Business processing method and device, electronic equipment, storage medium and program product

Abstract

The invention provides a business processing method and device, electronic equipment, a storage medium and a program product. Relates to the technical field of data processing. The method is applied to a server, and comprises the following steps: receiving a service amount prediction request sent by a client to determine user feature data and historical time sequence data of a target user, and performing supervised binning operation on continuous variables based on difference data of target variables to obtain corresponding binning results, and generating a corresponding binning time sequence feature based on the binning result, inputting the historical time sequence data and the binning time sequence feature into a trained data prediction model, and predicting the fluctuation trend of the business amount based on the data prediction model to obtain a corresponding prediction result. Based on the business processing method provided by the invention, the accuracy of predicting the fluctuation of the business amount can be improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a business processing method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] One of the core functions of banking institutions is to predict the future flow trends of pre-defined business amounts to support fund adjustments and strategy formulation. In practice, business payment behavior is closely related to user characteristics (such as age, payment duration, and income level). Significant differences exist among user groups with different individual characteristics in payment amounts, loan application frequency, and fund usage preferences, and these differences are directly reflected in fluctuations in business amounts.

[0003] Currently, related technologies typically predict fluctuations in transaction amounts by relying on the autocorrelation of users' historical payment data. However, this ignores the potential impact of continuous variables in user characteristics (such as user age and payment duration) on transaction amounts, resulting in low accuracy in predicting fluctuations in transaction amounts. Summary of the Invention

[0004] This application provides a business processing method, apparatus, electronic device, storage medium, and program product to improve the accuracy of predicting fluctuations in business volume.

[0005] In a first aspect, embodiments of this application provide a business processing method applied to a server, comprising:

[0006] Receive a business amount prediction request sent by the client. The business amount prediction request includes the target user's user characteristic data and historical time series data.

[0007] Identify the continuous variables in the feature data, and the target variables corresponding to the time series data;

[0008] Determine multiple initial binning intervals corresponding to continuous variables;

[0009] Determine the difference data of the target variable within any two adjacent initial binning intervals;

[0010] Supervised binning is performed on continuous variables based on the differential data of the target variables to obtain the corresponding binning results. The initial binning results include multiple target binning intervals corresponding to the continuous variables.

[0011] Determine the characteristic data of the preset indicators corresponding to each target binning interval in the binning results;

[0012] Based on the feature data of the preset indicators corresponding to each target binning interval, the corresponding binning time series features are generated.

[0013] Historical time series data and binned time series features are input into a trained data prediction model to obtain prediction results. The prediction results are used to indicate the business flow data of the target user's pre-set business.

[0014] Send the prediction results to the client.

[0015] In one possible implementation, supervised binning is performed on continuous variables based on the variance data of the target variable to obtain the corresponding binning results, including:

[0016] Determine the distributional complexity of the target variable;

[0017] By combining the distributional complexity of the standard variable and the difference data of the target variable, the initial binning intervals are merged to obtain multiple corresponding target binning intervals.

[0018] In one possible implementation, before determining the multiple initial binning intervals corresponding to the continuous variables, the method further includes:

[0019] By plotting histograms of the target variable on continuous variables, the distribution pattern of the target variable can be identified through the histograms;

[0020] Based on the distribution pattern of the target variable, the preset interval length is determined.

[0021] In one possible implementation, corresponding binning time series features are generated based on the binning results, including:

[0022] Determine the target binning interval corresponding to the binning results;

[0023] Determine the percentage of users in the target binning interval at the preset time point;

[0024] By using user number percentage data, we can determine the changing trend of the user number percentage corresponding to the target container interval;

[0025] The changing trend of the proportion of users was identified as a time series feature of binning.

[0026] In one possible implementation, it also includes:

[0027] Determine the real-time changes in the characteristics of the binning time series;

[0028] The binning results are updated based on real-time changes in binning time series characteristics.

[0029] In one possible implementation, it also includes:

[0030] Obtain the pre-defined distribution differences of economic indicators;

[0031] The binning results are updated based on the pre-defined differences in the distribution of economic indicators.

[0032] Secondly, embodiments of this application provide a business processing apparatus applied to a server, comprising:

[0033] The receiving module is used to receive business amount prediction requests sent by the client. The business amount prediction requests include user characteristic data and historical time series data of the target user.

[0034] The processing module is used to determine the continuous variables in the feature data and the target variables corresponding to the time series data;

[0035] The processing module is also used to determine multiple initial binning intervals corresponding to continuous variables;

[0036] The processing module is also used to determine the difference data of the target variable within any two adjacent initial binning intervals;

[0037] The processing module is also used to perform supervised binning operations on continuous variables based on the difference data of the target variable to obtain the corresponding binning results, wherein the initial binning results include multiple target binning intervals corresponding to the continuous variables.

[0038] The determination module is also used to determine the feature data of preset indicators corresponding to each target binning interval in the binning results;

[0039] The processing module is also used to generate corresponding bin time series features based on the feature data of preset indicators corresponding to each target bin interval;

[0040] The processing module is also used to input historical time series data and binned time series features into the trained data prediction model to obtain prediction results. The prediction results are used to indicate the business flow data of the target user's pre-set business.

[0041] The sending module is used to send the prediction results to the client.

[0042] In one possible implementation, the processing module is specifically used to determine the distributional complexity of the target variable;

[0043] The processing module is also used to combine the distribution complexity of the standard variable and the difference data of the target variable to merge the initial binning intervals to obtain multiple corresponding target binning intervals.

[0044] In one possible implementation, the processing module is further configured to identify the distribution pattern of the target variable by plotting a histogram of the target variable on a continuous variable.

[0045] The processing module is also used to determine the preset interval length based on the distribution pattern of the target variable.

[0046] In one possible implementation, the processing module is further configured to determine the target binning interval corresponding to the binning result;

[0047] The processing module is also specifically used to determine the percentage of users in the target bin area at a preset time point;

[0048] The processing module is also used to determine the trend of the change in the proportion of users in the target bin area by using the user proportion data;

[0049] The processing module is also used to determine the trend of changes in the proportion of users as binning time series features.

[0050] In one possible implementation, the processing module is also used to determine real-time change data of binning time series characteristics;

[0051] The processing module is also used to update the binning results based on real-time changes in binning time series characteristics.

[0052] In one possible implementation, the processing module is further configured to obtain preset economic indicator distribution differences;

[0053] The processing module is also used to update the binning results based on the differences in the distribution of preset economic indicators.

[0054] Thirdly, embodiments of this application provide an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0055] The memory stores the instructions that the computer executes;

[0056] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0057] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0058] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0059] The business processing method, apparatus, electronic device, storage medium, and program product provided in this application analyze user characteristic data to determine continuous variables and target variables within the user characteristic data. Supervised binning generates binning results strongly correlated with the target variables. Corresponding binning time series features are obtained from the binning results. Historical time series data and binning time series features are input into a trained data prediction model to obtain prediction results. This application, based on binning time series features, can capture the dynamic changes in user group behavior, making the prediction results of business amount fluctuation trends more closely match the behavioral patterns of specific user groups, thereby improving the accuracy of the prediction results. Attached Figure Description

[0060] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0061] Figure 1 This application provides a schematic diagram of the scenario.

[0062] Figure 2 Flowchart of the business processing method provided in this application Figure 1 ;

[0063] Figure 3 Flowchart of the business processing method provided in this application Figure 2 ;

[0064] Figure 4 Flowchart of the business processing method provided in this application Figure 3 ;

[0065] Figure 5 Flowchart of the business processing method provided in this application Figure 4 ;

[0066] Figure 6 Flowchart of the business processing method provided in this application Figure 5 ;

[0067] Figure 7 Flowchart of the business processing method provided in this application Figure 6 ;

[0068] Figure 8 A schematic diagram of the business processing device provided in this application;

[0069] Figure 9 A schematic diagram of the structure of the electronic device provided in this application.

[0070] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0071] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0072] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0073] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0074] One of the core functions of banking institutions is to predict the future flow trends of pre-defined business amounts to support fund adjustments and strategy formulation. In practice, business payment behavior is closely related to user characteristics (such as age, payment duration, and income level). Significant differences exist among user groups with different individual characteristics in payment amounts, loan application frequency, and fund usage preferences, and these differences are directly reflected in fluctuations in business amounts.

[0075] Currently, related technologies typically predict fluctuations in transaction amounts by relying on the autocorrelation of users' historical payment data. However, this ignores the potential impact of continuous variables in user characteristics (such as user age and payment duration) on transaction amounts, resulting in low accuracy in predicting fluctuations in transaction amounts.

[0076] Figure 1 The scenario diagram provided for this application is as follows: Figure 1As shown, the specific application scenarios of this application include servers and data prediction models. The server analyzes user characteristic data to determine continuous variables and target variables within the user characteristic data. Supervised binning generates binning results strongly correlated with the target variables. The corresponding binning time series features can be obtained from the binning results. Historical time series data and binning time series features are input into a trained data prediction model to obtain prediction results. This application, based on binning time series features, can capture the dynamic changes in user group behavior, making the prediction results of business volume fluctuation trends more closely match the behavioral patterns of specific user groups, thereby improving the accuracy of the prediction results.

[0077] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0078] Figure 2 Flowchart of the business processing method provided in this application Figure 1 ,like Figure 2 As shown, this method is applied to a server and includes:

[0079] S201. Receive a business amount prediction request sent by the client. The business amount prediction request includes user characteristic data and historical time series data of the target user.

[0080] Combined with scenario examples, Figure 1 In this embodiment, the execution entity is a server. The business amount prediction request can be a prediction request for a preset business amount fluctuation. The preset business can be determined according to the actual situation, for example, it can be a housing provident fund. User characteristic data mainly includes the target user's age, name, housing provident fund contribution period, and income level. Historical time series data refers to the target user's historical contribution amount, loan application records, etc. When the preset business is a housing provident fund, the historical contribution amount can be the target user's historical housing provident fund contribution amount.

[0081] S202. Determine the continuous variables in the feature data and the target variables corresponding to the time series data.

[0082] In the context of specific scenarios, continuous variables refer to variables that take any real value within a certain range, such as a user's age or contribution period. Target variables refer to continuous variables that need to be predicted, such as the amount of housing provident fund contributions.

[0083] S203. Determine multiple initial binning intervals corresponding to continuous variables.

[0084] Using a scenario example and age as a continuous variable, multiple initial binning intervals can be obtained by using 10-year intervals, such as: [20, 30), [30, 40), [40, 50), and [50, 60).

[0085] S204. Determine the difference data of the target variable within any two adjacent initial binning intervals.

[0086] In conjunction with scenario examples, the difference data can be the mean or variance of the target variable. For example, the difference data of the target variable can be the difference in the mean of the target variable. For instance, if the target variable is the housing provident fund contribution amount, the corresponding housing provident fund contribution amount in the interval [30, 40) is 1000 yuan, and the corresponding housing provident fund contribution amount in the interval [40, 50) is 1200 yuan, then the difference data of the target variable can be obtained as 200 yuan.

[0087] S205. Perform supervised binning on continuous variables based on the difference data of the target variables to obtain the corresponding binning results, wherein the initial binning results include multiple target binning intervals corresponding to the continuous variables.

[0088] In the context of a scenario example, supervised binning refers to the discretization process of continuous variables obtained from the difference data of the target variable. Optionally, supervised binning can be chi-square binning, which determines the initial binning intervals for the continuous variable and calculates the distribution differences of the target variable within adjacent initial binning intervals. The intervals are then merged or split to ensure that the binning results reflect the correlation between user behavior and the target variable, thus obtaining multiple corresponding target binning intervals.

[0089] Optional, Figure 3 Flowchart of the business processing method provided in this application Figure 2 ,like Figure 3 As shown, S205 includes:

[0090] S301. Determine the distributional complexity of the target variable.

[0091] In the context of scenario examples, distribution complexity refers to the complexity of the distribution pattern of the target variable over continuous variables, such as multimodal, unimodal, or uniform distributions. For example, if the contribution amount distribution of age data exhibits a bimodal shape, the distribution complexity is high. Analyze the distribution complexity of the target variable (e.g., contribution amount) over continuous variables (e.g., age). For example, use preset indicators (e.g., kurtosis, skewness) to determine if the data exhibits a multimodal distribution. If the distribution complexity is high (e.g., a multimodal pattern exists), retain more bins to reflect the behavioral differences among different user groups; if the distribution complexity is low (e.g., an approximately unimodal distribution), reduce the number of bins to avoid overfitting.

[0092] S302. Combining the distribution complexity of the standard variable and the difference data of the target variable, the initial binning intervals are merged to obtain multiple corresponding target binning intervals.

[0093] In the context of a scenario example, merging initial binning intervals refers to adjusting the boundaries of the binning intervals based on the magnitude of the mean difference. For instance, if the difference between adjacent binning intervals is less than a preset threshold, they are merged into one binning interval; if the difference is greater than the preset threshold, they are retained as independent binning intervals. The preset threshold can be determined based on the actual situation. If the preset threshold is 50 yuan, and the difference between the target variable in intervals [30, 40) and [40, 50) is 200 yuan, exceeding 50 yuan, then intervals [30, 40) and [40, 50) can be merged into one interval [30, 50), in which case interval [30, 50) becomes a target binning interval. Conversely, if the preset threshold is 500 yuan, then the difference between the target variables in intervals [30, 40) and [40, 50) does not exceed 500 yuan. In this case, intervals [30, 40) and [40, 50) are retained as two separate binning intervals, and intervals [30, 40) and [40, 50) are two target binning intervals respectively.

[0094] Based on the method provided in this example, refined adjustments to binning intervals are achieved by introducing quantitative calculations of distribution complexity. This technique avoids the problem of relying on subjective experience to divide intervals in traditional binning methods by quantifying the distribution differences of the target variable within adjacent intervals. Merging intervals reduces redundancy in the binning results. Furthermore, this quantitative calculation enhances the interpretability of the binning process, allowing for an intuitive understanding of the differences in contribution behavior among users of different age groups through the numerical values ​​of mean differences, thereby optimizing fund adjustment strategies.

[0095] Optional, Figure 4 Flowchart of the business processing method provided in this application Figure 3 ,like Figure 4 As shown, before S302, it also includes:

[0096] S401. By plotting histograms of the target variable on continuous variables, the distribution pattern of the target variable can be identified through the histograms.

[0097] Based on scenario examples, the distribution pattern of the target variable can be determined by the distribution complexity of the target variable. The distribution complexity of the target variable refers to the complexity of the distribution shape of the target variable on continuous variables. The distribution pattern can be multimodal, unimodal, or uniform.

[0098] Optionally, using age as a continuous variable, a histogram can be plotted based on the binning intervals of the age output described above, and the corresponding target variable within each interval, to determine the distribution pattern of the target variable. If the distribution of the target variable in the histogram shows two or more distinct peaks, it indicates that there are multiple major groups within the target variable's dataset. For example, the 20-30 year old and 40-50 year old groups among users may correspond to high and low housing provident fund contributions, respectively. In this case, the distribution pattern of the target variable can be determined as a multimodal distribution.

[0099] S402. Based on the distribution pattern of the target variable, determine the preset interval length.

[0100] Based on the distribution pattern obtained above, the interval length can be adaptively adjusted. For example, if the distribution complexity is high, such as a multi-peak pattern, more bins can be retained to reflect the behavioral differences among different user groups. In this case, a smaller interval length can be determined, such as 10 years old per interval. If the distribution complexity is low, such as an approximately unimodal distribution, fewer bins can be used to avoid overfitting. In this case, a larger interval length can be determined, such as 20 years old per interval.

[0101] Based on the method provided in this example, the interval length can be flexibly determined by the distribution pattern of the target variable, thereby reducing the randomness of the binning process.

[0102] S206. Determine the characteristic data of the preset indicators corresponding to each target binning interval in the binning results.

[0103] Based on scenario examples, the characteristic data of the preset indicator can be the percentage of users.

[0104] S207. Based on the feature data of the preset indicators corresponding to each target binning interval, generate the corresponding binning time series features.

[0105] With a scenario example, time series features refer to features that change over time, such as the changing trend of the proportion of users in each target bin interval.

[0106] Optional, Figure 5 Flowchart of the business processing method provided in this application Figure 4 ,like Figure 5 As shown, S205 includes:

[0107] S501. Determine the target binning interval corresponding to the binning results.

[0108] Based on the above content and in conjunction with the scenario example, if the intervals [30, 40) and [40, 50) are merged into the interval [30, 50), then the corresponding target binning intervals can be: [20, 30), [30, 50) and [50, 60).

[0109] S502. Determine the percentage of users in the target binning interval at the preset time point.

[0110] Based on the scenario example, the preset time point can be the beginning of each month (e.g., the 1st of each month) or a preset time point determined according to the actual situation. User percentage data refers to the ratio of the number of users within a specific target binning interval to the total number of users. For example, if the number of users in the age group corresponding to the [20, 30) interval accounts for 40% of the total number of users in January of a certain year, then the user percentage data for the [20, 30) interval in January of that year is 0.4.

[0111] S503. Determine the trend of user number percentage change corresponding to the target container interval by using user number percentage data.

[0112] In the context of a scenario, the trend refers to the pattern of change in the percentage of users over time, such as an increase, a decrease, or fluctuation.

[0113] S504. The trend of changes in the proportion of user numbers is determined as a time series feature of binning.

[0114] Using a scenario example, this step generates time-series features for each target binning interval by recording the percentage of users in different months for each target binning interval. For instance, if the binning interval is [20, 30), then the monthly change in the percentage of users in this interval is determined to obtain the time-series features for the [20, 30) interval. For example, the time-series feature for the [20, 30) interval could be "the percentage of users in the [20, 30) age group increases month by month." Similarly, the time-series features for each target binning interval can be obtained in this way.

[0115] Based on the method provided in this example, by incorporating the trend of changes in the proportion of users, the analysis of user behavior patterns is realized, enabling subsequent prediction models to simultaneously reflect the long-term trends and short-term fluctuations of the user group, and improving the accuracy of prediction results by combining changes in the user group.

[0116] S208. Input the historical time series data and binned time series features into the trained data prediction model to obtain the prediction results. The prediction results are used to indicate the business flow data of the target user's pre-set business.

[0117] In the context of a scenario example, the data prediction model can be the Autoregressive Integrated Moving Average (ARIMA) model. This model is a linear prediction model based on time series data. By analyzing the autocorrelation, stationarity, and error term characteristics of the data, it captures the long-term trends, seasonal fluctuations, and random fluctuations of the time series. It is suitable for short- to medium-term predictions of univariate and continuous time series. Therefore, the ARIMA model can be used to analyze historical time series data and binned time series characteristics to predict future fluctuation trends in business revenue.

[0118] S209. Send the prediction results to the client.

[0119] After obtaining the corresponding prediction results using scenario examples, the prediction results are finally sent to the user's client so that the user is aware of the prediction results for the business amount.

[0120] Based on the method provided in this example, the dynamic changes in user group behavior can be captured by binning time series characteristics, making the prediction results of business amount fluctuation trends more consistent with the behavior patterns of specific user groups, thereby improving the accuracy of the prediction results.

[0121] Optional, Figure 6 Flowchart of the business processing method provided in this application Figure 5 ,like Figure 6 As shown, it also includes:

[0122] S601. Determine the real-time changes in the binning time series characteristics.

[0123] Based on scenario examples, a sliding time window mechanism is introduced on top of chi-square binning. This mechanism divides time series data into continuous time windows, such as one month, and performs histogram discretization and chi-square binning independently within each window. Real-time changes in binned time series features can be monitored through the sliding time window mechanism, which updates the sliding window accordingly.

[0124] S602. Update the binning results based on real-time changes in binning time series characteristics.

[0125] With scenario examples, a sliding time window mechanism is used to dynamically adjust the binning intervals based on real-time changes in time series data. For instance, a monthly updated sliding window is used, with histogram discretization and chi-square binning performed independently within each window to update the binning results for each time window. This ensures that the binning results are always based on the latest data, avoiding the lag caused by fixed binning and improving the predictive model's responsiveness to short-term fluctuations.

[0126] Optional, Figure 7 Flowchart of the business processing method provided in this application Figure 6 ,like Figure 7 As shown, it also includes:

[0127] S701, Obtain the preset economic indicator distribution differences.

[0128] In conjunction with scenario examples, differences in the distribution of economic indicators can be data such as the trend of changes in regional Gross Domestic Product (GDP) and interest rate trends.

[0129] S702. Update the binning results based on the preset differences in the distribution of economic indicators.

[0130] In practical examples, changes in economic indicators can affect changes in the target variable. For instance, a decline in regional GDP may reduce users' contributions, and binning intervals may be more inclined to classify low-income, short-term contributors, thus providing early warnings of liquidity issues. Therefore, updating binning results based on differences in the distribution of economic indicators allows the binning results to be directly correlated with changes in the macroeconomic environment. By integrating external economic indicators, the predictive model achieves sensitivity modeling to changes in the macroeconomic environment, avoiding prediction biases caused by single factors.

[0131] Based on the business processing method provided in this application, continuous variables and target variables in user feature data can be determined by analyzing user feature data. Supervised binning generates binning results strongly correlated with the target variables. Corresponding binning time series features can be obtained from the binning results. Historical time series data and binning time series features are input into a trained data prediction model to obtain prediction results. This application, based on binning time series features, can capture the dynamic changes in user group behavior, making the prediction results of business amount fluctuation trends more closely match the behavioral patterns of specific user groups, thereby improving the accuracy of the prediction results.

[0132] Figure 8 The schematic diagram of the business processing device provided in this application is applied to a server, such as... Figure 8 As shown, it includes:

[0133] The receiving module 81 is used to receive a business amount prediction request sent by the client. The business amount prediction request includes user characteristic data and historical time series data of the target user.

[0134] Processing module 82 is used to determine the continuous variables in the feature data and the target variables corresponding to the time series data;

[0135] Processing module 82 is also used to determine multiple initial binning intervals corresponding to continuous variables;

[0136] The processing module 82 is also used to determine the difference data of the target variable within any two adjacent initial binning intervals;

[0137] The processing module 82 is also used to perform supervised binning operations on continuous variables based on the difference data of the target variable to obtain the corresponding binning results, wherein the initial binning results include multiple target binning intervals corresponding to the continuous variables.

[0138] The determination module 82 is also used to determine the feature data of the preset indicators corresponding to each target binning interval in the binning results;

[0139] The processing module 82 is also used to generate corresponding bin time series features based on the feature data of preset indicators corresponding to each target bin interval;

[0140] Processing module 82 is also used to input historical time series data and binned time series features into the trained data prediction model to obtain prediction results, which are used to indicate the business flow data of the target user's pre-set business.

[0141] The sending module 83 is used to send the prediction results to the client.

[0142] Optionally, processing module 82 is specifically used to determine the distributional complexity of the target variable;

[0143] The processing module 82 is also used to combine the distribution complexity of the standard variable and the difference data of the target variable to merge the initial binning intervals to obtain multiple target binning intervals.

[0144] Optionally, the processing module 82 is further configured to identify the distribution pattern of the target variable by plotting a histogram of the target variable on a continuous variable.

[0145] The processing module 82 is also used to determine the preset interval length based on the distribution pattern of the target variable.

[0146] Optionally, the processing module 82 is also used to determine the target binning interval corresponding to the binning result;

[0147] Processing module 82 is also specifically used to determine the percentage of users in the target bin area at a preset time point;

[0148] Processing module 82 is also used to determine the trend of user number percentage change corresponding to the target bin interval by using user number percentage data;

[0149] The processing module 82 is also used to determine the trend of changes in the proportion of user numbers as a binning time series feature.

[0150] Optionally, the processing module 82 is also used to determine the real-time changes in binning time series characteristics;

[0151] The processing module 82 is also used to update the binning results based on the real-time changing data of the binning time series characteristics.

[0152] Optionally, the processing module 82 is also used to obtain preset economic indicator distribution differences;

[0153] The processing module 82 is also used to update the binning results based on the preset differences in the distribution of economic indicators.

[0154] The business processing device provided in this embodiment can execute the business processing method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0155] Figure 9 A schematic diagram of the structure of the electronic device provided in this application. Figure 9 As shown, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the electronic device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus.

[0156] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0157] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0158] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0159] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0160] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0161] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0162] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0163] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0164] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0165] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0166] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0167] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0168] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0169] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0170] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A business processing method, characterized in that, Applied to servers, including: Receive a business amount prediction request sent by the client, the business amount prediction request including user characteristic data and historical time series data of the target user; Identify the continuous variables in the feature data, and the target variables corresponding to the time series data; Determine multiple initial binning intervals corresponding to the continuous variables; Determine the difference data of the target variable within any two adjacent initial binning intervals; Based on the difference data of the target variable, a supervised binning operation is performed on the continuous variable to obtain the corresponding binning results, wherein the initial binning results include multiple target binning intervals corresponding to the continuous variable; In the binning results, the characteristic data of the preset indicators corresponding to each target binning interval are determined; Based on the feature data of the preset indicators corresponding to each target compartment, the corresponding compartment time series features are generated. The historical time series data and the binned time series features are input into the trained data prediction model to obtain the prediction result, which is used to indicate the business flow data of the target user's pre-set business. The prediction result is sent to the client.

2. The method according to claim 1, characterized in that, The supervised binning operation on the continuous variable based on the difference data of the target variable to obtain the corresponding binning results includes: Determine the distributional complexity of the target variable; By combining the distribution complexity of the standard variable and the difference data of the target variable, the initial binning intervals are merged to obtain multiple corresponding target binning intervals.

3. The method according to claim 2, characterized in that, Before determining the multiple initial binning intervals corresponding to the continuous variable, the method further includes: By plotting histograms of the target variable on the continuous variable, the distribution pattern of the target variable can be identified through the histograms; Based on the distribution pattern of the target variable, the preset interval length is determined.

4. The method according to claim 1, characterized in that, The step of generating corresponding binning time series features based on the binning results includes: Determine the target binning interval corresponding to the binning results; Determine the percentage of users in the target binning interval at a preset time point; By using the user number percentage data, the trend of user number percentage change corresponding to the target container interval is determined; The trend of the change in the proportion of the number of users is determined as the time series feature of the binning.

5. The method according to any one of claims 1-4, characterized in that, Also includes: Determine the real-time changes in the binning time series characteristics; The binning results are updated based on the real-time changes in the binning time series characteristics.

6. The method according to claim 5, characterized in that, Also includes: Obtain the pre-defined distribution differences of economic indicators; The binning results are updated based on the pre-defined differences in the distribution of economic indicators.

7. A business processing apparatus, characterized in that, Applied to servers, including: The receiving module is used to receive a business amount prediction request sent by the client. The business amount prediction request includes user characteristic data and historical time series data of the target user. The processing module is used to determine the continuous variables in the feature data and the target variables corresponding to the time series data; The processing module is also used to determine multiple initial binning intervals corresponding to the continuous variables; The processing module is also used to determine the difference data of the target variable within any two adjacent initial binning intervals; The processing module is also used to perform supervised binning on the continuous variable based on the difference data of the target variable to obtain the corresponding binning result, wherein the initial binning result includes multiple target binning intervals corresponding to the continuous variable; The determining module is also used to determine the feature data of the preset indicators corresponding to each target binning interval in the binning results; The processing module is also used to generate corresponding bin time series features based on the feature data of the preset indicators corresponding to each target bin interval; The processing module is also used to input the historical time series data and the binned time series features into the trained data prediction model to obtain prediction results, which are used to indicate the business flow data of the target user's pre-set business. The sending module is used to send the prediction result to the client.

8. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-6.

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

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-6.

Images