Data processing method and device, computer equipment and storage medium

Abstract

The invention relates to a data processing method and device, computer equipment, a storage medium and a computer program product, and relates to the technical field of big data. The method comprises the following steps: acquiring a historical data set corresponding to a target item of a user; the target project comprises a deduction project and an association project associated with the deduction project; preprocessing the historical data set to obtain a first target historical data set and a second target historical data set; determining a first dimension prediction result based on the time sequence model and the first target historical data set, and determining a second dimension prediction result according to the cyclic network model and the second target historical data set; according to the first dimension prediction result and the second dimension prediction result, constructing a prediction result of the reduction project; and initiating a deduction program of the next time period of the deduction project based on the prediction result. By adopting the method, the efficiency of the data processing method can be improved.

Description

Technical Field

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

[0002] At present, banks will collect fees from the payment items of various enterprises to obtain intermediary business income. For some special types of exemption items that meet the requirements for fee reduction, banks will use data processing methods to reduce the fees of each exemption item.

[0003] The current data processing method is to manually determine the user's exemption items and the corresponding exemption ratios for the next time period based on the user's historical charges and exemption records. Then, the staff manually initiates the exemption procedure for the exemption items, and then completes the fee exemption for the exemption items.

[0004] However, in the current data processing method, when there are many exemption items, it takes a lot of time to manually determine the exemption items and the exemption ratios of the exemption items, which results in low efficiency of the current data processing method. Summary of the invention

[0005] Based on this, it is necessary to provide a data processing method, apparatus, computer device, computer-readable storage medium and computer program product to address the above technical issues.

[0006] In a first aspect, the present application provides a data processing method, comprising:

[0007] Acquire a historical data set corresponding to a user's target project; the target project includes a reduction or exemption project and associated projects associated with the reduction or exemption project;

[0008] Preprocessing the historical data set to obtain a first target historical data set and a second target historical data set;

[0009] Determine a first dimension prediction result based on a time series model and the first target historical data set, and determine a second dimension prediction result based on a recurrent network model and the second target historical data set;

[0010] Constructing the prediction result of the exemption project according to the prediction result of the first dimension and the prediction result of the second dimension;

[0011] Based on the prediction results, the exemption procedure for the next time period of the exemption project is initiated.

[0012] In one embodiment, obtaining a historical data set corresponding to a user's target project includes:

[0013] Determining the exemption item corresponding to the user identifier according to the user identifier;

[0014] Determining the associated project associated with the exemption project according to the target type to which the exemption project belongs;

[0015] The project history data set corresponding to the exemption project and the associated history data set corresponding to the associated project are obtained, and the history data set corresponding to the target project is constructed according to the project history data set and the associated history data set.

[0016] In one embodiment, the historical data set includes a project historical data set and an associated historical data set, and the preprocessing of the historical data set to obtain a first target historical data set and a second target historical data set includes:

[0017] Backing up the historical exemption trend data in the project historical data set, and grouping the backed-up project historical data set to obtain a first project historical data set and a second project historical data set;

[0018] Backing up the associated historical exemption trend data in the associated historical data set, and performing grouping processing on the backed up associated historical data set to obtain a first associated historical data set and a second associated historical data set;

[0019] Performing weighted processing on the first project historical data set and the first associated historical data set to obtain a first target historical data set;

[0020] A second target historical data set is constructed according to the second project historical data set and the second associated historical data set.

[0021] In one embodiment, determining the prediction result of the first dimension based on the time series model and the first target historical data set, and determining the prediction result of the second dimension based on the recurrent network model and the second target historical data set, includes:

[0022] Performing forecasting processing on the first target historical data set according to the time series model to obtain a forecast result identifier and a first forecast trend, and constructing a first dimension forecast result according to the forecast result identifier and the first forecast trend;

[0023] The second target historical data set is predicted and processed according to the recurrent network model to obtain a predicted exemption ratio and a second predicted trend, and a second dimension prediction result is constructed according to the predicted exemption ratio and the second predicted trend.

[0024] In one embodiment, constructing the prediction result of the exemption project according to the prediction result of the first dimension and the prediction result of the second dimension includes:

[0025] Determine whether the prediction result identifier in the prediction result of the first dimension is a target prediction result identifier;

[0026] If the prediction result identifier is a target prediction result identifier, determining whether a first prediction trend in the prediction result of the first dimension is consistent with a second prediction trend in the prediction result of the second dimension;

[0027] If the first prediction trend is consistent with the second prediction trend, the prediction result of the exemption project is constructed according to the predicted exemption ratio in the first dimension prediction result and the second dimension prediction result.

[0028] In one of the embodiments, after determining whether the prediction result identifier in the prediction result of the first dimension is a target prediction result identifier, the method further includes:

[0029] If the prediction result identifier is not the target prediction result identifier, determining that resource reduction will not be performed in the next time period of the reduction item;

[0030] Determining that resource reduction will not be performed in the next time period of the reduction and exemption item is a prediction result of the reduction and exemption item.

[0031] In one embodiment, after determining whether the first prediction trend in the first dimension prediction result is consistent with the second prediction trend in the second dimension prediction result, the method further includes:

[0032] If the first prediction trend is inconsistent with the second prediction trend, execute the steps of determining the first dimension prediction result based on the time series model and the first target historical data set, and determining the second dimension prediction result based on the recurrent network model and the second target historical data set until the first prediction trend is consistent with the second prediction trend.

[0033] In a second aspect, the present application further provides a data processing device, comprising:

[0034] An acquisition module, used to acquire a historical data set corresponding to a user's target project; the target project includes a reduction or exemption project and an associated project associated with the reduction or exemption project;

[0035] A processing module, used for preprocessing the historical data set to obtain a first target historical data set and a second target historical data set;

[0036] A determination module, configured to determine a first dimension prediction result based on a time series model and the first target historical data set, and to determine a second dimension prediction result based on a recurrent network model and the second target historical data set;

[0037] A construction module, used to construct the prediction result of the exemption project according to the prediction result of the first dimension and the prediction result of the second dimension;

[0038] A sending module is used to initiate the exemption procedure for the next time period of the exemption project based on the prediction result.

[0039] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0040] Acquire a historical data set corresponding to a user's target project; the target project includes a reduction or exemption project and associated projects associated with the reduction or exemption project;

[0041] Preprocessing the historical data set to obtain a first target historical data set and a second target historical data set;

[0042] Determine a first dimension prediction result based on a time series model and the first target historical data set, and determine a second dimension prediction result based on a recurrent network model and the second target historical data set;

[0043] Constructing the prediction result of the exemption project according to the prediction result of the first dimension and the prediction result of the second dimension;

[0044] Based on the prediction results, the exemption procedure for the next time period of the exemption project is initiated.

[0045] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0046] Acquire a historical data set corresponding to a user's target project; the target project includes a reduction or exemption project and associated projects associated with the reduction or exemption project;

[0047] Preprocessing the historical data set to obtain a first target historical data set and a second target historical data set;

[0048] Determine a first dimension prediction result based on a time series model and the first target historical data set, and determine a second dimension prediction result based on a recurrent network model and the second target historical data set;

[0049] Constructing the prediction result of the exemption project according to the prediction result of the first dimension and the prediction result of the second dimension;

[0050] Based on the prediction results, the exemption procedure for the next time period of the exemption project is initiated.

[0051] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0052] Acquire a historical data set corresponding to a user's target project; the target project includes a reduction or exemption project and associated projects associated with the reduction or exemption project;

[0053] Preprocessing the historical data set to obtain a first target historical data set and a second target historical data set;

[0054] Determine a first dimension prediction result based on a time series model and the first target historical data set, and determine a second dimension prediction result based on a recurrent network model and the second target historical data set;

[0055] Constructing the prediction result of the exemption project according to the prediction result of the first dimension and the prediction result of the second dimension;

[0056] Based on the prediction results, the exemption procedure for the next time period of the exemption project is initiated.

[0057] The above data processing method, device, computer equipment, storage medium and computer program product obtain the historical data set corresponding to the user's target project; the target project includes the exemption project and the associated project associated with the exemption project; the historical data set is preprocessed to obtain the first target historical data set and the second target historical data set; the first dimension prediction result is determined based on the time series model and the first target historical data set, and the second dimension prediction result is determined based on the cyclic network model and the second target historical data set; the prediction result of the exemption project is constructed based on the first dimension prediction result and the second dimension prediction result; the exemption program for the next time period of the exemption project is initiated based on the prediction result. By adopting this method, the first target historical data set is processed by the time series model, and the second target historical data set is processed by the cyclic network model, so that the first dimension prediction result and the second dimension prediction result of different dimensions of the exemption project can be quickly obtained. Then, the prediction result of the exemption project is determined according to the first dimension prediction result and the second dimension prediction result of different dimensions, so as to realize the automatic prediction of the exemption situation of the next cycle of the exemption project, save data processing time, and improve the efficiency of the data processing method. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 is a flow chart of a data processing method in one embodiment;

[0060] Figure 2 A schematic diagram of a process for obtaining a historical data set in one embodiment;

[0061] Figure 3 A schematic diagram of a process flow of steps for processing a historical data set in one embodiment;

[0062] Figure 4 A schematic diagram of a flow chart of the steps of determining a first dimension prediction result and a second dimension prediction result in one embodiment;

[0063] Figure 5 A schematic diagram of a process for constructing a prediction result step in one embodiment;

[0064] Figure 6 A schematic diagram of a process for determining a prediction result in one embodiment;

[0065] Figure 7 A schematic diagram of the framework of a handling fee prediction and reduction system in one embodiment;

[0066] Figure 8 A unit structure diagram of a handling fee reduction prediction module in one embodiment;

[0067] Fig. 9 A flowchart of a fee reduction prediction module in one embodiment;

[0068] Fig.10 It is a unit structure diagram of a handling fee reduction linkage processing module in one embodiment;

[0069] Fig.11 It is a flowchart of a fee reduction linkage processing module in one embodiment;

[0070] Fig.12 is a structural block diagram of a data processing device in one embodiment;

[0071] Fig.13 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0073] In one embodiment, Figure 1As shown, a data processing method is provided. The embodiment of the present application takes the method applied to a computer device as an example for explanation. The embodiment of the present application does not limit the execution device of the data processing method, and includes the following steps 102 to 110:

[0074] Step 102: Obtain a historical data set corresponding to the user's target project.

[0075] The target project includes the exemption project and the associated projects associated with the exemption project. The exemption project is the project corresponding to the user that needs to be exempted from resources. The number of exemption projects is at least one.

[0076] In implementation, the computer device determines the exemption project corresponding to the user ID in the database according to the user ID of the user. Then, the computer device determines the associated project that belongs to the same project type as the exemption project. The computer device obtains the project history data set corresponding to the exemption project and the associated history data set corresponding to the associated project, and constructs the history data set according to the project history data set and the associated history data set.

[0077] Step 104 , preprocessing the historical data set to obtain a first target historical data set and a second target historical data set.

[0078] The historical data set includes a project historical data set corresponding to the exemption project and an associated historical data set corresponding to the associated project.

[0079] In implementation, the computer device backs up the historical exemption trend data in the project historical data set and the associated historical exemption trend data in the associated historical data set, respectively, to obtain the backed-up project historical data set and the backed-up associated historical data set. Then, the computer device performs grouping and weighting processing on the backed-up project historical data set and the backed-up associated historical data set to obtain the first target historical data set and the second target historical data set.

[0080] Step 106, determining a first dimension prediction result based on the time series model and the first target historical data set, and determining a second dimension prediction result based on the recurrent network model and the second target historical data set.

[0081] In implementation, the computer device performs prediction processing on the first target historical data set through the time series model to obtain a first dimension prediction result. Then, the computer device performs prediction processing on the second target historical data through the recurrent network model to obtain a second dimension prediction result.

[0082] Step 108, constructing the prediction results of the exemption items according to the prediction results of the first dimension and the prediction results of the second dimension.

[0083] The first dimension prediction result includes a prediction result identifier and a first prediction trend, and the second dimension prediction result includes a second prediction trend and a predicted reduction ratio.

[0084] In implementation, the computer device determines whether the prediction result identifier is the target prediction result identifier. If the prediction result identifier is the target prediction result identifier, the computer device determines whether the first prediction trend is consistent with the second prediction trend. If the first prediction trend is consistent with the second prediction trend, the computer device constructs the prediction result of the exemption project according to the first prediction trend, the prediction result identifier and the predicted exemption ratio.

[0085] Step 110, initiating the exemption procedure for the next time period of the exemption project based on the prediction result.

[0086] In implementation, the computer device determines the value of the resource to be exempted according to the predicted exemption ratio in the prediction result and the resource value of the exemption item. Then, the computer device initiates the exemption procedure for the next time period of the exemption item according to the value of the resource to be exempted.

[0087] In an optional embodiment, if the prediction result is that the exemption project will no longer perform resource exemption in the next time period, the computer device does not initiate the exemption procedure for the next time period of the exemption project.

[0088] In the above data processing method, the first target historical data set is processed by the time series model, and the second target historical data set is processed by the recurrent network model, so that the first dimension prediction results and the second dimension prediction results of different dimensions of the exemption items can be quickly obtained. Then, the prediction results of the exemption items are determined according to the first dimension prediction results and the second dimension prediction results of different dimensions, so as to realize the automatic prediction of the exemption situation of the exemption items in the next cycle, save data processing time, and improve the efficiency of the data processing method.

[0089] In an exemplary embodiment, Figure 2 As shown, the specific processing process of step 102 includes steps 202 to 206. Among them:

[0090] Step 202: Determine the exemption item corresponding to the user ID according to the user ID.

[0091] In implementation, the computer device obtains the user ID of the user, and then determines the exemption item corresponding to the user ID in the database according to the user ID.

[0092] For example, in the banking field, the computer device obtains the user ID of the user, and then determines the payment items corresponding to the user ID in the database according to the user ID, and determines the exemption items that need to be exempted from the handling fee among the payment items.

[0093] Optionally, the user identifier may be, but is not limited to, a user number, name, etc. The embodiment of the present application does not limit the user identifier.

[0094] Step 204, determining the associated projects associated with the exemption project according to the target type to which the exemption project belongs.

[0095] In implementation, the computer device determines the target type of the exemption project according to the information of the exemption project, and then determines the associated projects of the same target type as the exemption project in the database.

[0096] In an exemplary embodiment, if there are multiple exemption items, the computer device determines the target type corresponding to the exemption item in the database for each exemption item. For example, if the exemption item is a transfer of tuition and miscellaneous fees, the computer device determines the target type of the exemption item to be tuition and miscellaneous fees. Then, the computer device determines the associated items of the associated number of the target type in the database according to the preset associated number. For example, if the associated number is 3, the computer device determines 3 associated items of the target type in the database.

[0097] Optionally, the number of associations is determined based on the number of projects of the target type of the exemption project and data processing requirements. The embodiment of the present application does not limit the number of associations.

[0098] Step 206, obtaining the project historical data set corresponding to the exemption project and the associated historical data set corresponding to the associated project, and constructing the historical data set corresponding to the target project based on the project historical data set and the associated historical data set.

[0099] In implementation, the computer device obtains the project history data set corresponding to the exemption project from the database according to the project identifier of the exemption project. At the same time, the computer device obtains the associated history data set corresponding to the associated project from the database according to the project identifier of the associated project. Then, the computer device combines the project history data set and the associated history data set into a history data set corresponding to the target project.

[0100] Specifically, the historical data set of the exemption project includes historical project exemption identification data, historical project exemption trend data and historical project exemption ratio data for each historical time period of the exemption project. The associated historical data set includes associated historical exemption identification data, associated historical exemption trend data and associated historical exemption ratio data for each historical time period of the associated project.

[0101] In an exemplary embodiment, a data acquisition unit is provided in a computer device. The computer device obtains a historical data set of the target project through the data acquisition unit. Specifically, the main function of the data acquisition unit is to obtain the target project information corresponding to the user from the payment project database, and obtain the resource value and resource exemption record corresponding to the target project from the resource processing system to obtain a historical data set. Then, the data acquisition unit sends the collected historical data set to the subsequent model unit. The historical data set predicts the exemption situation in the next time period. For example, if the target project is a bank's payment project, the data contained in a payment project include: a timestamp, which represents the time when the transaction occurred. The payment item type is used to identify the type of payment item, such as party dues, fiscal payments, tuition and miscellaneous fees, etc. Charging data: including the payment item handling fee charging amount, charging cycle, handling fee exemption amount, exemption ratio customer information, etc.

[0102] In this embodiment, by acquiring the project historical data set corresponding to the exemption project and the associated historical data set corresponding to the associated projects, the historical data set corresponding to the target project is obtained, that is, the historical data set used to predict the exemption results of the exemption project is obtained, which facilitates subsequent data processing.

[0103] In an exemplary embodiment, the historical data set includes a project historical data set and an associated historical data set, such as Figure 3 As shown, the specific processing process of step 104 includes steps 302 to 308. Among them:

[0104] Step 302 , backing up the historical exemption trend data in the project historical data set, and performing grouping processing on the backed up project historical data set to obtain a first project historical data set and a second project historical data set.

[0105] The project history data set includes historical project exemption identification data, historical project exemption trend data and historical project exemption ratio data. The historical project exemption identification data indicates whether the exemption project has carried out resource exemption in the historical time period.

[0106] In implementation, the computer device backs up the historical exemption trend data in the project historical data set to obtain the backed-up project historical data set. The computer device constructs the historical project exemption identification data and the historical project exemption trend data into a first project historical data set. Then, the computer device constructs the backed-up historical project exemption trend data and the historical project exemption ratio data into a second project historical data set.

[0107] Step 304 , backing up the associated historical exemption trend data in the associated historical data set, and performing grouping processing on the backed up associated historical data set to obtain a first associated historical data set and a second associated historical data set.

[0108] The associated historical data set includes associated historical exemption identification data, associated historical exemption trend data, and associated historical exemption ratio data. The associated historical exemption identification data indicates whether the associated project has been exempted from resources within a historical time period.

[0109] In implementation, the computer device performs a backup process on the associated historical exemption trend data in the associated historical data set to obtain a backed-up associated historical data set. The computer device constructs the associated historical exemption identification data and the associated historical exemption trend data into a first associated historical data set. Then, the computer device constructs the backed-up associated historical exemption trend data and the associated historical exemption ratio data into a second associated historical data set.

[0110] Step 306: Perform weighted processing on the first project historical data set and the first associated historical data set to obtain a first target historical data set.

[0111] Among them, the sum of the weights of exemption items and related items is 1.

[0112] In implementation, the weights of the exemption items and the associated items are pre-set in the computer device. Then, for each item historical data in each item historical data set, the computer device determines the associated historical data corresponding to the item historical data in the first associated historical data set. According to the weights of the exemption items and the associated items, the item historical data and the associated historical data are weighted to obtain the target historical data. The first target historical data set is constructed according to each target historical data.

[0113] Specifically, the computer device is pre-set with weights of exemption items and associated items. The project history data in the first project history data set includes historical project exemption identification data and historical project exemption trend data. The associated historical data in the first associated historical data set includes associated historical exemption identification data and associated historical exemption trend data. The computer device determines the associated historical exemption identification data corresponding to the historical project exemption identification data. Then, the computer device performs weighted processing on the historical project exemption identification data and the associated historical exemption identification data according to the weight of the exemption project and the weight of the associated project to obtain the target historical exemption identification data. The computer device determines the associated historical exemption trend data corresponding to the historical project exemption trend data. Then, the computer device performs weighted processing on the historical project exemption trend data and the associated historical exemption trend data according to the weight of the exemption project and the weight of the associated project to obtain the target historical exemption trend data.

[0114] For example, the exemption project corresponds to two associated projects, namely the first associated project and the second associated project. The historical project exemption identification data is 1, indicating that the exemption project has carried out resource exemptions during the historical time period. The first associated historical exemption identification data is 0, indicating that the associated project has not carried out resource exemptions during the historical time period. The first associated historical exemption identification data is 1. The weight of the exemption project is 0.5. The weights of the first associated project and the second project are 0.25 respectively. The computer device performs weighted processing on the historical project exemption identification data and the associated historical exemption identification data according to the weight of the exemption project and the weight of the associated project, and obtains 0.75. Since 0.75 is greater than 0.5 (the pre-set exemption identification threshold), the computer device determines that the target historical exemption ratio data is 1.

[0115] Step 308: construct a second target historical data set according to the second project historical data set and the second associated historical data set.

[0116] In implementation, the computer device combines the second project historical data set and the second associated historical data set into a second target historical data set.

[0117] In this embodiment, by grouping and weighting the historical data sets, a first target historical data set representing the resource exemption situation of the first dimension of the exemption project and a second target historical data set representing the resource exemption situation of the second dimension of the exemption project are obtained, which facilitates the subsequent determination of the resource exemption situation of the exemption project in the next time period based on the first target historical data set and the second target historical data set.

[0118] In an exemplary embodiment, Figure 4 As shown, the specific processing process of step 106 includes steps 402 to 404. Among them:

[0119] Step 402, predictively process the first target historical data set according to the time series model to obtain a prediction result identifier and a first prediction trend, and construct a first dimension prediction result according to the prediction result identifier and the first prediction trend.

[0120] The time series model is an ARIMA model (Autoregressive Integrated Moving Average Model). The first target historical data set contains target historical exemption identification data and target historical exemption trend data. The first dimension represents the exemption trend of the exemption project in the next time period.

[0121] In implementation, the computer device inputs the target historical exemption identification data into the ARIMA model, and performs prediction processing on the target historical exemption identification data through the ARIMA model to obtain a prediction result identification. Then, the computer device inputs the target historical exemption trend data into the ARIMA model, and performs prediction processing on the target historical exemption trend data through the ARIMA model to obtain a first prediction trend. The computer device constructs a first dimension prediction result based on the prediction result identification and the first prediction trend.

[0122] In an exemplary embodiment, the computer device performs differential processing on the first target historical data set to obtain a processed first target historical data set. The ARIMA model generally requires stable time series data. The differential operation can be implemented by calculating the difference between the current time period and the previous time period. Then, the computer device determines the parameters of the ARIMA model based on the autocorrelation function (ACF) and partial autocorrelation function (PACF) graphs. The computer device selects appropriate AR (Autoregressive model), I (difference order) and MA (Moving average) values ​​based on the information in the graph. The computer device fits the ARIMA model based on the selected parameter values. Among them, the formula of the ARIMA model is shown in the following formula (1):

[0123] (1)

[0124] In the above formula (1), is the observed value of the time series, c is a constant, is the autoregressive coefficient, is the sliding mean coefficient, is the error term.

[0125] After completing the preparation of the above-mentioned ARIMA model, the computer device inputs the processed first target historical data set into the ARIMA model, performs prediction processing on the processed first target historical data set through the ARIMA model, and obtains a prediction result identifier and a first prediction trend.

[0126] Step 404 , predictively process the second target historical data set according to the recurrent network model to obtain a predicted exemption ratio and a second predicted trend, and construct a second dimension prediction result according to the predicted exemption ratio and the second predicted trend.

[0127] The recurrent network model is an LSTM (Long Short-Term Memory) model. The second dimension represents the tax relief details for the next time period of the tax relief project.

[0128] In implementation, the computer device inputs the second target historical data set into the LSTM model, performs prediction processing on the second target historical data set through the LSTM model, and obtains the predicted exemption ratio and the second predicted trend. Then, the computer device constructs the second dimension prediction result according to the predicted exemption ratio and the second predicted trend.

[0129] In an exemplary embodiment, before executing step 404, it is necessary to build and train the LSTM network. The computer device obtains an initial historical sample data set. The initial historical sample data set includes sample data of historical project exemption ratios, sample data of historical project exemption trends, historical project types, and timestamps. Then, the computer device converts the initial historical sample data set into a sequence sample data set suitable for LSTM. Among them, the sequence sample data set usually needs to define the length of the input sequence and the time step of the output sequence. Then, the computer device builds an LSTM network, that is, the computer device builds a deep learning model, including an LSTM layer, a fully connected layer, and an appropriate activation function. The LSTM layer may include multiple neurons to capture long-term dependencies in the time series. Then, the computer device trains the LSTM model based on the sequence sample data set to obtain a trained LSTM model. Among them, the LSTM model learns patterns and trends in time series data. The computer device uses loss functions such as Mean-Square Error (MSE) for training, and uses optimization algorithms such as Stochastic Gradient Descent (SGD) to update the parameters of the LSTM model. The computer equipment evaluates the performance of the model based on the validation set data, and tunes the model hyperparameters according to the validation results, such as adjusting the number of layers, number of neurons, learning rate, etc.

[0130] In this embodiment, the first target historical data set is predicted and processed by the time series model to obtain the first dimension prediction result, thereby quickly clarifying the tax reduction trend of the next time period of the tax reduction project. The second target historical data set is predicted and processed by the recurrent network model to obtain the second dimension prediction result, thereby quickly clarifying the tax reduction details of the next time period of the tax reduction project.

[0131] In an exemplary embodiment, Figure 5 As shown, the specific processing process of step 108 includes steps 502 to 506. Among them:

[0132] Step 502, determining whether the prediction result identifier in the first dimension prediction result is a target prediction result identifier.

[0133] The first dimension prediction result includes a prediction result identifier and a first prediction trend. The target prediction result identifier indicates that the resource reduction will be carried out for the reduction and exemption project in the next time period.

[0134] In implementation, the computer device determines whether the prediction result identifier is the target prediction result identifier. If the prediction result identifier is the target prediction result identifier, the computer device determines that the resource reduction will be performed in the next time period of the reduction and exemption project.

[0135] In an exemplary embodiment, 1 indicates that the exemption project will be exempted from resources in the next time period, that is, 1 is the target prediction result identifier, and 0 indicates that the exemption project will not be exempted from resources in the next time period. The computer device determines whether the prediction result identifier in the first dimension prediction result is 1. If the prediction result identifier is 1, the computer device determines that the exemption project will be exempted from resources in the next time period.

[0136] Step 504: If the prediction result identifier is the target prediction result identifier, determine whether the first prediction trend in the first dimension prediction result and the second prediction trend in the second dimension prediction result are consistent.

[0137] Among them, the second dimension prediction results include the second prediction trend and the predicted exemption ratio.

[0138] In implementation, if the prediction result identifier is a target prediction result identifier, the computer device determines whether the first prediction trend is consistent with the second prediction trend.

[0139] For example, if the first predicted trend is rising and the second predicted trend is falling, the computer device determines that the first predicted trend and the second predicted trend are inconsistent. If the first predicted trend is rising and the second predicted trend is rising, the computer device determines that the first predicted trend and the second predicted trend are consistent.

[0140] Step 506: If the first prediction trend is consistent with the second prediction trend, construct the prediction results of the exemption items according to the predicted exemption ratios in the first dimension prediction results and the second dimension prediction results.

[0141] During implementation, if the first prediction trend and the second prediction trend are consistent, the computer device constructs the prediction result of the exemption project based on the first prediction trend, the prediction result identifier and the predicted exemption ratio.

[0142] In this embodiment, when the prediction result identifier is the target prediction result identifier and the first prediction trend and the second prediction trend are consistent, the prediction result of the exemption project is determined based on the first prediction trend and the predicted exemption ratio, and the resource exemption situation of the exemption project in the next time period is automatically determined, thereby improving data processing efficiency.

[0143] In an exemplary embodiment, when the prediction result identifier is not the target prediction result identifier, it is also necessary to construct the prediction result of the exemption project. Figure 6 As shown, after step 502 is executed, the data processing method further includes steps 602 to 604. Among them:

[0144] Step 602: If the prediction result identifier is not the target prediction result identifier, it is determined that resource reduction will not be performed in the next time period of the reduction item.

[0145] In implementation, if the prediction result identifier is not the target prediction result identifier, the computer device determines that the exemption item will not perform resource exemption in the next time period.

[0146] Step 604: determining that resource reduction will not be performed in the next time period of the reduction or exemption project is a prediction result of the reduction or exemption project.

[0147] In implementation, the computer device determines that the exemption project will not perform resource exemption in the next time period as a prediction result of the exemption project.

[0148] In this embodiment, when the prediction result identifier is not the target prediction result identifier, the exemption item will not perform resource exemption in the next time period as the prediction result of the exemption item, so that the exemption procedure of the exemption item will not be initiated subsequently, saving communication resources.

[0149] In an exemplary embodiment, when the first prediction trend is different from the second prediction trend, it is necessary to redetermine the prediction trend. After step 504 is executed, the data processing method further includes:

[0150] If the first prediction trend and the second prediction trend are inconsistent, the steps of determining the first dimension prediction result based on the time series model and the first target historical data set, and determining the second dimension prediction result based on the recurrent network model and the second target historical data set are performed until the first prediction trend and the second prediction trend are consistent.

[0151] In implementation, if the first predicted trend and the second predicted trend are inconsistent, the computer device executes the above step 106 until the first predicted trend and the second predicted trend are consistent. The specific processing process of step 106 has been described in the above embodiment, and the embodiment of the present application will not be repeated here.

[0152] In an exemplary embodiment, a prediction number threshold is preset in the computer device. After executing step 106, the computer device updates the current prediction number. Then, the computer device determines whether the current prediction number reaches the prediction number threshold. If the current prediction number reaches the prediction number threshold, the computer device stops executing the data processing method and returns information that the current prediction number reaches the prediction number threshold.

[0153] In this embodiment, when the first prediction trend and the second prediction trend are inconsistent, the prediction result of the first dimension and the prediction result of the second dimension are re-determined, thereby verifying the prediction result and improving the accuracy of the prediction result.

[0154] In an exemplary embodiment, the data processing method of the present application is applied to the field of bank fee reduction, and a fee forecasting and reduction system can be obtained. Figure 7 This is a schematic diagram of the framework of the fee prediction and reduction system. Figure 7 As shown, the fee prediction and reduction system includes two parts: a fee reduction and exemption prediction module and a fee reduction and exemption linkage processing module.

[0155] Figure 8 This is the unit structure diagram of the fee reduction prediction module. Figure 8 As shown, the fee reduction prediction module consists of five parts: a data acquisition unit, an ARIMA model unit, an LSTM model unit, a combined model prediction unit, and a result output unit. The data acquisition unit is used to obtain the historical data set of the target project and preprocess the historical data set to obtain the first target historical data set and the second target historical data set. That is, the main function of the data acquisition unit is to obtain the payment item information from the bank payment item database, obtain the charging and reduction records from the charging system, and send the collected data information to the subsequent model unit. The ARIMA model unit is used to determine the prediction result of the first dimension. That is, the ARIMA model unit uses the ARIMA model to analyze the historical data based on the data information of the data acquisition unit to predict the trend and annual change of the fee, and obtain the first prediction trend and prediction result identifier. The LSTM model unit is used to determine the second dimension de-test result, that is, the LSTM model unit uses the LSTM model to analyze the historical data based on the data information of the data acquisition unit to predict the occurrence and reduction ratio of the fee reduction, and obtain the second prediction trend and predicted reduction ratio. The combined model prediction unit is used to determine the prediction results, that is, the combined model prediction unit combines the outputs of ARIMA and LSTM to formulate the reduction and exemption strategy for the next charging cycle, including whether to exempt the handling fee and the reduction and exemption ratio. The result output unit sends the handling fee reduction and exemption linkage processing module according to the handling fee amount and reduction and exemption ratio of the exemption item.

[0156] Fig. 9This is the flowchart of the fee reduction prediction module. Fig. 9 As shown, the processing steps of the fee reduction prediction module include:

[0157] Step 901, the handling fee prediction and exemption system obtains payment item information and fee exemption information, obtains a historical data set, and pre-processes the historical data set to obtain a first target historical data set and a second target historical data set;

[0158] Step 902, construct an ARIMA model, analyze the first target historical data set, and obtain a first dimension prediction result.

[0159] Step 903, construct an LSTM model, analyze the second target historical data set, and obtain the second dimension prediction result.

[0160] Step 904, construct a combined model of ARIMA and LSTM, and output the prediction result of the next charging cycle.

[0161] Step 905: Send the prediction result to the fee reduction linkage processing module.

[0162] Fig.10 This is the unit structure diagram of the fee reduction linkage processing module. Fig.10 As shown, the fee reduction linkage processing module consists of two parts: an adaptive selection unit and an automatic processing unit for reduction and exemption procedures. The main function of the adaptive selection unit is to adaptively match and calculate the reduction and exemption ratio and fee amount calculated by the fee reduction and exemption prediction module, and pass the calculated fee reduction and exemption amount to the automatic processing unit for reduction and exemption procedures. When the reduction and exemption amount and ratio results of the payment item change, the system will automatically trigger the update of the reduction and exemption procedure. The main function of the automatic processing unit for reduction and exemption procedures is to receive the fee reduction and exemption amount of the payment item transmitted by the adaptive selection unit, automatically process the relevant reduction and exemption process in the charging system, and notify the branch to which the payment item belongs.

[0163] Fig.11 This is a flow chart of the fee reduction linkage processing module. Fig.11 As shown, the processing steps of the fee reduction linkage processing module include:

[0164] Step 1101, based on the prediction results, the handling fee prediction and reduction system adaptively matches the handling fee reduction strategy of the reduction and exemption project.

[0165] Step 1102, automatically process the reduction or exemption process of relevant fuel fee items and notify the affiliated branch.

[0166] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0167] This application uses the ARIMA and LSTM combined model and adaptive selection technology to build an intelligent and automated fee forecasting and exemption system. By analyzing the historical charging data of the payment items, the system can quickly and accurately predict when and under what circumstances the fee reduction will be made without manual intervention. And by combining time series analysis and deep learning technology, the advantages of both methods are fully utilized. This model integration can improve the reliability of the exemption prediction and ensure that the needs of different situations are better met. In addition, the fee forecasting and exemption system of this application also provides an automated processing mechanism that can trigger the reduction and exemption of fees for intermediary business income, thereby reducing the need for manual intervention by banks and improving business processing efficiency.

[0168] Based on the same inventive concept, the embodiment of the present application also provides a data processing device for implementing the data processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the one or more data processing device embodiments provided below can refer to the limitations on the data processing method above, and will not be repeated here.

[0169] In an exemplary embodiment, Fig.12 As shown, a data processing device 1200 is provided, comprising: an acquisition module 1201, a processing module 1202, a determination module 1203, a construction module 1204 and a sending module 1205, wherein:

[0170] The acquisition module 1201 is used to acquire the historical data set corresponding to the user's target project; the target project includes the exemption project and the associated project associated with the exemption project.

[0171] The processing module 1202 is used to pre-process the historical data set to obtain a first target historical data set and a second target historical data set.

[0172] The determination module 1203 is used to determine the prediction result of the first dimension based on the time series model and the first target historical data set, and to determine the prediction result of the second dimension based on the recurrent network model and the second target historical data set.

[0173] The construction module 1204 is used to construct the prediction results of the exemption items according to the prediction results of the first dimension and the prediction results of the second dimension.

[0174] The sending module 1205 is used to initiate the exemption procedure for the next time period of the exemption project based on the prediction result.

[0175] In an exemplary embodiment, the acquisition module 1201 includes:

[0176] The first determination submodule is used to determine the exemption item corresponding to the user identification according to the user identification.

[0177] The second determination submodule is used to determine the associated projects associated with the exemption project according to the target type to which the exemption project belongs.

[0178] The first acquisition submodule is used to acquire the project historical data set corresponding to the exemption project and the associated historical data set corresponding to the associated project, and construct the historical data set corresponding to the target project based on the project historical data set and the associated historical data set.

[0179] In an exemplary embodiment, the historical data set includes a project historical data set and a related historical data set, and the processing module 1202 includes:

[0180] The first grouping submodule is used to back up the historical exemption trend data in the project historical data set, and to perform grouping processing on the backed up project historical data set to obtain a first project historical data set and a second project historical data set.

[0181] The second grouping submodule is used to back up the associated historical exemption trend data in the associated historical data set, and to perform grouping processing on the backed up associated historical data set to obtain a first associated historical data set and a second associated historical data set.

[0182] The first processing submodule is used to perform weighted processing on the first project historical data set and the first associated historical data set to obtain a first target historical data set.

[0183] The first construction submodule is used to construct a second target historical data set according to the second project historical data set and the second associated historical data set.

[0184] In an exemplary embodiment, the processing module 1203 includes:

[0185] The second processing submodule is used to perform prediction processing on the first target historical data set according to the time series model, obtain the prediction result identifier and the first prediction trend, and construct the first dimension prediction result according to the prediction result identifier and the first prediction trend.

[0186] The third processing submodule is used to perform prediction processing on the second target historical data set according to the recurrent network model to obtain the predicted exemption ratio and the second predicted trend, and to construct a second dimension prediction result according to the predicted exemption ratio and the second predicted trend.

[0187] In an exemplary embodiment, the construction module 1204 includes:

[0188] The first judgment submodule is used to judge whether the prediction result identifier in the first dimension prediction result is a target prediction result identifier.

[0189] The second judgment submodule is used to judge whether the first prediction trend in the first dimension prediction result and the second prediction trend in the second dimension prediction result are consistent if the prediction result identifier is the target prediction result identifier.

[0190] The second construction submodule is used to construct the prediction results of the exemption items according to the predicted exemption ratios in the first dimension prediction results and the second dimension prediction results if the first prediction trend and the second prediction trend are consistent.

[0191] In an exemplary embodiment, the data processing device 1200 further includes:

[0192] The second determination module is used to determine that resource reduction will not be performed in the next time period of the reduction or exemption project if the prediction result identifier is not the target prediction result identifier.

[0193] The third determining module is used to determine that resource reduction will not be performed in the next time period of the reduction or exemption project as a prediction result of the reduction or exemption project.

[0194] In an exemplary embodiment, the data processing device 1200 further includes:

[0195] An execution module is used to, if the first prediction trend and the second prediction trend are inconsistent, execute the steps of determining the first dimension prediction result based on the time series model and the first target historical data set, and determining the second dimension prediction result based on the recurrent network model and the second target historical data set, until the first prediction trend and the second prediction trend are consistent.

[0196] Each module in the above data processing device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0197] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Fig.13 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a data processing method is realized. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0198] Those skilled in the art will understand that Fig.13 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0199] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0200] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0201] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0202] 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, use and processing of relevant data must comply with relevant regulations. In addition, the information collected by this application is 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 relevant data comply with the relevant laws, regulations and standards of relevant countries and regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation portals for users to choose to authorize or refuse.

[0203] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0204] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0205] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A data processing method, characterized in that: The method comprises: Acquire a historical data set corresponding to a user's target project; the target project includes a reduction or exemption project and associated projects associated with the reduction or exemption project; Preprocessing the historical data set to obtain a first target historical data set and a second target historical data set; Determine a first dimension prediction result based on a time series model and the first target historical data set, and determine a second dimension prediction result based on a recurrent network model and the second target historical data set; Constructing the prediction result of the exemption project according to the prediction result of the first dimension and the prediction result of the second dimension; Based on the prediction results, the exemption procedure for the next time period of the exemption project is initiated.

2. The method according to claim 1, characterized in that: The step of obtaining a historical data set corresponding to the user's target project includes: Determining the exemption item corresponding to the user identifier according to the user identifier; Determining the associated project associated with the exemption project according to the target type to which the exemption project belongs; The project history data set corresponding to the exemption project and the associated history data set corresponding to the associated project are obtained, and the history data set corresponding to the target project is constructed according to the project history data set and the associated history data set.

3. The method according to claim 1, characterized in that The historical data set includes a project historical data set and an associated historical data set, and the preprocessing of the historical data set to obtain a first target historical data set and a second target historical data set includes: Backing up the historical exemption trend data in the project historical data set, and grouping the backed-up project historical data set to obtain a first project historical data set and a second project historical data set; Backing up the associated historical exemption trend data in the associated historical data set, and performing grouping processing on the backed up associated historical data set to obtain a first associated historical data set and a second associated historical data set; Performing weighted processing on the first project historical data set and the first associated historical data set to obtain a first target historical data set; A second target historical data set is constructed according to the second project historical data set and the second associated historical data set.

4. The method according to claim 1, characterized in that The determining of the first dimension prediction result based on the time series model and the first target historical data set, and determining the second dimension prediction result based on the recurrent network model and the second target historical data set, includes: Performing forecasting processing on the first target historical data set according to the time series model to obtain a forecast result identifier and a first forecast trend, and constructing a first dimension forecast result according to the forecast result identifier and the first forecast trend; The second target historical data set is predicted and processed according to the recurrent network model to obtain a predicted exemption ratio and a second predicted trend, and a second dimension prediction result is constructed according to the predicted exemption ratio and the second predicted trend.

5. The method according to claim 1, characterized in that The constructing the prediction result of the exemption project according to the prediction result of the first dimension and the prediction result of the second dimension includes: Determine whether the prediction result identifier in the prediction result of the first dimension is a target prediction result identifier; If the prediction result identifier is a target prediction result identifier, determining whether a first prediction trend in the prediction result of the first dimension is consistent with a second prediction trend in the prediction result of the second dimension; If the first prediction trend is consistent with the second prediction trend, the prediction result of the exemption project is constructed according to the predicted exemption ratio in the first dimension prediction result and the second dimension prediction result.

6. The method according to claim 5, characterized in that After determining whether the prediction result identifier in the prediction result of the first dimension is a target prediction result identifier, the method further includes: If the prediction result identifier is not the target prediction result identifier, determining that resource reduction will not be performed in the next time period of the reduction item; Determining that resource reduction is not performed in the next time period of the reduction and exemption item is a prediction result of the reduction and exemption item.

7. The method according to claim 5, characterized in that After determining whether the first prediction trend in the first dimension prediction result and the second prediction trend in the second dimension prediction result are consistent, the method further includes: If the first prediction trend is inconsistent with the second prediction trend, execute the steps of determining the first dimension prediction result based on the time series model and the first target historical data set, and determining the second dimension prediction result based on the recurrent network model and the second target historical data set until the first prediction trend is consistent with the second prediction trend.

8. A data processing device, characterized in that: The device comprises: An acquisition module, used to acquire a historical data set corresponding to a user's target project; the target project includes a reduction or exemption project and an associated project associated with the reduction or exemption project; A processing module, used for preprocessing the historical data set to obtain a first target historical data set and a second target historical data set; A determination module, configured to determine a first dimension prediction result based on a time series model and the first target historical data set, and to determine a second dimension prediction result based on a recurrent network model and the second target historical data set; A construction module, used to construct the prediction result of the exemption project according to the prediction result of the first dimension and the prediction result of the second dimension; A sending module is used to initiate the exemption procedure for the next time period of the exemption project based on the prediction result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images