Resource transfer behavior data prediction method and device, computer equipment, readable storage medium and program product

By obtaining user and product identification, combining the first and second resource transfer historical behavior data to generate sample data, and using neural network model training, the problem of inaccurate prediction of resource transfer behavior data in the existing technology is solved, and the prediction accuracy and Data processing efficiency.

CN118916785APending Publication Date: 2024-11-08IND CONSUMER FINANCE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411041287.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

In the prior art, the data prediction method for resource transfer behavior is inaccurate due to insufficient completed data.

Method used

By obtaining user and product identification, combining the first and second resource transfer historical behavior data, sample data is generated, and using neural network model training, a resource transfer behavior data prediction model is generated, expanding the data volume to improve prediction accuracy.

Benefits of technology

It improves the accuracy of the data prediction model of resource transfer behavior, reduces the inaccuracy of prediction results, and enhances data processing efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118916785A_ABST
    Figure CN118916785A_ABST
Patent Text Reader

Abstract

The invention relates to a resource transfer behavior data prediction method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps of obtaining a resource transfer behavior data prediction request, and obtaining first resource transfer historical behavior data according to a user identifier and a product identifier carried in the resource transfer behavior data prediction request; acquiring second resource transfer historical behavior data according to the user identifier; generating sample data according to the first resource transfer historical behavior data and the second resource transfer historical behavior data; training a neural network model by using the sample data to obtain a resource transfer behavior data prediction model; and inputting the target prediction time and the user identifier into a resource transfer behavior data prediction model to obtain a resource transfer behavior data prediction result output by the resource transfer behavior data prediction model. By adopting the method, the accuracy of resource transfer behavior data prediction can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a resource transfer behavior data prediction method, device, computer equipment, computer-readable storage medium and computer program product. Background Art

[0002] With the development of computer technology, more and more companies use computer technology to process daily business, such as approving product application requests, predicting resource transfer behavior data after approval, etc.

[0003] In the related art, the prediction of resource transfer behavior data generally uses completed resource transfer behavior data to predict subsequent resource transfer behavior data. However, the completed resource transfer behavior data has defects such as insufficient data volume. Therefore, the prediction method of resource transfer behavior data in the related art has inaccurate predictions. Summary of the invention

[0004] Based on this, it is necessary to provide a resource transfer behavior data prediction method, device, computer equipment, computer-readable storage medium and computer program product that can improve data accuracy in response to the above technical problems.

[0005] In a first aspect, the present application provides a resource transfer behavior data prediction method, comprising:

[0006] Obtaining a resource transfer behavior data prediction request, where the resource transfer behavior data prediction request carries a user identifier and a product identifier;

[0007] Acquire first resource transfer history behavior data according to the user identifier and the product identifier;

[0008] Acquire second resource transfer history behavior data according to the user identifier, where the second resource transfer history behavior data is data used to calculate the credit data of the user corresponding to the user identifier;

[0009] Generate sample data according to the first resource transfer history behavior data and the second resource transfer history behavior data;

[0010] The neural network model is trained using sample data to obtain a resource transfer behavior data prediction model;

[0011] The target prediction time and the user identifier are input into the resource transfer behavior data prediction model to obtain the resource transfer behavior data prediction result output by the resource transfer behavior data prediction model.

[0012] In one embodiment, sample data is generated based on first resource transfer historical behavior data and second resource transfer historical behavior data, and the resource transfer behavior data prediction method also includes: obtaining a first start time and a first end time of the first resource transfer historical behavior data, and determining time interval data based on the first start time and the first end time; when the time interval data is less than a preset time interval, determining the second start time based on the preset time interval and the first end time; obtaining first resource transfer supplementary behavior data from the second resource transfer historical behavior data based on the first start time and the second start time; and generating sample data based on the first resource transfer historical behavior data and the first resource transfer supplementary behavior data.

[0013] In one of the embodiments, after generating sample data based on the first resource transfer historical behavior data and the first resource transfer supplementary behavior data, the resource transfer behavior data prediction method also includes: when there is a missing data in the first resource transfer historical behavior data, obtaining the first time information corresponding to the data missing location; obtaining the second resource transfer supplementary behavior data from the second resource transfer historical behavior data based on the first time information; and updating the sample data based on the second resource transfer supplementary behavior data.

[0014] In one of the embodiments, obtaining the second resource transfer historical behavior data based on the user identifier includes: obtaining the initial credit behavior data used to calculate the user credit data based on the user identifier; obtaining the parameter classification identifier included in the first resource transfer historical behavior data; obtaining the pending credit behavior data corresponding to the parameter classification identifier from the initial credit behavior data; sorting the pending credit behavior data according to the second time information corresponding to the pending credit behavior data, and generating the second resource transfer historical behavior data.

[0015] In one of the embodiments, after obtaining the first resource transfer historical behavior data according to the user identifier and the product identifier, the resource transfer behavior data prediction method also includes: obtaining third time information of the product corresponding to the product identifier; the third time information is used to characterize the validity period of the product; obtaining resource product application request data within the third time information according to the user identifier, the resource product application request data including the number of resource product application requests and the number of resource product application requests passed; generating third resource transfer historical behavior data according to the number of resource product application requests and the number of resource product application requests passed; generating sample data according to the first resource transfer historical behavior data and the second resource transfer historical behavior data, including: generating sample data according to the first resource transfer historical behavior data, the second resource transfer historical behavior data and the third resource transfer historical behavior data.

[0016] In an optional embodiment, after generating sample data, the resource transfer behavior data prediction method also includes: obtaining the behavior data of completed resource transfer and the fourth time information corresponding to the behavior data of completed resource transfer, and the behavior data of incomplete resource transfer and the fifth time information corresponding to the behavior data of incomplete resource transfer included in the sample data; generating positive sample data according to the behavior data of completed resource transfer and the fourth time information corresponding to the behavior data of completed resource transfer, and generating negative sample data according to the behavior data of incomplete resource transfer and the fifth time information corresponding to the behavior data of incomplete resource transfer; using the sample data to train the neural network model to obtain the resource transfer behavior data prediction model, including: training the neural network model according to the positive sample data and the negative sample data to obtain the resource transfer behavior data prediction model.

[0017] In a second aspect, the present application also provides a resource transfer behavior data prediction device, comprising:

[0018] A prediction request acquisition module, used to acquire a resource transfer behavior data prediction request, wherein the resource transfer behavior data prediction request carries a user identifier and a product identifier;

[0019] A first data acquisition module, which acquires first resource transfer history behavior data according to the user identifier and the product identifier;

[0020] A second data acquisition module, which acquires second resource transfer history behavior data according to the user identifier, where the second resource transfer history behavior data is used to calculate the credit data of the user corresponding to the user identifier;

[0021] A sample data generating module, configured to generate sample data according to the first resource transfer history behavior data and the second resource transfer history behavior data;

[0022] A model generation module is used to train the neural network model using sample data to obtain a resource transfer behavior data prediction model;

[0023] The result acquisition module is used to input the target prediction time and the user identifier into the resource transfer behavior data prediction model to obtain the resource transfer behavior data prediction result output by the resource transfer behavior data prediction model.

[0024] 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 the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0025] In a fourth aspect, the present application further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0026] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0027] The above-mentioned resource transfer behavior data prediction method, device, computer equipment, computer-readable storage medium and computer program product generate sample data based on the first resource transfer historical behavior data and the second resource transfer historical behavior data, expand the sample data, thereby improving the accuracy of the resource transfer behavior data prediction model, and further improving the accuracy of the resource transfer behavior data prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0029] FIG1 is an application environment diagram of a resource transfer behavior data prediction method according to an embodiment;

[0030] FIG2 is a schematic diagram of a flow chart of a resource transfer behavior data prediction method in one embodiment;

[0031] FIG3 is a schematic diagram of a flow chart of a resource transfer behavior data prediction method in another embodiment;

[0032] FIG4 is a schematic diagram of a flow chart of a resource transfer behavior data prediction method in yet another embodiment;

[0033] FIG5 is a structural block diagram of a resource transfer behavior data prediction device in one embodiment;

[0034] FIG. 6 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0035] 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.

[0036] The resource transfer behavior data prediction method provided in the embodiment of the present application can be applied to the application environment shown in Figure 1. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 initiates a resource transfer behavior data prediction request, which carries a user identifier and a product identifier, and sends the resource transfer behavior data prediction request to the server 104. The server 104 obtains the resource transfer behavior data prediction request, obtains the first resource transfer history behavior data according to the user identifier and the product identifier; obtains the second resource transfer history behavior data according to the user identifier, and the second resource transfer history behavior data is the data used when calculating the credit data of the user corresponding to the user identifier; generates sample data according to the first resource transfer history behavior data and the second resource transfer history behavior data; uses the sample data to train the neural network model to obtain a resource transfer behavior data prediction model; inputs the target prediction time and the user identifier into the resource transfer behavior data prediction model, and receives the resource transfer behavior data prediction result output by the resource transfer behavior data prediction model. The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. Head-mounted devices may be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server 104 may be an independent physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides cloud computing services.

[0037] In an exemplary embodiment, as shown in FIG2 , a resource transfer behavior data prediction method is provided, which is described by taking the method applied to the server in FIG1 as an example, and includes the following steps 202 to 212. Among them:

[0038] Step 202: Obtain a resource transfer behavior data prediction request.

[0039] The resource transfer behavior data prediction request is used to request prediction of resource transfer behavior data occurring in a target time period for a target prediction product or user. Optionally, the resource transfer behavior data may include, but is not limited to, active resource transfer behavior data, passive resource transfer behavior data, and the like, such as active repayment, active repayment times, active repayment amount, passive repayment, passive repayment times, passive repayment amount, and the like.

[0040] In one embodiment, the resource transfer behavior data prediction request carries a user identifier and a product identifier.

[0041] The user identifier is used to identify the user who performs resource transfer behavior data prediction. Optionally, the user identifier may be account information, or other identification information that can be used to identify the user.

[0042] The product identification is used to determine the target prediction product for predicting the user resource transfer behavior data. Optionally, the product identification can be a product number, or other identification information that can be used to determine the product identification.

[0043] In an optional embodiment, the resource transfer behavior data prediction request may be generated by the terminal, and the terminal sends the resource transfer behavior data prediction request to the server. After receiving the resource transfer behavior data prediction request, the server performs resource transfer behavior data prediction.

[0044] Optionally, a salesperson may operate the terminal to determine the user ID of the user to be predicted and the product ID of the product to be predicted, and encapsulate the user ID and product ID to form a resource transfer behavior data prediction request, and send the resource transfer behavior data prediction request to the server.

[0045] Step 204: Acquire first resource transfer history behavior data according to the user identifier and the product identifier.

[0046] The first resource transfer historical behavior data is used to represent the formed resource transfer behavior data.

[0047] Optionally, the to-be-determined product corresponding to the user may be determined according to the user identifier, and the target predicted product may be further determined according to the product identifier, the first start time and the first end time may be determined according to the determined target predicted product, and the current time may be determined, and the first resource transfer historical behavior data may be determined according to the first start time, the first end time, and the current time. Exemplarily, if the current time is between the first start time and the first end time, the resource transfer behavior data occurring between the first start time and the current time is obtained as the first resource transfer historical behavior data; if the current time is after the first end time, the resource transfer behavior data occurring between the first start time and the first end time is obtained as the first resource transfer historical behavior data.

[0048] In one embodiment, a resource transfer record table may be obtained, and the first resource transfer historical behavior data may be obtained from the resource transfer record table. Optionally, the resource transfer record table may correspond to a product, and the corresponding resource transfer record table may be obtained through a product identifier.

[0049] In one embodiment, the resource transfer record table records resource transfer behavior data of all users within the organization, and the corresponding first resource transfer behavior data can be obtained using the user identifier.

[0050] Step 206: Acquire the second resource transfer history behavior data according to the user identifier.

[0051] The second resource transfer historical behavior data is data used to calculate the credit data of the user corresponding to the user identifier.

[0052] In one embodiment, it can be determined whether to obtain the second resource transfer history behavior data based on the first resource transfer history behavior data. Exemplarily, when the data volume of the first resource transfer history behavior data is insufficient, the second resource transfer history behavior data is obtained, and the first resource transfer history behavior data is supplemented with the second resource transfer history behavior data, so that the data volume of the sample data formed for model training is sufficient, thereby improving the accuracy of the trained model, thereby improving the accuracy of the resource transfer behavior data prediction. Optionally, when there are defects in the data in the first resource transfer history behavior data, the second resource transfer history behavior data can be obtained, and the data in the first resource transfer history behavior data can be replaced with the second resource transfer history behavior data, so that the formed sample data is accurate, thereby improving the accuracy of the model trained based on the sample data, and improving the accuracy of the resource transfer behavior data prediction.

[0053] Optionally, during the implementation process, the second resource transfer historical behavior data can also be directly obtained, and then further determined whether to use the second resource transfer historical behavior data to supplement or adjust the first resource transfer historical behavior data based on whether the data volume of the first resource transfer historical behavior data meets the demand or whether there are defects in the first resource transfer historical behavior data.

[0054] In an optional embodiment, the second resource transfer historical behavior data may be data used to calculate user credit data when the user applies to obtain the target prediction product, to determine whether the user is allowed to obtain the target prediction product.

[0055] In one embodiment, the initial credit behavior data used to calculate the user's credit data can be obtained based on the user identifier, the parameter classification identifier included in the first resource transfer historical behavior data can be obtained, the pending credit behavior data corresponding to the parameter classification identifier can be obtained from the initial credit behavior data, the pending credit behavior data can be sorted according to the second time information corresponding to the pending credit behavior data, and the second resource transfer historical behavior data can be generated.

[0056] Among them, the parameter classification identifier is used to identify different dimensions of the first resource transfer historical behavior data, for example, the maximum overdue days, the minimum overdue days, the maximum overdue amount, the minimum overdue amount, etc.

[0057] In one embodiment, the data for calculating whether the user has the authority to obtain the target prediction product when the user applies for the target prediction product can be obtained and recorded as the initial credit behavior data. Optionally, the credit behavior data to be processed can be obtained from the initial credit behavior data according to the parameter classification identifier, and the credit behavior data to be processed can be obtained in a targeted manner to achieve targeted data acquisition, thereby improving the accuracy of data acquisition, improving the accuracy of the formed sample data, and reducing the noise of the sample data, improving the accuracy of the model obtained by training based on the sample data, thereby improving the accuracy of the resource transfer behavior data prediction.

[0058] Optionally, the obtained credit investigation behavior data to be processed can be sorted according to the second time information of the credit investigation behavior data to be processed to obtain the second resource transfer history behavior data. The second time information can be the time when the credit investigation behavior data to be processed occurs. The credit investigation behavior data to be processed is sorted by using the second time information of the credit investigation behavior data to be processed, so that the credit investigation behavior data to be processed can be sorted according to the sequence of the second time information to form the second resource transfer history behavior data, so that when the second resource transfer history behavior data is used to supplement or correct the first resource transfer history behavior data, the supplemented or corrected data can be obtained more conveniently, thereby improving the efficiency of data acquisition and data processing efficiency.

[0059] Step 208: Generate sample data according to the first resource transfer history behavior data and the second resource transfer history behavior data.

[0060] In one embodiment, it is possible to determine whether the data volume of the first resource transfer historical behavior data meets the demand based on the first start time and the first end time of the first resource transfer historical behavior data. If it does not meet the demand, the second resource transfer historical behavior data can be used to supplement the first resource transfer historical behavior data, thereby expanding the data volume of the formed sample data, ensuring the training sample data volume, improving the accuracy of the model training, and thus improving the accuracy of the resource transfer behavior data prediction results. Optionally, generating sample data based on the first resource transfer historical behavior data and the second resource transfer historical behavior data may include: obtaining the first start time and the first end time of the first resource transfer historical behavior data, and determining the time interval data based on the first start time and the first end time; when the time interval data is less than the preset time interval, determining the second start time based on the preset time interval and the first end time; obtaining the first resource transfer supplementary behavior data from the second resource transfer historical behavior data based on the first start time and the second start time, and generating sample data based on the first resource transfer historical behavior data and the first resource transfer supplementary behavior data.

[0061] Among them, the preset time interval is used to determine whether the data volume of the first resource transfer historical behavior data meets the demand. Optionally, when the time interval data between the first start time and the first end time of the first resource transfer historical behavior data is less than the preset time interval data, it indicates that the data volume of the first resource transfer historical behavior data does not meet the demand, and the second resource transfer historical behavior data is used to supplement the first resource transfer historical behavior data. Optionally, the second start time when the data volume is met can be determined based on the preset time interval data and the first end time, and then the data between the first start time and the second start time can be obtained from the second resource transfer historical behavior data as the first resource transfer supplementary behavior data. The obtained first resource transfer supplementary behavior data is merged with the first resource transfer historical behavior data to obtain sample data. In one embodiment, the preset time interval data can be 24 months.

[0062] Optionally, when the time interval between the first start time and the first end time of the first resource transfer historical behavior data is greater than or equal to a preset time interval, it can be considered that the data volume of the first resource transfer historical behavior data is sufficient, and the first resource transfer historical behavior data can be used for model training.

[0063] In one embodiment, it can be determined whether there are defects in the data of the first resource transfer history behavior data, such as whether there are missing data. When there are defects, the second resource transfer history behavior data can be used to correct the data at the defective part of the first resource transfer history behavior data. Optionally, it can include: when there is a missing data in the first resource transfer history behavior data, obtaining the first time information corresponding to the missing data, obtaining the second resource transfer supplementary behavior data from the second resource transfer history behavior data according to the first time information, and generating sample data according to the first resource transfer history behavior data and the second resource transfer supplementary behavior data.

[0064] Optionally, it can be determined whether the data volume of the first resource transfer historical behavior data meets the demand. If not, the second resource transfer historical behavior data is used to supplement it. It is further determined whether there is a missing in the first resource transfer historical behavior data. If there is a missing, the first resource transfer historical behavior data is corrected by the second resource transfer historical behavior data. Optionally, it can include: obtaining the first start time and the first end time of the first resource transfer historical behavior data, and determining the time interval data according to the first start time and the first end time; when the time interval data is less than the preset time interval, determining the second start time according to the preset time interval and the first end time; obtaining the first resource transfer supplementary behavior data from the second resource transfer historical behavior data according to the first start time and the second start time; generating sample data according to the first resource transfer historical behavior data and the first resource transfer supplementary behavior data; if there is a missing in the first resource transfer historical behavior data, obtaining the first time information corresponding to the data missing, obtaining the second resource transfer supplementary behavior data from the second resource transfer historical behavior data according to the first time information, and updating the sample data according to the second resource transfer supplementary behavior data.

[0065] Optionally, it is also possible to first detect whether there are defects in the first resource transfer historical behavior data. If there are defects, obtain the first time information corresponding to the missing data, obtain the second resource transfer supplementary behavior data from the second resource transfer historical behavior data according to the first time information, and generate sample data according to the first resource transfer historical behavior data and the second resource transfer supplementary behavior data. Then determine the first start time and the first end time of the first resource transfer historical behavior data. When the time interval data between the first start time and the first end time is less than the preset time interval, determine the second start time according to the first end time and the preset time interval, obtain the first resource transfer supplementary behavior data from the second resource transfer historical behavior data according to the first start time and the second start time, and update the sample data according to the first resource transfer supplementary behavior data.

[0066] Optionally, when it is determined that there are omissions in the first resource transfer historical behavior data and the data amount does not meet the requirements, the first resource transfer supplementary behavior data and the second resource transfer supplementary behavior data can be obtained from the second resource transfer historical behavior data, and sample data can be generated based on the first resource transfer historical behavior data, the first resource transfer supplementary behavior data and the second resource transfer supplementary behavior data.

[0067] In an optional embodiment, third time information of the product corresponding to the product identifier can also be obtained, and resource product application request data in the third time information can be obtained according to the user identifier. The resource product application request data includes the number of resource product application requests and the number of resource product application requests passed. Third resource transfer historical behavior data is generated according to the number of resource product application requests and the number of resource product application requests passed. Sample data is generated according to the first resource transfer historical behavior data, the second resource transfer historical behavior data and the third resource transfer historical behavior data.

[0068] The third time information is used to characterize the validity period of the product, for example, the credit period of the credit product. Optionally, the third time information can be determined by the first start time and the first end time, and the third time information = the first end time - the first start time.

[0069] Optionally, within the third time information, it is possible to obtain whether the user corresponding to the user identifier has other resource product application requests, and supplement the sample data according to the resource product application requests corresponding to the user identifier to expand the sample data, thereby improving the accuracy of model training and thus improving the accuracy of the resource transfer behavior data prediction results.

[0070] In one embodiment, the number of passes and the number of fails corresponding to the resource product application request can be obtained, the number of passes can be used as positive samples, and the number of fails can be used as negative samples, the sample data can be divided finely, the accuracy of the sample data can be improved, the accuracy of the model training can be improved, and thus the accuracy of the resource transfer behavior data prediction results can be improved.

[0071] In one embodiment, the sample data may be saved in the form of a matrix, for example:

[0072]

[0073] in, Indicates the value of the dimension data corresponding to the nth parameter classification identifier in the mth month in the credit data. Indicates the value of the resource transfer behavior data of the target forecast product in the dimension data corresponding to the nth parameter classification identifier in the mth month. Represents the time series behavior data of the i-th customer.

[0074] Step 210: Use sample data to train the neural network model to obtain a resource transfer behavior data prediction model.

[0075] Step 211, input the target prediction time and the user identifier into the resource transfer behavior data prediction model to obtain the resource transfer behavior data prediction result output by the resource transfer behavior data prediction model.

[0076] In one embodiment, the sample data of all users included in the organization can be aggregated to form a data matrix, for example:

[0077]

[0078] In an optional embodiment, label data may be added to the resource transfer behavior data included in the sample data of different users. For example, if the resource transfer behavior data is active repayment behavior, an active label is added to the resource transfer behavior data; if the resource transfer behavior data is overdue repayment behavior, an overdue repayment label is added to the resource transfer behavior data.

[0079] Alternatively, the data can be represented in the form of a matrix, for example:

[0080]

[0081] in, Represents the label data of the i-th customer at time T+1.

[0082] In an optional embodiment, after the sample data is added with label data, the neural network model is trained using the sample data. For example, the neural network model can be trained based on D and Y to obtain a resource transfer behavior data prediction model for realizing resource transfer behavior data prediction. Optionally, the output result data of the resource transfer behavior data prediction model can include the resource transfer behavior prediction result and the corresponding probability. For example, the probability of actively executing resource transfer behavior in the next month is 90%.

[0083] In one embodiment, the convergence step size can be determined during model training based on the time series data. For example, if the time series data includes m months of data, the convergence step size can be set to m. In one embodiment, during the model training process, grid search or other parameter tuning methods can also be used to tune parameters to improve the accuracy and efficiency of model training. Optionally, the adjusted parameters can include but are not limited to batch_size, epochs, etc.

[0084] Optionally, when training a neural network model based on D and Y, a test data set may be set, and the test data set may be input into the trained resource transfer behavior data prediction model, and whether the trained resource transfer behavior data prediction model meets the requirements may be determined based on the output test result data. Exemplarily, the evaluation criteria for the resource transfer behavior data prediction model may include, but are not limited to, the range of the cumulative distribution function (KS index) and the area under the ROC curve (receiver operating characteristic curve) (AUC index), etc.

[0085] In an optional embodiment, after generating the resource transfer behavior data prediction model, the target prediction time is obtained, and the obtained target prediction time is input into the resource transfer behavior data prediction model to obtain the prediction result output by the resource transfer behavior data prediction model.

[0086] Optionally, the target prediction time can be determined by the user operating the terminal and sent by the terminal to the server. In one embodiment, the target prediction time can be a month, and the resource transfer behavior data prediction result is the prediction result of the month; optionally, the target prediction time can also be a year, and the resource transfer behavior data prediction result can be the prediction result of the whole year.

[0087] In the resource transfer behavior data prediction request method, sample data is generated based on the first resource transfer history behavior data and the second resource transfer history behavior data, sample data is expanded, and sample data is enriched, thereby improving the accuracy of the resource transfer behavior data prediction model, thereby improving the accuracy of the resource transfer behavior data prediction results. Further, when the first resource transfer history behavior data has insufficient data or is missing, the second resource transfer history behavior data is used to supplement the data in a targeted manner to improve the accuracy of the formed sample data, and the redundancy of the formed sample data can be reduced, the accuracy of the model training can be improved, and thus the accuracy of the resource transfer behavior data prediction results can be improved. Moreover, according to the parameter classification identifier included in the first resource transfer history behavior data, data can be extracted from the initial credit behavior data, and the extracted data can be sorted based on the second time information to generate the second resource transfer history behavior data, and the generation of the second resource transfer history behavior data can be achieved in a targeted manner, which can effectively ensure the accuracy of the second resource transfer behavior data, thereby improving the accuracy of the sample data, improving the accuracy of the model training, and improving the accuracy of the resource transfer behavior data prediction results. Furthermore, the sample data may be further expanded with respect to the resource product application request data in the third time information corresponding to the product identifier, thereby improving the accuracy of model training and the accuracy of resource transfer behavior data prediction results.

[0088] In an exemplary embodiment, as shown in FIG3 , a resource transfer behavior data prediction method is provided, which is described by taking the method applied to the server in FIG1 as an example, and includes the following steps 302 to 324:

[0089] Step 302: Obtain a resource transfer behavior data prediction request.

[0090] Optionally, the resource transfer behavior data prediction request carries a user identifier and a product identifier.

[0091] Step 304: Acquire first resource transfer history behavior data according to the user identifier and the product identifier.

[0092] Step 306, obtaining initial credit behavior data used in calculating the user's credit data based on the user identifier.

[0093] Step 308: Obtain the parameter classification identifier included in the first resource transfer historical behavior data.

[0094] Step 310, obtaining the credit investigation behavior data to be processed corresponding to the parameter classification identifier from the initial credit investigation behavior data.

[0095] Step 312: sort the credit investigation behavior data to be processed according to the second time information corresponding to the credit investigation behavior data to be processed, and generate second resource transfer historical behavior data.

[0096] Step 314: Obtain third time information of the product corresponding to the product identifier.

[0097] The third time information is used to represent the validity period of the product.

[0098] Step 316, obtaining the resource product application request data of the user in the fourth time information and the third time information according to the user identifier.

[0099] Optionally, the resource product application request data includes the number of resource product application requests and the number of resource product application requests passed.

[0100] Step 318: Generate third resource transfer historical behavior data according to the number of resource product application requests and the number of resource product application request approvals.

[0101] Step 320: Generate sample data according to the first resource transfer history behavior data, the second resource transfer history behavior data and the third resource transfer history behavior data.

[0102] Step 322: Use the sample data to train the neural network model to obtain a resource transfer behavior data prediction model.

[0103] Step 324, input the target prediction time and the user identifier into the resource transfer behavior data prediction model to obtain the resource transfer behavior data prediction result output by the resource transfer behavior data prediction model.

[0104] The above-mentioned resource transfer behavior data prediction method can further expand the sample data according to the resource product application request data in the third time information corresponding to the product identifier, improve the accuracy of model training, and improve the accuracy of resource transfer behavior data prediction results.

[0105] In an exemplary embodiment, as shown in FIG. 4 , a resource transfer behavior data prediction method is provided, which is described by taking the method applied to the server in FIG. 1 as an example, and includes the following steps 402 to 428:

[0106] Step 402: Obtain a resource transfer behavior data prediction request.

[0107] Optionally, the resource transfer behavior data prediction request carries a user identifier and a product identifier.

[0108] Step 404: Acquire first resource transfer history behavior data according to the user identifier and the product identifier.

[0109] Step 406: Acquire the second resource transfer history behavior data according to the user identifier.

[0110] The second resource transfer historical behavior data is data used to calculate the credit data of the user corresponding to the user identifier.

[0111] Step 408: Obtain a first start time and a first end time of the first resource transfer historical behavior data, and determine time interval data according to the first start time and the first end time.

[0112] Step 410 , determining whether the time interval data is less than a preset time interval, if so, executing step 412 , if not, executing step 416 .

[0113] Step 412: determine a second start time according to the preset time interval and the first end time.

[0114] Step 414: Acquire the first resource transfer supplementary behavior data from the second resource transfer historical behavior data according to the first start time and the second start time.

[0115] Step 416 , determining whether there is any missing in the first resource transfer history behavior data, if so, executing step 418 , if not, executing step 424 .

[0116] Step 418, obtaining the first time information corresponding to the missing part.

[0117] Step 420: Acquire second resource transfer supplementary behavior data from second resource transfer historical behavior data according to the first time information.

[0118] Step 422: Generate sample data according to the first resource historical behavior data, the first resource transfer supplementary behavior data, and the second resource transfer supplementary behavior data.

[0119] Step 424: Generate sample data based on the historical behavior data of the first resource.

[0120] Step 426, using the sample data to train the neural network model to obtain a resource transfer behavior data prediction model.

[0121] Step 428, input the target prediction time and the user identifier into the resource transfer behavior data prediction model to obtain the resource transfer behavior data prediction result output by the resource transfer behavior data prediction model.

[0122] In this embodiment, when the first resource transfer historical behavior data is insufficient or missing, the second resource transfer historical behavior data is used to perform targeted data supplementation to improve the accuracy of the formed sample data, and reduce the redundancy of the formed sample data, thereby improving the accuracy of model training and improving the accuracy of the resource transfer behavior data prediction results.

[0123] 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.

[0124] Based on the same inventive concept, the embodiment of the present application also provides a resource transfer behavior data prediction device for implementing the resource transfer behavior data prediction 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 one or more resource transfer behavior data prediction device embodiments provided below can refer to the above limitations on the resource transfer behavior data prediction method, and will not be repeated here.

[0125] In an exemplary embodiment, as shown in FIG5 , a resource transfer behavior data prediction device is provided, comprising: a prediction request acquisition module 502, a first data acquisition module 504, a second data acquisition module 506, a sample data generation module 508, a model generation module 510 and a result acquisition module 512, wherein:

[0126] A prediction request acquisition module 502 is used to acquire a resource transfer behavior data prediction request, wherein the resource transfer behavior data prediction request carries a user identifier and a product identifier;

[0127] A first data acquisition module 504 acquires first resource transfer history behavior data according to the user identifier and the product identifier;

[0128] A second data acquisition module 506, which acquires second resource transfer history behavior data according to the user identifier, where the second resource transfer history behavior data is used to calculate the credit data of the user corresponding to the user identifier;

[0129] A sample data generating module 508, configured to generate sample data according to the first resource transfer history behavior data and the second resource transfer history behavior data;

[0130] A model generation module 510 is used to train a neural network model using sample data to obtain a resource transfer behavior data prediction model;

[0131] The result acquisition module 512 is used to input the target prediction time and the user identifier into the resource transfer behavior data prediction model to obtain the resource transfer behavior data prediction result output by the resource transfer behavior data prediction model.

[0132] In one embodiment, the sample data generation module 508 is also used to obtain the first start time and the first end time of the first resource transfer historical behavior data, and determine the time interval data according to the first start time and the first end time; when the time interval data is less than the preset time interval, determine the second start time according to the preset time interval and the first end time; obtain the first resource transfer supplementary behavior data from the second resource transfer historical behavior data according to the first start time and the second start time; generate sample data according to the first resource transfer historical behavior data and the first resource transfer supplementary behavior data.

[0133] In one of the embodiments, the sample data generation module 508 is also used to obtain the first time information corresponding to the missing data when there is a missing data in the first resource transfer historical behavior data; obtain the second resource transfer supplementary behavior data from the second resource transfer historical behavior data according to the first time information; and update the sample data according to the second resource transfer supplementary behavior data.

[0134] In one of the embodiments, the sample data generation module 508 is also used to obtain initial credit behavior data used to calculate user credit data based on the user identifier; obtain the parameter classification identifier included in the first resource transfer historical behavior data; obtain the credit behavior data to be processed corresponding to the parameter classification identifier from the initial credit behavior data; sort the credit behavior data to be processed according to the second time information corresponding to the credit behavior data to be processed, and generate the second resource transfer historical behavior data.

[0135] In one embodiment, the sample data generation module 508 is also used to obtain third time information of the product corresponding to the product identifier; the third time information is used to characterize the validity period of the product; obtain resource product application request data in the third time information according to the user identifier, the resource product application request data including the number of resource product application requests and the number of resource product application requests passed; generate third resource transfer historical behavior data according to the number of resource product application requests and the number of resource product application requests passed; generate sample data according to the first resource transfer historical behavior data and the second resource transfer historical behavior data, including: generating sample data according to the first resource transfer historical behavior data, the second resource transfer historical behavior data and the third resource transfer historical behavior data.

[0136] In an optional embodiment, the sample data generation module 508 is also used to obtain the behavior data of completed resource transfer and the fourth time information corresponding to the behavior data of completed resource transfer, and the behavior data of incomplete resource transfer and the fifth time information corresponding to the behavior data of incomplete resource transfer included in the sample data; generate positive sample data according to the behavior data of completed resource transfer and the fourth time information corresponding to the behavior data of completed resource transfer, and generate negative sample data according to the behavior data of incomplete resource transfer and the fifth time information corresponding to the behavior data of incomplete resource transfer; use the sample data to train the neural network model to obtain a resource transfer behavior data prediction model, including: training the neural network model according to the positive sample data and the negative sample data to obtain a resource transfer behavior data prediction model

[0137] Each module in the resource transfer behavior data prediction 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.

[0138] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG6. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via 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, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a resource transfer behavior data prediction method is implemented.

[0139] Those skilled in the art will understand that the structure shown in FIG. 6 is merely a block diagram of a partial 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 shown in the figure, or combine certain components, or have a different arrangement of components.

[0140] In an exemplary embodiment, a computer device is 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-mentioned method embodiments when executing the computer program.

[0141] 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.

[0142] 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.

[0143] 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.

[0144] 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 memory 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), magnetic 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, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0145] The technical features of the above embodiments may be combined arbitrarily. 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 application.

[0146] The above 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 resource transfer behavior data prediction method, characterized in that: The method comprises: Obtaining a resource transfer behavior data prediction request, wherein the resource transfer behavior data prediction request carries a user identifier and a product identifier; Acquire first resource transfer history behavior data according to the user identifier and the product identifier; Acquire second resource transfer history behavior data according to the user identifier, where the second resource transfer history behavior data is data used when calculating the credit data of the user corresponding to the user identifier; Generate sample data according to the first resource transfer historical behavior data and the second resource transfer historical behavior data; Using the sample data to train a neural network model to obtain a resource transfer behavior data prediction model; The target prediction time and the user identifier are input into the resource transfer behavior data prediction model to obtain the resource transfer behavior data prediction result output by the resource transfer behavior data prediction model.

2. The method according to claim 1, characterized in that The generating sample data according to the first resource transfer history behavior data and the second resource transfer history behavior data includes: Obtain a first start time and a first end time of the first resource transfer historical behavior data, and determine time interval data according to the first start time and the first end time; When the time interval data is less than a preset time interval, determining a second start time according to the preset time interval and the first end time; Acquire first resource transfer supplementary behavior data from the second resource transfer historical behavior data according to the first start time and the second start time; The sample data is generated according to the first resource transfer historical behavior data and the first resource transfer supplementary behavior data.

3. The method according to claim 2, characterized in that After generating the sample data according to the first resource transfer historical behavior data and the first resource transfer supplementary behavior data, the method further includes: When there is a missing portion in the first resource transfer historical behavior data, obtaining first time information corresponding to the missing portion of the data; Acquire second resource transfer supplementary behavior data from the second resource transfer historical behavior data according to the first time information; The sample data is updated according to the supplementary behavior data transferred by the second resource.

4. The method according to claim 1, characterized in that The acquiring the second resource transfer historical behavior data according to the user identifier includes: Acquiring initial credit behavior data used in calculating user credit data according to the user identifier; Obtaining a parameter classification identifier included in the first resource transfer historical behavior data; Acquire the credit investigation behavior data to be processed corresponding to the parameter classification identifier from the initial credit investigation behavior data; The credit investigation behavior data to be processed are sorted according to the second time information corresponding to the credit investigation behavior data to be processed, and the second resource transfer historical behavior data is generated.

5. The method according to claim 1, characterized in that After acquiring the first resource transfer history behavior data according to the user identifier and the product identifier, the method includes: Acquire third time information of the product corresponding to the product identifier; the third time information is used to represent the validity period of the product; Acquire resource product application request data within the third time information according to the user identifier, wherein the resource product application request data includes the number of resource product application requests and the number of resource product application requests passed; Generating third resource transfer historical behavior data according to the number of resource product application requests and the number of resource product application request approvals; The generating sample data according to the first resource transfer history behavior data and the second resource transfer history behavior data includes: The sample data is generated according to the first resource transfer history behavior data, the second resource transfer history behavior data, and the third resource transfer history behavior data.

6. The method according to any one of claims 1 to 5, characterized in that: After generating the sample data, the method further includes: Acquire the behavior data of completed resource transfer and fourth time information corresponding to the behavior data of completed resource transfer, and the behavior data of uncompleted resource transfer and fifth time information corresponding to the behavior data of uncompleted resource transfer included in the sample data; Generate a positive sample according to the behavior data of the completed resource transfer and the fourth time information corresponding to the behavior data of the completed resource transfer, and generate a negative sample according to the behavior data of the uncompleted resource transfer and the fifth time information corresponding to the behavior data of the uncompleted resource transfer; The method of using the sample data to train a neural network model to obtain a resource transfer behavior data prediction model includes: The neural network model is trained according to the positive samples and the negative samples to obtain the resource transfer behavior data prediction model.

7. A resource transfer behavior data prediction device, characterized in that: The device comprises: A prediction request acquisition module, used to acquire a resource transfer behavior data prediction request, wherein the resource transfer behavior data prediction request carries a user identifier and a product identifier; A first data acquisition module, acquiring first resource transfer history behavior data according to the user identifier and the product identifier; A second data acquisition module, which acquires second resource transfer history behavior data according to the user identifier, wherein the second resource transfer history behavior data is data used when calculating the credit data of the user corresponding to the user identifier; A sample data generating module, configured to generate sample data according to the first resource transfer historical behavior data and the second resource transfer historical behavior data; A model generation module, used to train a neural network model using the sample data to obtain a resource transfer behavior data prediction model; The result acquisition module is used to input the target prediction time and the user identifier into the resource transfer behavior data prediction model to obtain the resource transfer behavior data prediction result output by the resource transfer behavior data prediction model.

8. 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 6 are implemented.

9. 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 6 are implemented.

10. A computer program product, comprising a computer program, 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 6 are implemented.