Daytime batch processing method and device

By predicting the bank transaction volume trend and flexibly adjusting the startup time of day batch processing, the online transaction performance decline and accounting problems caused by fixed startup time in the existing technology are solved, and more efficient transaction processing and accurate accounting processing are achieved. .

CN113971495BActive Publication Date: 2024-11-19BANK OF CHINA
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202111286674.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-11-02
Publication Date
2024-11-19
Estimated Expiration
2041-11-02

AI Technical Summary

Technical Problem

The existing day batch processing methods are unable to cope with fluctuations in transaction volume due to fixed startup time, resulting in decreased online transaction performance and accounting problems.

Method used

By obtaining transaction data from the banking business system, using neural network models to predict the next day's transaction volume trend, flexibly adjust the start time of day batch processing, and determine the appropriate time window to start processing.

Benefits of technology

It improves customers' online transaction experience, ensures the accuracy of bank internal accounts and the efficiency of transaction processing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113971495B_ABST
    Figure CN113971495B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and device for daytime batch processing, which relates to the financial field, wherein the method comprises: obtaining the transaction data of the day in the banking system, wherein the transaction data of the day includes transaction feature data; inputting the transaction feature data into a transaction volume prediction model to obtain the trend of the next day's daytime transaction volume of the banking system, wherein the transaction volume prediction model is obtained by training a neural network model based on the historical transaction data of the banking system and the corresponding historical next day's daytime transaction volume; determining the time window for executing the daytime batch according to the trend of the next day's daytime transaction volume of the banking system; selecting the starting time of the time window as the start time of the daytime batch processing; and starting the daytime batch processing at the start time of the daytime batch processing. The present invention can flexibly determine the start time of the daytime batch processing, improve the customer's online transaction experience, and protect the accuracy of the bank's internal accounts.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the financial field, and in particular to a daytime batch processing method and device. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] In the financial industry, there is a need for daily batch processing, such as batch collection and payment, which can generate a large amount of transaction request data in advance and does not require high transaction response timeliness, or periodic deductions, interest payments and other internal accounting processing of the bank system. At present, the initiation time of daily batch task processing is fixed, which may cause the daily batch task to be started at the peak of online transactions on the same day, affecting the performance of online transactions and thus affecting the customer experience; or it may cause the daily batch transactions that need to be completed on the same day to not be completed in the daily trading window due to excessive transaction volume, thus causing accounting problems. Summary of the invention

[0004] The embodiment of the present invention provides a daytime batch processing method for flexibly determining the start time of the daytime batch processing, improving the online transaction experience of customers and protecting the accuracy of the internal accounts of the bank. The method includes:

[0005] Acquire the current day transaction data in the banking system, wherein the current day transaction data includes transaction feature data;

[0006] Input the transaction feature data into the transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training the neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume;

[0007] Determine the time window for executing the intraday batch according to the intraday trading volume trend of the next day in the banking system;

[0008] Select the start time of the time window as the start time of the daytime batch processing;

[0009] At the start time of the day batch processing, the day batch processing is started.

[0010] The embodiment of the present invention also provides a daytime batch processing device, which is used to flexibly determine the start time of daytime batch processing, improve the customer's online transaction experience and protect the accuracy of the bank's internal accounts. The device includes:

[0011] A data acquisition module is used to acquire the daily transaction data in the banking system, wherein the daily transaction data includes transaction feature data;

[0012] The transaction volume prediction module is used to input the transaction feature data into the transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training the neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume;

[0013] A time window determination module is used to determine the time window for executing the daily batch according to the trend of the daily transaction volume of the next day of the banking business system;

[0014] The start time setting module is used to select the start time of the time window as the start time of the daytime batch processing;

[0015] The daytime batch processing module is used to start the daytime batch processing at the start time of the daytime batch processing.

[0016] An embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned daytime batch processing method when executing the computer program.

[0017] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for executing the above-mentioned daytime batch processing method.

[0018] In an embodiment of the present invention, the transaction data of the day in the banking system is obtained, wherein the transaction data of the day includes transaction feature data; the transaction feature data is input into a transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training a neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume; according to the next day's intraday transaction volume trend of the banking system, the time window for executing the intraday batch is determined; the starting time of the time window is selected as the start time of the intraday batch processing; and the intraday batch processing is started at the start time of the intraday batch processing. Compared with the existing intraday batch timing start method, the present invention can predict the next day's intraday transaction volume trend according to the current day's transaction data in the banking system through the transaction volume prediction model, so that the start time of the intraday batch processing can be flexibly adjusted according to the next day's intraday transaction volume trend, thereby ensuring the customer's online transaction experience and the accuracy of the bank's internal accounts. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0020] Figure 1 A process flow of a daytime batch processing method provided in an embodiment of the present invention Figure 1 ;

[0021] Figure 2 A flow chart of a method for training a transaction volume prediction model in daily batch processing provided in an embodiment of the present invention;

[0022] Figure 3 A process flow of a daytime batch processing method provided in an embodiment of the present invention Figure 2 ;

[0023] Figure 4 A process flow of a daytime batch processing method provided in an embodiment of the present invention Figure 3 ;

[0024] Figure 5 A process flow of a daytime batch processing method provided in an embodiment of the present invention Figure 4 ;

[0025] Figure 6 A flow chart of a method for optimizing a transaction volume prediction model in daily batch processing provided in an embodiment of the present invention;

[0026] Figure 7 A schematic diagram of a daytime batch processing device provided in an embodiment of the present invention Figure 1 ;

[0027] Figure 8 A schematic diagram of a daytime batch processing device provided in an embodiment of the present invention Figure 2 ;

[0028] Fig. 9 A schematic diagram of a daytime batch processing device provided in an embodiment of the present invention Figure 3 ;

[0029] Fig.10 A schematic diagram of a daytime batch processing device provided in an embodiment of the present invention Figure 4 . DETAILED DESCRIPTION

[0030] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0031] In the description of this specification, the terms "include", "including", "have", "contain", etc. are all open terms, which mean including but not limited to. The descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps is not limited and can be appropriately adjusted as needed.

[0032] Terminology explanation:

[0033] Daytime batch processing: The bank's daytime batch processing refers to the bank system simulating the front-end initiated transactions in batches in an online manner during the day. It can be mainly divided into two categories. One is batch business processing such as batch collection and payment, batch opening of customers, etc., which can generate a large amount of transaction request data in advance and do not require high transaction response timeliness; the other is internal batch account processing such as regular deductions and interest payments.

[0034] Research has found that the existing daytime batch processing methods are all scheduled to start, which cannot cope with unexpected situations during the batch processing process. With the advent of the era of big data and artificial intelligence, deep learning is increasingly used in all aspects of people's lives. Therefore, it is considered to apply deep learning technology to the daytime batch processing of the banking system to achieve flexible adjustment of the start time of daytime batch processing.

[0035] Regarding the above studies, Figure 1 As shown, an embodiment of the present invention provides a daytime batch processing method, which includes the following steps:

[0036] Step 101, obtaining the current day transaction data in the banking system, wherein the current day transaction data includes transaction feature data;

[0037] Step 102, inputting the transaction feature data into a transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training a neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume;

[0038] Step 103, determining a time window for executing a daily batch according to the daily transaction volume trend of the banking system for the next day;

[0039] Step 104, selecting the start time of the time window as the start time of the daytime batch processing;

[0040] Step 105, at the start time of the daytime batch processing, start the daytime batch processing.

[0041] In an embodiment of the present invention, the transaction data of the day in the banking system is obtained, wherein the transaction data of the day includes transaction feature data; the transaction feature data is input into a transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training a neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume; according to the next day's intraday transaction volume trend of the banking system, the time window for executing the intraday batch is determined; the starting time of the time window is selected as the start time of the intraday batch processing; and the intraday batch processing is started at the start time of the intraday batch processing. Compared with the existing intraday batch timing start method, the present invention can predict the next day's intraday transaction volume trend based on the current day's transaction data in the banking system through the transaction volume prediction model, so that the start time of the intraday batch processing can be flexibly adjusted according to the next day's intraday transaction volume trend, thereby ensuring the customer's online transaction experience and the accuracy of internal accounts.

[0042] In the above step 101, the online transaction information in the banking system can be monitored by the monitoring module, and the transaction data of the day in the banking system can be obtained after the daily batch processing task is completed.

[0043] During the specific implementation, it is necessary to perform desensitization, cleansing and other processing on the day's transaction data to make it meet data security requirements; then, extract the transaction feature data from the processed day's transaction data, and standardize the transaction feature data.

[0044] The transaction characteristic data may include the transaction response time, the transaction volume corresponding to the transaction response time, and whether the transaction response time is within a preset special time period. For example, the preset special time period may be a Double 11 flash sale, during which the transaction volume increases significantly. Therefore, whether the transaction response time is within the preset special time period may be used as the transaction characteristic data.

[0045] In the above step 102, the transaction feature data extracted in the above step 101 is input into the transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system.

[0046] Prior to this, the historical transaction data of the banking system and the corresponding historical next-day intraday trading volume can be used to train the neural network model to obtain a transaction volume prediction model.

[0047] In one embodiment of the present invention, Figure 2 As shown, a flow chart of a method for training a transaction volume prediction model in daily batch processing is provided, and the method comprises the following steps:

[0048] Step 201, obtaining historical transaction data in a banking system and corresponding historical next-day intraday transaction volume as sample data, and constructing a training set and a test set, wherein the historical transaction data includes historical transaction feature data.

[0049] In specific implementation, it is necessary to desensitize and clean the historical transaction data in the banking business system, and use the processed historical transaction data and the corresponding historical next day's daily transaction volume as sample data to meet data security requirements; then, extract the historical transaction feature data in the sample data, and standardize the historical transaction feature data, and use the standardized historical transaction feature data and the corresponding historical next day's daily transaction volume to construct a training set and a test set in a preset ratio. For example, the standardized historical transaction feature data can be divided into a training set and a test set in a ratio of 3:1.

[0050] Step 202: Use the training set to train the neural network model to obtain a transaction volume prediction model.

[0051] The neural network model may be a long short-term memory network (LSTM). By training the LSTM using a training set, a recurrent neural network model for predicting the next day's trading volume trend, namely, a trading volume prediction model, may be obtained.

[0052] Step 203: Test the transaction volume prediction model using the test set.

[0053] In this way, the historical transaction data in the banking system and the corresponding historical next-day intraday trading volume are used to train the neural network model, making the final transaction volume prediction model more accurate.

[0054] In the above step 103, the time window for executing the daily batch can be determined according to the intraday trading volume trend of the next day predicted by the trading volume prediction model.

[0055] In one embodiment of the present invention, Figure 3 As shown, the above step 103 may specifically include:

[0056] Step 301, determine the execution time required for daily batch processing according to the trend of daily trading volume on the next day.

[0057] During specific implementation, the next day's intraday trading volume can be obtained based on the next day's intraday trading volume trend, and then the execution time required to process the next day's intraday trading volume can be determined based on the next day's intraday trading volume.

[0058] Among them, since the system performance of the banking business system has a great impact on the efficiency of batch processing, in order to ensure that the accurate execution time of the daily batch processing can be obtained, the above step 301 may specifically include: determining the execution time required for the daily batch processing based on the daily transaction volume trend of the next day and the system performance of the banking business system.

[0059] For example, when the system performance of the banking business system is good, the execution time of the daytime batch processing will be shortened, and when the system performance of the banking business system is poor, the execution time of the daytime batch processing will be extended.

[0060] Step 302, taking the execution time required for the daily batch processing as the size of the sliding window, and determining the time window for executing the daily batch according to the sliding window algorithm and the trend of the daily trading volume of the next day.

[0061] In specific implementation, according to the sliding window algorithm, a time window in which the intraday batch can be executed can be selected by sliding in the trend chart of the intraday trading volume trend of the next day.

[0062] The determined time window for executing the daytime batch may be one or more, and there may be no time window that meets the requirements for executing the daytime batch.

[0063] In the above step 104, according to the time window for executing the daytime batch determined in the above step 103, the start time of the time window is selected as the start time of the daytime batch processing.

[0064] In specific implementation, when the determined time window for executing the daytime batch is one, the above step 104 includes: selecting the starting time of the time window as the starting time of the daytime batch processing.

[0065] In specific implementation, when there are multiple time windows for executing daytime batches, such as Figure 4 As shown, the above step 104 includes: step 401, taking the starting time of the median-th time window among the multiple time windows as the starting time of the daytime batch processing.

[0066] In specific implementation, if the time window for executing the daily batch is not determined, the daily batch task can be divided into individual sessions for processing, such as Figure 5As shown, it also includes: Step 501, dividing the next day's day trading volume into multiple sessions for daytime batch processing according to the next day's day trading volume trend.

[0067] Then, for the next day's intraday trading volume of each session, the above steps 301 and 302 are executed. That is, according to the next day's intraday trading volume of each session, the execution time required for the intraday batch processing of each session is determined; the execution time required for the intraday batch processing of each session is used as the size of the sliding window, and the time window for executing the intraday batch of each session is determined according to the sliding window algorithm and the trend of the next day's intraday trading volume of each session. By dividing the next day's intraday trading volume into multiple sessions for intraday batch processing, the size of the time window can be shortened, and then a time window for executing the intraday batch can be selected.

[0068] In the above step 105, the daytime batch processing is started at the start time of the daytime batch processing.

[0069] In one embodiment of the present invention, after the daily batch processing is completed, the latest daily transaction feature data and the corresponding historical next day's daily transaction volume can be regularly obtained to optimize the transaction volume prediction model. Figure 6 As shown, the following steps may be specifically included:

[0070] Step 601, obtaining daily transaction feature data within a preset time period and the corresponding historical next day's daily transaction volume.

[0071] During specific implementation, the preset time period can be set according to the specific implementation scenario, and can be the latest week, the latest 15 days, or the latest month, etc., and is not limited here.

[0072] Step 602, updating sample data with daily transaction feature data within a preset time period and the corresponding historical next day intraday transaction volume;

[0073] Step 603, using the updated sample data, update the training set and the test set, and continue to train and test the transaction volume prediction model.

[0074] In this way, regular optimization and training of the trading volume prediction model can ensure the real-time and accuracy of the trading volume prediction model.

[0075] The present invention also provides a daytime batch processing device, as described in the following embodiments. Since the principle of the device to solve the problem is similar to that of the daytime batch processing method, the implementation of the device can refer to the implementation of the daytime batch processing method, and the repeated parts will not be repeated.

[0076] like Figure 7FIG. 1 is a schematic diagram of a daytime batch processing device provided by an embodiment of the present invention, wherein the device comprises:

[0077] The data acquisition module 701 is used to acquire the daily transaction data in the banking system, wherein the daily transaction data includes transaction feature data;

[0078] The transaction volume prediction module 702 is used to input the transaction feature data into the transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training a neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume;

[0079] The time window determination module 703 is used to determine the time window for executing the daily batch according to the trend of the daily transaction volume of the banking system on the next day;

[0080] A start time setting module 704 is used to select the start time of the time window as the start time of the daytime batch processing;

[0081] The daytime batch processing module 705 is used to start the daytime batch processing at the start time of the daytime batch processing.

[0082] In one embodiment of the present invention, the transaction characteristic data includes a transaction response time, a transaction volume corresponding to the transaction response time, and whether the transaction response time is within a preset special time period.

[0083] In one embodiment of the present invention, Figure 8 As shown, the device further includes: a model training and testing module 801, which is used before the transaction volume prediction module inputs the transaction feature data into the transaction volume prediction model:

[0084] Obtain historical transaction data in a banking business system and the corresponding historical next-day intraday trading volume as sample data to construct a training set and a test set, wherein the historical transaction data includes historical feature data;

[0085] Use the training set to train the neural network model and obtain the transaction volume prediction model;

[0086] The trading volume prediction model is tested using the test set.

[0087] In one embodiment of the present invention, Fig. 9 As shown, the device also includes: a data updating and model optimization module 901, which is specifically used to:

[0088] Obtain daily transaction feature data within a preset time period and the corresponding historical next day's intraday trading volume;

[0089] Update sample data with daily transaction feature data within a preset time period and the corresponding historical next day intraday trading volume;

[0090] Using the updated sample data, the training set and test set are updated, and the trading volume prediction model is continued to be trained and tested.

[0091] In one embodiment of the present invention, the time window determination module is specifically used to:

[0092] Determine the execution time required for daily batch processing based on the next day's intraday trading volume trend;

[0093] The execution time required for daily batch processing is used as the size of the sliding window, and the time window for executing daily batches is determined based on the sliding window algorithm and the trend of the next day's daily trading volume.

[0094] In one embodiment of the present invention, the time window determination module is further configured to:

[0095] Determine the time window for executing the daily batch as multiple time windows;

[0096] The startup time setting module is specifically used for:

[0097] The start time of the median time window among the multiple time windows is used as the start time of the daytime batch processing.

[0098] In one embodiment of the present invention, the time window determination module is further specifically used to: determine the time window for executing the daytime batch;

[0099] like Fig.10 As shown, the device also includes:

[0100] The trading volume division module 1001 is used to divide the next day's daily trading volume into multiple sessions for daily batch processing according to the next day's daily trading volume trend.

[0101] In one embodiment of the present invention, the time window determination module is further configured to:

[0102] The execution time required for daily batch processing is determined based on the next day's intraday trading volume trend and the system performance of the banking system.

[0103] In an embodiment of the present invention, the transaction data of the day in the banking system is obtained, wherein the transaction data of the day includes transaction feature data; the transaction feature data is input into a transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training a neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume; according to the next day's intraday transaction volume trend of the banking system, the time window for executing the intraday batch is determined; the starting time of the time window is selected as the start time of the intraday batch processing; and the intraday batch processing is started at the start time of the intraday batch processing. Compared with the existing intraday batch timing start method, the present invention can predict the next day's intraday transaction volume trend based on the current day's transaction data in the banking system through the transaction volume prediction model, so that the start time of the intraday batch processing can be flexibly adjusted according to the next day's intraday transaction volume trend, thereby ensuring the customer's online transaction experience and the accuracy of internal accounts.

[0104] An embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned daytime batch processing method when executing the computer program.

[0105] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for executing the above-mentioned daytime batch processing method.

[0106] In an embodiment of the present invention, the transaction data of the day in the banking system is obtained, wherein the transaction data of the day includes transaction feature data; the transaction feature data is input into a transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training a neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume; according to the next day's intraday transaction volume trend of the banking system, the time window for executing the intraday batch is determined; the starting time of the time window is selected as the start time of the intraday batch processing; and the intraday batch processing is started at the start time of the intraday batch processing. Compared with the existing intraday batch timing start method, the present invention can predict the next day's intraday transaction volume trend based on the current day's transaction data in the banking system through the transaction volume prediction model, so that the start time of the intraday batch processing can be flexibly adjusted according to the next day's intraday transaction volume trend, thereby ensuring the customer's online transaction experience and the accuracy of internal accounts.

[0107] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0109] 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.

[0110] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0111] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A daytime batch processing method, It is characterized in that include: Acquire the transaction data of the day in the banking system, wherein the transaction data of the day includes transaction characteristic data, and the transaction characteristic data includes transaction response time, transaction volume corresponding to the transaction response time, and whether the transaction response time is within a preset special time period; Input the transaction feature data into the transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training the neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume; Determine the time window for executing the intraday batch according to the intraday trading volume trend of the next day in the banking system; Select the start time of the time window as the start time of the daytime batch processing; At the start time of the day batch processing, start the day batch processing; Among them, according to the trend of the next day's intraday trading volume of the banking system, the time window for executing the intraday batch is determined, including: Determine the execution time required for daily batch processing based on the next day's intraday trading volume trend; The execution time required for daily batch processing is used as the size of the sliding window. The time window for executing daily batches is determined based on the sliding window algorithm and the trend of the next day's daily trading volume. Determine the time window for executing intraday batches, including: Determine the time window for executing the daily batch as multiple time windows; Select the start time of the time window as the start time of the daytime batch processing, including: The start time of the median time window among the multiple time windows is used as the start time of the daytime batch processing.

2. The method according to claim 1, It is characterized in that Before inputting transaction feature data into the transaction volume prediction model, it also includes: Obtain historical transaction data in a banking business system and the corresponding historical next-day intraday transaction volume as sample data, and construct a training set and a test set, wherein the historical transaction data includes historical transaction feature data; Use the training set to train the neural network model and obtain the transaction volume prediction model; The trading volume prediction model is tested using the test set.

3. The method according to claim 2, It is characterized in that Also includes: Obtain daily transaction feature data within a preset time period and the corresponding historical next day's intraday trading volume; Update sample data with daily transaction feature data within a preset time period and the corresponding historical next day intraday trading volume; Using the updated sample data, the training set and test set are updated, and the trading volume prediction model is continued to be trained and tested.

4. The method according to claim 1, It is characterized in that Determine the execution time required for daily batch processing, including: The execution time required for daily batch processing is determined based on the next day's intraday trading volume trend and the system performance of the banking system.

5. A daytime batch processing device, It is characterized in that include: A data acquisition module, used to acquire the transaction data of the day in the banking system, wherein the transaction data of the day includes transaction characteristic data, and the transaction characteristic data includes transaction response time, transaction volume corresponding to the transaction response time, and whether the transaction response time is within a preset special time period; The transaction volume prediction module is used to input the transaction feature data into the transaction volume prediction model to obtain the next day's intraday transaction volume trend of the banking system, wherein the transaction volume prediction model is obtained by training the neural network model based on the historical transaction data of the banking system and the corresponding historical next day's intraday transaction volume; A time window determination module is used to determine the time window for executing the daily batch according to the trend of the daily transaction volume of the next day of the banking business system; The start time setting module is used to select the start time of the time window as the start time of the daytime batch processing; The daytime batch processing module is used to start the daytime batch processing at the start time of the daytime batch processing; The time window determination module is specifically used for: Determine the execution time required for daily batch processing based on the next day's intraday trading volume trend; The execution time required for daily batch processing is used as the size of the sliding window. The time window for executing daily batches is determined based on the sliding window algorithm and the trend of the next day's daily trading volume. The time window determination module is also used for: Determine the time window for executing the daily batch as multiple time windows; The startup time setting module is specifically used for: The start time of the median time window among the multiple time windows is used as the start time of the daytime batch processing.

6. The device as claimed in claim 5, It is characterized in that Also includes: The model training and testing module is used before the transaction volume prediction module inputs the transaction feature data into the transaction volume prediction model: Obtain historical transaction data in a banking business system and the corresponding historical next-day intraday trading volume as sample data to construct a training set and a test set, wherein the historical transaction data includes historical feature data; Use the training set to train the neural network model and obtain the transaction volume prediction model; The trading volume prediction model is tested using the test set.

7. The device according to claim 6, It is characterized in that Also includes: Data update and model optimization module, specifically used for: Obtain daily transaction feature data within a preset time period and the corresponding historical next day's intraday trading volume; Update sample data with daily transaction feature data within a preset time period and the corresponding historical next day intraday trading volume; Using the updated sample data, the training set and test set are updated, and the trading volume prediction model is continued to be trained and tested.

8. The device according to claim 5, It is characterized in that The time window determination module is also used for: The execution time required for daily batch processing is determined based on the next day's intraday trading volume trend and the system performance of the banking system.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.

10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Citation Information

Patent Citations

  • Method and apparatus for predicting foreign exchange transaction volume

    CN109359758A

  • Automatic modeling method and device and electronic equipment

    CN111008707A

  • Batch operation simulation operation model establishing method and device

    CN111461867A

  • Bank transaction volume prediction method, resource allocation method, computing device and medium

    CN112200653A

  • Data processing method and device and server

    CN113159464A