target outlier detection model, and determine whether the second preset warning condition is met based on the second deviation value between the total transaction feature volume in the target total transaction data and the total transaction volume forecast value, and output the second warning information when the condition is met, so as to realize outlier detection for the target cycle and identify abnormal transaction volume within the target cycle.

[0148] In the embodiment of the present application, the electronic device can not only detect the abnormal transaction volume in real time and dynamically in each current time period, but also perform the overall transaction volume outlier detection within the target period, without the need to manually set the threshold for abnormal transaction volume detection, thereby improving the accuracy and real-time performance of abnormal transaction volume detection, and thus improving the stability and reliability of the trading system. Specifically, the transaction data processing method in the present application does not require manual setting of the transaction volume threshold, and can be predicted based on the training model of historical transaction data. In the related art, the manual setting of the transaction volume threshold requires industry experts to formulate specific thresholds, which usually requires a lot of professional knowledge, and the threshold needs to be updated from time to time, with high technical barriers and high human resource consumption. In the embodiment of the present application, the electronic device performs model training based on a large amount of historical transaction data, and can realize the prediction of transaction volume based on the target regression model and the target outlier detection model, and as the historical transaction data is updated, the model can also be automatically updated and iterated, with higher flexibility and accuracy, and also saves labor costs.

[0149] In addition, the electronic device in the present application performs outlier detection on the target period and determines that there is abnormal transaction volume data within the target period, which effectively improves the problem of handling huge transaction anomalies in the financial industry's payment and collection system, and the prediction of transaction volume is more reasonable and accurate. The target total transaction data of the target period can also be used as historical total transaction data for model optimization and analysis, which can further improve the accuracy of model prediction. This will also make it easier for relevant personnel to understand the dynamic trend of market transaction volume and assist relevant personnel in making product recommendation decisions.

[0150] Figure 6 This is a schematic diagram of the structure of a transaction data processing device provided by this application. Figure 6 , the transaction data processing device 60 may include:

[0151] The first determination module 61 is used to determine the target transaction data of the current time period; the target transaction data includes transaction feature quantities and time period feature quantities; the current time period is a sliding time window that changes according to a preset time interval;

[0152] The second determination module 62 is used to determine the transaction volume prediction value corresponding to the current time period according to the time period characteristic quantity and the target regression model;

[0153] A third determination module 63, configured to determine a first deviation value corresponding to the current time period according to the transaction characteristic quantity and the transaction volume prediction value;

[0154] The output module 64 is used to output a first warning message when the first deviation value meets a first preset warning condition.

[0155] In a possible implementation manner, the first determining module 61 is specifically configured to:

[0156] Obtaining transaction log data of at least one transaction system in the current time period;

[0157] For each transaction system, the transaction log data is parsed and processed, the transaction feature quantity of the current time period is calculated, and the time period feature quantity corresponding to the current time period is determined.

[0158] In a possible implementation manner, the device 60 is further used for:

[0159] Acquire historical transaction data, where the historical transaction data includes historical time period feature quantities and historical transaction feature quantities corresponding to multiple historical time periods;

[0160] Perform weighted processing on historical transaction data to obtain sample data;

[0161] The sample data is randomly divided according to a preset ratio to obtain training sample data and decision sample data corresponding to the sample data;

[0162] According to the training sample data and the decision sample data, the preset regression model is iteratively trained to obtain the target regression model.

[0163] In a possible implementation manner, the second determining module 62 is specifically configured to:

[0164] If the current time period corresponds to the first time period and the second time period, determining a first proportion of the current time period in the first time period, and determining a second proportion of the current time period in the second time period;

[0165] Determine a first transaction volume prediction value corresponding to the first time period according to a time period feature quantity corresponding to the first time period and a target regression model;

[0166] Determine a second transaction volume prediction value corresponding to the second time period according to the time period feature quantity corresponding to the second time period and the target regression model;

[0167] The transaction volume forecast value corresponding to the current time period is determined according to the first proportion, the second proportion, the first transaction volume forecast value and the second transaction volume forecast value.

[0168] In a possible implementation manner, the device 60 is further used for:

[0169] In each preset period, the transaction volume forecast value corresponding to each historical time period is updated based on the target regression model.

[0170] In a possible implementation manner, the device 60 is further used for:

[0171] Determine the target total transaction data within the target period; the target total transaction data includes the total transaction feature amount corresponding to the target period and the target period feature amount corresponding to the target period;

[0172] Determine the predicted value of the total transaction volume corresponding to the target period based on the target period characteristic quantity and the target outlier detection model;

[0173] Determine a second deviation value between the total amount of transaction features and the predicted value of the total amount of transactions;

[0174] When the second deviation value satisfies the second preset warning condition, a second warning message is output.

[0175] The transaction data processing device 60 provided in the embodiment of the present application can execute the technical solution shown in the above method embodiment, and its implementation principle and beneficial effects are similar, which will not be repeated here.

[0176] Figure 7 A schematic diagram of the structure of a transaction data processing device provided for this application. Figure 7 The transaction data processing device 70 may include a memory 71 and a processor 72. Exemplarily, the memory 71 and the processor 72 are interconnected via a bus 73.

[0177] The memory 71 is used to store program instructions;

[0178] The processor 72 is used to execute the program instructions stored in the memory to implement the transaction data processing method shown in the above embodiment.

[0179] Figure 7 The transaction data processing device 70 shown can execute the technical solution shown in the above method embodiment, and its implementation principle and beneficial effects are similar, which will not be repeated here.

[0180] An embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the above-mentioned transaction data processing method.

[0181] The embodiment of the present application may also provide a computer program product, including a computer program, which, when executed by a processor, can implement the above-mentioned transaction data processing method.

[0182] It should be noted that the processor mentioned in the embodiments of the present application may be a central processing unit (CPU), or 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 the processor may be any conventional processor, etc.

[0183] It should be understood that the memory mentioned in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct RAM bus random access memory (DR RAM). It should be noted that when the processor is a general purpose processor, DSP, ASIC, FPGA or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, the memory (storage module) is integrated in the processor. It should be noted that the memory described herein is intended to include but is not limited to these and any other suitable types of memory.

[0184] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0185] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processing unit of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0186] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0187] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0188] Regarding the various modules / units contained in the various devices and products described in the above embodiments, they may be software modules / units, or hardware modules / units, or they may be partially software modules / units and partially hardware modules / units. Each device and product may be applied to or integrated in a chip, a chip module or a terminal device. Exemplarily, for each device and product applied to or integrated in a chip, each module / chip contained therein may be implemented in the form of hardware such as circuits, or at least some modules / units may be implemented in the form of software programs, which run on a processor integrated inside the chip, and the remaining modules / units may be implemented in the form of hardware such as circuits.

[0189] In the present application, the term "include" and its variations may refer to non-restrictive inclusion; the term "or" and its variations may refer to "and / or". The terms "first", "second", etc. in the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. In the present application, "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the previously associated objects are in an "or" relationship.

[0190] The above are only some embodiments of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A transaction data processing method, characterized in that: include: Determine the target transaction data for the current time period; The target transaction data includes transaction feature quantities and time period feature quantities; the current time period is a sliding time window that changes according to a preset time interval; Determine the transaction volume forecast value corresponding to the current time period according to the time period characteristic quantity and the target regression model; Determine a first deviation value corresponding to a current time period according to the transaction characteristic quantity and the transaction volume prediction value; When the first deviation value satisfies a first preset warning condition, a first warning message is output.

2. The method according to claim 1, characterized in that The determining of the target transaction data for the current time period includes: Obtaining transaction log data of at least one transaction system in the current time period; For each transaction system, the transaction log data is parsed and processed, the transaction feature quantity of the current time period is calculated, and the time period feature quantity corresponding to the current time period is determined.

3. The method according to claim 1, characterized in that The method further comprises: Acquiring historical transaction data, wherein the historical transaction data includes historical time period feature quantities and historical transaction feature quantities corresponding to a plurality of historical time periods; Performing weighted processing on the historical transaction data to obtain sample data; Randomly dividing the sample data according to a preset ratio to obtain training sample data and decision sample data corresponding to the sample data; The preset regression model is iteratively trained according to the training sample data and the decision sample data to obtain the target regression model.

4. The method according to claim 1, characterized in that: Determining the transaction volume prediction value corresponding to the current time period according to the time period feature value and the target regression model includes: If the current time period corresponds to a first time period and a second time period, determining a first proportion of the current time period in the first time period, and determining a second proportion of the current time period in the second time period; Determine a first transaction volume prediction value corresponding to the first time period according to the time period feature quantity corresponding to the first time period and the target regression model; Determine a second transaction volume prediction value corresponding to the second time period according to the time period feature value corresponding to the second time period and the target regression model; The transaction volume prediction value corresponding to the current time period is determined according to the first proportion, the second proportion, the first transaction volume prediction value, and the second transaction volume prediction value.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: In each preset period, based on the target regression model, the transaction volume forecast value corresponding to each historical time period is updated.

6. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Determine target total transaction data within a target period; the target total transaction data includes the total transaction feature amount corresponding to the target period and the target period feature amount corresponding to the target period; Determine a predicted value of the total transaction volume corresponding to the target period according to the target period characteristic quantity and the target outlier detection model; Determining a second deviation value between the transaction feature total amount and the transaction total amount prediction value; When the second deviation value satisfies a second preset warning condition, a second warning message is output.

7. A transaction data processing device, characterized in that: include: A first determination module, used to determine target transaction data for a current time period; The target transaction data includes transaction feature quantities and time period feature quantities; the current time period is a sliding time window that changes according to a preset time interval; A second determination module is used to determine the transaction volume prediction value corresponding to the current time period according to the time period characteristic quantity and the target regression model; A third determination module, configured to determine a first deviation value corresponding to a current time period according to the transaction characteristic quantity and the transaction volume prediction value; The output module is used to output a first warning message when the first deviation value meets a first preset warning condition.

8. A transaction data processing device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor executes the transaction data processing method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the transaction data processing method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the transaction data processing method according to any one of claims 1 to 6 is implemented.

Images