Data distribution method and device

Abstract

The invention discloses a data distribution method and device, and relates to the technical field of computers. A specific embodiment of the method comprises the steps of obtaining consumption data of a target user and a preset adjustment user number, generating a predicted quota usage amount of the target user based on a prediction model and the consumption data, and further generating a quota of the target user based on a target model, and determining a target adjustment user by combining the predicted quota usage amount of the target user and the preset adjustment user number, and adjusting the allocation quota of the target adjustment user. According to the embodiment, automatic allocation of the user quota is realized, and the accuracy of user quota allocation is further improved through a model prediction mode.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for data distribution. Background Art

[0002] At present, in the management process of logistics enterprises, the management of customer accounts is very important. Since customers may have risky behaviors, it will affect the liquidity of enterprises. In order to deal with the above risks, enterprises need to strictly control the credit limits of customers. Through the refined management of customer credit limits, risks can be avoided from the source. When managing customers, due to the large number of customers and the limitations on information resource acquisition, there will be certain deficiencies in the inventory analysis of customer data, and there will also be certain errors in the customers set based on this.

[0003] The existing customer credit limit analysis and setting process mainly relies on corporate employees to make judgments based on the obtained customer data to set customer credit limits. The manual judgment method has problems of low efficiency and low accuracy. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method for data allocation, which can achieve efficient and accurate setting of user credit limits and reduce enterprise risks.

[0005] To achieve the above object, according to a first aspect of an embodiment of the present invention, a method for data allocation is provided, comprising:

[0006] Obtain consumption data of target users and preset the number of adjusted users;

[0007] Input the consumption data of the target user into the prediction model to obtain the predicted usage of the target user's credit limit. The prediction model is trained based on the historical consumption data of the target user.

[0008] The predicted credit usage of the target user is input into the target model to obtain the credit of the target user. The target model is a learning model established based on the user's consumption information and cost information;

[0009] Determine a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user and the preset number of adjustment users;

[0010] Based on the quota of the target user, the allocation quota of the target adjustment user is adjusted.

[0011] Optionally, before inputting the predicted credit limit usage of the target user into the target model to obtain the credit limit of the target user, the method further includes:

[0012] Setting the objective function of the target model based on the predicted quota usage of the target user;

[0013] Set the quota constraint conditions of the target model based on the target user's predicted quota usage and preset income parameters;

[0014] Set the cost constraints of the target model based on the preset adjustment number of users.

[0015] Optionally, the predicted credit usage of the target user is input into the target model to obtain the credit usage of the target user, including:

[0016] Input the target user's predicted credit limit usage into the target model to obtain the target user's initial credit limit;

[0017] Determine whether the preset income conditions are met based on the target user's initial credit limit;

[0018] When the judgment result is that the preset income condition is met, the initial credit limit of the target user is determined as the credit limit of the target user.

[0019] Optionally, the method further comprises:

[0020] When the judgment result is that the preset income condition is not met, obtaining updated data of the preset income parameter;

[0021] Update the quota constraint conditions corresponding to the preset income parameters based on the updated data;

[0022] The updated initial credit limit of the target user is determined based on the updated credit limit constraint condition, and whether the preset income condition is met is determined again based on the initial credit limit of the target user until the preset income condition is met.

[0023] Optionally, obtain consumption data of target users, including:

[0024] Obtain original consumption information, target user information and industry information of target users;

[0025] Based on the target user's information and industry information, the target user's original consumption information is supplemented with features to obtain the target user's consumption data.

[0026] Optionally, before obtaining consumption data of target users and presetting the number of adjusted users, the method further includes:

[0027] Obtain labor cost data and preset cost limits;

[0028] The preset adjusted number of users is calculated based on labor cost data and preset cost limits.

[0029] Optionally, determining a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user, and the preset number of adjustment users includes:

[0030] Based on the predicted credit limit usage of the target user and the credit limit of the target user, determining a first user among the target users for whom credit limit adjustment is required;

[0031] According to the size of the first users, users among the first users who meet a preset number of adjusted users are determined as target adjusted users.

[0032] Optionally, the method further comprises:

[0033] Based on the difference between the predicted quota usage of the target user and the quota of the target user, high-risk users among the target users are determined for monitoring the high-risk users.

[0034] According to a second aspect of an embodiment of the present invention, there is provided a data distribution device, comprising:

[0035] An acquisition module is used to acquire consumption data of target users and preset the number of adjusted users;

[0036] An input module is used to input the target user's data into a prediction model to obtain the output of the target user's predicted credit usage, where the prediction model is trained based on the target user's historical consumption data; input the target user's predicted credit usage into a target model to obtain the target user's credit output by the target model, where the target model is a learning model established based on the user's consumption information and cost information;

[0037] A determination module, used to determine a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user and the preset number of adjustment users;

[0038] The adjustment module is used to adjust the allocation quota of the target adjustment user based on the quota of the target user.

[0039] Optionally, the device further comprises:

[0040] The setting module is used to set the objective function of the target model based on the predicted quota usage of the target user; set the quota constraint of the target model based on the predicted quota usage and preset income parameters of the target user; and set the cost constraint of the target model based on the preset adjustment number of users.

[0041] Optionally, an input module is used to input the predicted credit limit usage of the target user into the target model to obtain the credit limit of the target user output by the target model, including:

[0042] An input module, used to input the predicted credit limit usage of the target user into the target model to obtain the initial credit limit of the target user;

[0043] A judgment module is used to judge whether the preset income condition is met based on the initial credit limit of the target user;

[0044] The determination module is used to determine the initial credit limit of the target user as the credit limit of the target user when the judgment result is that the preset income condition is met.

[0045] Optionally, the device further comprises: a judgment module, configured to obtain updated data of a preset income parameter when the judgment result is that the preset income condition is not met;

[0046] An updating module, used for updating the quota constraint conditions corresponding to the preset income parameters based on the updated data;

[0047] The determination module is used to determine the updated initial credit limit of the target user based on the updated credit limit constraint condition and to determine again whether the preset income condition is met based on the initial credit limit of the target user until the preset income condition is met.

[0048] Optionally, an acquisition module is used to acquire consumption data of target users and preset the number of adjusted users, including:

[0049] The acquisition module is used to obtain the original consumption information of the target user, the information of the target user and the industry information;

[0050] The supplement module is used to supplement the original consumption information of the target user with features based on the target user's information and industry information to obtain the consumption data of the target user.

[0051] Optionally, the device further comprises:

[0052] The acquisition module is used to obtain labor cost data and preset cost limits;

[0053] The calculation module is used to calculate the preset adjusted number of users based on the labor cost data and the preset cost limit.

[0054] Optionally, a determination module is used to determine a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user, and the preset number of adjustment users, including:

[0055] The determination module is used to determine the first user among the target users who needs to adjust the credit limit based on the predicted credit limit usage of the target users and the credit limit of the target users; according to the size of the first users, determine the users among the first users who meet the preset adjustment user number as the target adjustment users.

[0056] Optionally, the device further comprises:

[0057] The determination module is used to determine high-risk users among the target users based on the difference between the predicted credit limit usage of the target users and the credit limit of the target users, so as to monitor the high-risk users.

[0058] An embodiment of the above invention has the following advantages or beneficial effects: it can realize automatic prediction of user quota and quota usage based on user consumption data and pre-trained models, and further determine users who need to be adjusted according to the preset adjustment number of users, and adjust the allocated quota of users who need to be adjusted according to the predicted quota, thereby realizing accurate allocation of user allocated quota, while improving the efficiency of user quota setting and reducing enterprise risks.

[0059] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 is a schematic diagram of the main process of the method for data distribution according to an embodiment of the present invention;

[0061] Figure 2 is a flow chart of the merchant credit limit determination method of the present invention;

[0062] Figure 3 is a schematic diagram of main modules of a data distribution device according to an embodiment of the present invention;

[0063] Figure 4 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0064] Figure 5 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0065] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0066] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, storage and other aspects of user personal information involved in the technical solution of this disclosure are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security and national security.

[0067] Figure 1 FIG. 1 is a schematic diagram of the main flow of the method for data distribution according to an embodiment of the present invention. Figure 1As shown, the data distribution method of the embodiment of the present invention mainly includes the following steps S101 to S105.

[0068] Step S101, obtaining consumption data of target users and preset adjustment number of users.

[0069] Obtain the consumption data of the target users stored in the enterprise system and external systems, and obtain the number of preset adjustment users, where the consumption data of the target users obtained is the data of the users who have received the adjustment quota application; the number of preset adjustment users is the number set by the enterprise based on actual cost and other data. Specifically, obtain the target users' basic information, transaction information, waybill information and other consumption data through the enterprise system; obtain the target users' industry information and other consumption data through the external system. Among them, the target users can refer to users who have registered in the enterprise system or have had historical consumption behavior, specifically merchants.

[0070] Optionally, the consumption data of the target user includes, for example, basic information, transaction information, industry information, sales behavior, abnormal transaction information, merchant waybill information, consumption information, information on user product selection, and other data. Among them, basic information includes the merchant company registration time, merchant historical registration revocation information, merchant legal person information, merchant cash flow status and merchant insurance data; transaction information includes the frequency of cooperation with logistics customers, cooperation duration, average monthly credit usage amount, average monthly number of application adjustments, average monthly adjustment amount, average monthly number of adjustment rejections, reasons for adjustment applications, etc.; industry information includes the merchant's main categories, merchant influence, and the road area of ​​the merchant's brand; sales behavior includes the historical adjustment frequency under the name of the salesperson, the average monthly number of adjustments, the length of time the salesperson contacts the merchant, the number of visits the salesperson makes to the merchant, the size of the merchant under the salesperson's name, and abnormal adjustments under the salesperson's name; abnormal transaction information includes the merchant's recent claims information, customer complaints, whether large-scale consumption is carried out, and whether the payment method is changed; merchant waybill information includes the merchant's average monthly order volume fluctuations, the weight of a single item of the merchant, and whether the merchant's single item income is abnormal; consumption information includes the merchant's daily consumption amount, historical consumption amount fluctuations, holiday consumption data, and average daily consumption flow.

[0071] Step S102, inputting the consumption data of the target user into a prediction model to obtain the predicted quota usage of the target user, wherein the prediction model is trained based on the historical consumption data of the target user.

[0072] The consumption data of the target user is input into the pre-trained LightGBM model, i.e., the prediction model, to obtain the output data of the model, which is the predicted usage of the target user's quota. The prediction model is trained with historical consumption data as a sample.

[0073] Optionally, before inputting the consumption data of the target user into the prediction model, the consumption data also needs to be labeled. Specifically, based on the acquired merchant's consumption data, determine the merchant's overdue payment or bad debt record caused by inaccurate credit limit adjustment, and label the corresponding data. For the labeled merchant's data, it is necessary to pay attention to the adjustment amount: the adjustment amount generated by the merchant, including the initial amount and the amount after the credit limit adjustment; the number of adjustments: the number of times the merchant has initiated a credit limit adjustment request within a period of time and the amount involved and the merchant situation; the adjustment loss: the impact of the merchant-initiated credit limit adjustment on the overall logistics revenue; the adjustment process, the merchant's current credit limit approval process.

[0074] Among them, the labeling of merchants can be judged based on historical data, and specifically can be labeled based on the merchant's sales data for the past three years.

[0075] Optionally, the credit limit management strategy can be optimized based on the predicted credit limit usage output by the model, including adjusting the merchant's credit limit, improving risk control measures, providing merchants with personalized services, etc., which helps to better understand the merchants' needs and consumption behaviors, and thus provide better services.

[0076] Step S103, inputting the predicted credit usage of the target user into the target model to obtain the credit usage of the target user, where the target model is a learning model established based on the user's consumption information and cost information.

[0077] The predicted credit usage of the target user output by the prediction model is used as input data into the trained target model to obtain the output result of the target model, that is, the credit of the target user. The target model is a pre-trained model. Specifically, a machine learning model is established based on user consumption information and cost information, and solved by a professional solver to obtain the output data of the model.

[0078] Step S104: determining a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user, and the preset number of adjustment users.

[0079] According to the predicted quota usage of the determined target user and the quota calculation of the target user, the predicted quota difference allocated to the target user is allocated, and based on the predicted quota difference and the preset number of adjusted users, users who meet the preset number of adjusted users are screened from the target users as target adjusted users.

[0080] Step S105: adjusting the allocation quota of the target adjustment user based on the quota of the target user.

[0081] According to the determined quota of the target user, the allocated quota of the target adjustment user is adjusted. Specifically, the allocated quota of the target adjustment user is adjusted to the quota of the corresponding target user.

[0082] The data allocation method provided in an embodiment of the present invention can realize automatic prediction of user quotas and quota usage based on user consumption data and pre-trained models, and further determine users who need to be adjusted according to a preset number of adjusted users, and adjust the allocated quotas of users who need to be adjusted according to the predicted quotas, thereby achieving accurate allocation of user allocated quotas, while improving the efficiency of user quota setting and reducing enterprise risks.

[0083] Optionally, before inputting the target user's predicted credit limit usage into the target model and obtaining the target user's credit limit, the method also includes: setting the target function of the target model based on the target user's predicted credit limit usage; setting the credit limit constraint of the target model based on the target user's predicted credit limit usage and preset income parameters; and setting the cost constraint of the target model based on a preset adjusted number of users.

[0084] Before making a quota prediction for a target user, you need to set model parameters, objective functions and other data. Specifically, the objective function of the target model is set based on the predicted quota usage of the target user, which can be set as the sum of the predicted quota usage of the target user and the quota difference of the target user; based on the predicted quota usage of the target user and the preset income parameter, the actual income of each product line and business line of the enterprise is set to at least reach the preset income parameter, that is, the total income determined according to the predicted quota usage allocated to each customer must reach the preset income, where the income of each product line and business line is the sum of the income of each merchant; the labor time cost constraint is set based on the preset number of adjusted users, that is, the labor time required for the set adjustment users cannot exceed the upper limit of the enterprise's labor time.

[0085] Optionally, constraint conditions may also be set based on whether the target user's credit usage rate meets a preset specific target or range; or based on whether each user's credit is within a set upper and lower limit.

[0086] Optionally, the specific formula of the objective function is as follows:

[0087]

[0088] Where i represents the merchant, S is the set of merchants applying for quota adjustment, is the allocation amount of merchant i, is the predicted usage of merchant i’s quota.

[0089] Optionally, the specific formula of the constraint condition is as follows:

[0090] Income constraints:

[0091] in, For product lines, is the business line, r is the preset income parameter corresponding to the usage of each unit of quota for different business lines and product lines, and R is the preset target income.

[0092] Manual time constraints:

[0093]

[0094] in, The labor cost limit is set, that is, the upper limit of labor time. The manual time spent by logistics settlement personnel to adjust merchant quotas, The time required to adjust the credit limit for a single merchant; A single number of the amount to be adjusted for abnormal bill adjustment processing.

[0095] , c( ) is the manual time required to adjust the credit limit of a single merchant.

[0096] Merchant quota constraints:

[0097] Meet a specific goal or scope.

[0098]

[0099] in, The merchant limit. This is the lower limit of the merchant's credit limit.

[0100] Optionally, the target model may be solved using a solver such as Gurobi or CPLEX to obtain a model output result.

[0101] Optionally, the target user's predicted credit limit usage is input into the target model to obtain the target user's credit limit, including: inputting the target user's predicted credit limit usage into the target model to obtain the target user's initial credit limit; judging whether the preset income condition is met based on the target user's initial credit limit; when the judgment result is that the preset income condition is met, determining the target user's initial credit limit as the target user's credit limit.

[0102] Optionally, when the judgment result is that the preset income condition is not met, updated data of the preset income parameter is obtained; the quota constraint condition corresponding to the preset income parameter is updated based on the updated data; the updated initial quota of the target user is determined based on the updated quota constraint condition and whether the preset income condition is met again based on the initial quota of the target user, until the preset income condition is met.

[0103] The predicted quota usage of the target user output by the prediction model is input into the target model to obtain the initial quota output by the target model. Based on the initial quota, it is judged whether the preset income condition is met, that is, based on the initial quota, it is judged whether the total merchant income meets the preset target income R. Specifically, after allocating the merchant quota based on the initial quota, the income of each merchant based on the allocated quota is obtained, and based on the income, it is judged whether the target income R is met. When it is judged that the preset target income is met, the initial quota is determined to be the allocated quota of each merchant, that is, the quota of the target user. When it is judged that the preset target income is not met, the updated target income is obtained. It can be seen that the updated target income should be less than the original target income R. Based on the updated target income, it is judged again whether the merchant income corresponding to the allocated quota meets the conditions. If not, the target income is continuously updated until the target income condition is met.

[0104] Optionally, obtaining consumption data of the target user includes: obtaining original consumption information of the target user, information of the target user and industry information; and supplementing the original consumption information of the target user with features based on the information of the target user and the industry information to obtain the consumption data of the target user.

[0105] The original consumption information, information and industry information of the target user are obtained, and the original consumption information, information and industry information of the target user are combined by means of feature supplementation to obtain the consumption data of the target user.

[0106] Optionally, before obtaining the consumption data of the target user and the preset adjusted number of users, the method further includes: obtaining labor cost data and a preset cost limit; and calculating the preset adjusted number of users based on the labor cost data and the preset cost limit.

[0107] Obtain the enterprise's cost data, including labor cost limits and the preset time required for per capita quota adjustment, and calculate the maximum number of users that can be adjusted based on the labor cost limits and the preset time required for per capita quota adjustment.

[0108] Optionally, target adjustment users among the target users are determined based on the target user's predicted quota usage, the target user's quota and a preset number of adjustment users, including: determining a first user among the target users who needs to have a quota adjustment based on the target user's predicted quota usage and the target user's quota; and determining, based on the size of the first user, users among the first users who meet the preset number of adjustment users as target adjustment users.

[0109] According to the difference between the predicted quota usage of the target user and the quota, the first user among the target users who needs to adjust the quota is determined, and further, according to the preset number of adjusted users, the number of users who can be adjusted is determined, and the first user is selected based on the number of users. Specifically, the first users can be sorted according to the size of the users, and users whose number meets the preset number of adjusted users are selected in order from large to small to adjust the allocated quota.

[0110] Optionally, the method further includes: determining high-risk users among the target users based on the difference between the predicted credit limit usage of the target user and the credit limit of the target user, so as to monitor the high-risk users.

[0111] According to the difference between the predicted credit limit usage of the target user and the credit limit of the target user, the user whose difference is greater than the preset threshold is determined to be a high-risk user, and the high-risk user needs to be further monitored.

[0112] Optionally, for the training of the prediction model, you can input the labeled sample data into the LightGBM model for training. For each sample i, calculate its predicted value under the current model:

[0113]

[0114] Where K represents the number of decision trees in the model. is the predicted value of sample i by the kth decision tree in the model.

[0115] During the training process, we need to calculate the loss function under the current model state:

[0116]

[0117] in represents the loss function of sample i, and represents the regularization term of the k-th decision tree.

[0118] LightGBM uses the gradient boosting algorithm to update the decision tree in each iteration and improves the performance of the model by optimizing the loss function. It uses a histogram-based algorithm to optimize computational efficiency and maintain efficient computing speed when processing large-scale data.

[0119] In each iteration, the model adjusts the structure of the decision tree and selects the optimal split point to increase the leaf node, thereby minimizing the value of the loss function. When the number of iterations reaches the predetermined number of rounds or the loss function value is lower than the set threshold, the LightGBM model trained based on the historical merchant credit information sample has completed the training.

[0120] Integrate the receivables information of merchants with quota information for the current month, including daily fluctuation attributes such as consumption information and order quantity. Then, use the trained LightGBM model to predict the merchant's quota usage in the future.

[0121] a. For each logistics merchant sample, its predicted quota usage needs to be calculated. In the LightGBM model, the predicted quota usage is obtained by summing up the output values ​​of multiple trees. The specific calculation method is as follows:

[0122]

[0123] Among them, (w[i]) is the weight of the i-th tree, (fi) is the output value of the i-th tree for the sample, and b is the bias term.

[0124] In the LightGBM model, the output value of each tree is obtained by the weight sum of its leaf nodes, which can be expressed as:

[0125]

[0126] Among them, (R[j]) is the sample region corresponding to the j-th leaf node, and (w[j]) is the weight of the j-th leaf node.

[0127] b. Output quota usage.

[0128] Based on the above calculation, the predicted credit usage of each merchant sample is obtained. This value can be directly used as an estimate of the credit amount that the merchant may use in the future.

[0129] According to the data allocation method provided by the embodiment of the present invention, it is possible to automatically and accurately predict the user's quota and the amount of quota used based on the user's consumption data and the pre-trained model, and further determine the users who need to be adjusted according to the preset number of adjusted users, and adjust the allocated quota of the users who need to be adjusted by the predicted quota, thereby achieving accurate allocation of user allocation quotas, while improving the efficiency of user quota setting and reducing corporate risks. Among them, accurate allocation of merchant quotas can be achieved through model constraints, making up for mistakes in manual judgment, optimizing the merchant quota allocation process, and helping to achieve rapid and accurate allocation of merchant quotas.

[0130] Figure 2This is a flow chart of the merchant credit determination method of the present invention. As shown in the figure, attributes are supplemented for each merchant based on the acquired data, and the historical credit usage and the latest credit usage of the month are distinguished. The credit usage of each merchant is marked to determine whether bad debts are generated, and the merchant's historical consumption data is processed, and the processed data is used to train the machine learning model M. The data of the current month after the supplemented attributes are input into the model M to predict the merchant credit for the current month. According to the prediction results and the manpower restrictions of the enterprise's monthly review, the merchants who can adjust the credit are calculated, and the merchant credit prediction is generated. Based on the prediction results, it is judged whether the income of the production line and business line meets the preset income coefficient. If it does, the collection data of the merchants who need to adjust the credit is issued. If it does not meet the requirements, the updated income coefficient is obtained and judged again until the income coefficient condition is met.

[0131] Figure 3 is a schematic diagram of a data distribution device 300 provided in an embodiment of the present invention, such as Figure 3 As shown, the data distribution device 300 according to the embodiment of the present invention mainly includes: an acquisition module 301 , an input module 302 , a determination module 303 and an adjustment module 304 .

[0132] An acquisition module 301 is used to acquire consumption data of target users and preset the number of adjusted users;

[0133] Input module 302, used to input the target user's data into the prediction model to obtain the output of the target user's predicted credit usage, the prediction model is trained based on the target user's historical consumption data; input the target user's predicted credit usage into the target model to obtain the target user's credit output by the target model, the target model is a learning model established based on the user's consumption information and cost information;

[0134] A determination module 303 is used to determine a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user and the preset number of adjustment users;

[0135] The adjustment module 304 is used to adjust the allocation quota of the target adjustment user based on the quota of the target user.

[0136] The data allocation device provided in accordance with the embodiment of the present invention can realize automatic prediction of user quotas and quota usage based on user consumption data and pre-trained models, and further determine users who need to be adjusted according to a preset number of adjusted users, and adjust the allocated quotas of users who need to be adjusted according to the predicted quotas, thereby achieving accurate allocation of user allocated quotas, while improving the efficiency of user quota setting and reducing enterprise risks.

[0137] Optionally, the device further comprises:

[0138] The setting module 305 is used to set the objective function of the target model based on the predicted quota usage of the target user; set the quota constraint of the target model based on the predicted quota usage of the target user and the preset income parameter; and set the cost constraint of the target model based on the preset adjustment number of users.

[0139] Optionally, the input module 302 is used to input the predicted credit usage of the target user into the target model to obtain the credit usage of the target user output by the target model, including:

[0140] An input module 302 is used to input the predicted credit limit usage of the target user into the target model to obtain the initial credit limit of the target user;

[0141] A determination module 306 is used to determine whether a preset income condition is met based on the initial credit limit of the target user;

[0142] The determination module 303 is used to determine the initial credit limit of the target user as the credit limit of the target user when the judgment result is that the preset income condition is met.

[0143] Optionally, the device further comprises: a judgment module 306, configured to obtain updated data of a preset income parameter when the judgment result is that the preset income condition is not met;

[0144] An updating module 307, used to update the quota constraint condition corresponding to the preset income parameter based on the updated data;

[0145] The determination module 303 is used to determine the updated initial credit limit of the target user based on the updated credit limit constraint condition and to determine again based on the initial credit limit of the target user whether the preset income condition is met until the preset income condition is met.

[0146] Optionally, the acquisition module 301 is used to acquire consumption data of target users and preset adjustment number of users, including:

[0147] An acquisition module 301 is used to acquire original consumption information of a target user, information of the target user, and industry information;

[0148] The supplement module 308 is used to perform feature supplement on the original consumption information of the target user based on the information of the target user and the industry information to obtain the consumption data of the target user.

[0149] Optionally, the device further comprises:

[0150] An acquisition module 301 is used to acquire labor cost data and preset cost limits;

[0151] The calculation module 309 is used to calculate the preset adjusted number of users based on the labor cost data and the preset cost limit.

[0152] Optionally, the determination module 303 is used to determine the target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user and the preset number of adjustment users, including:

[0153] The determination module 303 is used to determine the first user among the target users who needs to adjust the credit limit based on the predicted credit limit usage of the target users and the credit limit of the target users; and determine the users among the first users who meet the preset adjustment user number as the target adjustment users according to the size of the first users.

[0154] Optionally, the device further comprises:

[0155] The determination module 303 is used to determine high-risk users among the target users based on the difference between the predicted credit limit usage of the target user and the credit limit of the target user, so as to monitor the high-risk users.

[0156] According to the data allocation device provided by the embodiment of the present invention, it is possible to automatically and accurately predict the user's quota and the amount of quota used based on the user's consumption data and the pre-trained model, and further determine the users who need to be adjusted according to the preset number of adjusted users, and adjust the allocated quota of the users who need to be adjusted by the predicted quota, thereby achieving accurate allocation of user allocation quotas, while improving the efficiency of user quota setting and reducing corporate risks. Among them, accurate allocation of merchant quotas can be achieved through model constraints, making up for mistakes in manual judgment, optimizing the merchant quota allocation process, and helping to achieve rapid and accurate allocation of merchant quotas.

[0157] Figure 4 An exemplary system architecture 400 is shown to which the method or apparatus for data distribution according to an embodiment of the present invention may be applied.

[0158] like Figure 4 As shown, system architecture 400 may include terminal devices 401, 402, 403, network 404 and server 405. Network 404 is used to provide a medium for communication links between terminal devices 401, 402, 403 and server 405. Network 404 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0159] Users can use terminal devices 401, 402, 403 to interact with server 405 through network 404 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 401, 402, 403, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0160] The terminal devices 401 , 402 , and 403 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0161] Server 405 may be a server that provides various services, such as a background management server that provides support for websites browsed by users using terminal devices 401, 402, and 403 (for example only). The background management server may obtain the consumption data of the target user and the preset number of adjusted users from the received data such as the data allocation request; input the consumption data of the target user into the prediction model to obtain the predicted quota usage of the target user, the prediction model is trained based on the historical consumption data of the target user; input the predicted quota usage of the target user into the target model to obtain the quota of the target user, the target model is a learning model established based on user consumption information and cost information; determine the target adjusted user among the target users based on the predicted quota usage of the target user, the quota of the target user, and the preset number of adjusted users; adjust the allocated quota of the target adjusted user based on the quota of the target user, and feed back the processing result (such as the allocated quota of the target adjusted user - for example only) to the terminal device.

[0162] It should be noted that the data distribution method provided in the embodiment of the present invention is generally executed by the server 405 , and accordingly, the data distribution device is generally set in the server 405 .

[0163] It should be understood that Figure 4 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0164] Reference below Figure 5 , which shows a schematic diagram of the structure of a computer system 500 of a terminal device suitable for implementing an embodiment of the present invention. Figure 5 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0165] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage part 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the system 500 are also stored. The CPU 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0166] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed, so that a computer program read therefrom is installed into the storage section 508 as needed.

[0167] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 509, and / or installed from the removable medium 511. When the computer program is executed by the central processing unit (CPU) 501, the above-mentioned functions defined in the system of the present invention are executed.

[0168] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0169] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0170] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes an acquisition module, an input module, a determination module and an adjustment module. The names of these modules do not constitute a limitation on the modules themselves in some cases, for example, the acquisition module may also be described as "a module for acquiring consumption data of target users and presetting the number of adjusted users".

[0171] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes:

[0172] Obtain consumption data of target users and preset the number of adjusted users;

[0173] Input the consumption data of the target user into the prediction model to obtain the predicted usage of the target user's credit limit. The prediction model is trained based on the historical consumption data of the target user.

[0174] The predicted credit usage of the target user is input into the target model to obtain the credit of the target user. The target model is a learning model established based on the user's consumption information and cost information;

[0175] Determine a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user and the preset number of adjustment users;

[0176] Based on the quota of the target user, the allocation quota of the target adjustment user is adjusted.

[0177] According to the technical solution of the embodiment of the present invention, it is possible to automatically predict the user's credit limit and credit limit usage based on the user's consumption data and a pre-trained model, and further determine the users who need to be adjusted based on the preset number of adjusted users. The allocated credit limit of the users who need to be adjusted is adjusted according to the predicted credit limit, thereby achieving accurate allocation of user allocated credit limits, while improving the efficiency of user credit limit setting and reducing enterprise risks.

[0178] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for data distribution, characterized in that: include: Obtain consumption data of target users and preset the number of adjusted users; Inputting the consumption data of the target user into a prediction model to obtain a predicted usage amount of the target user, wherein the prediction model is trained based on the historical consumption data of the target user; Inputting the predicted credit usage of the target user into a target model to obtain the credit of the target user, wherein the target model is a learning model established based on user consumption information and cost information; Determine a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user and the preset number of adjustment users; Based on the quota of the target user, the allocation quota of the target adjustment user is adjusted.

2. The method according to claim 1, characterized in that Before inputting the predicted credit limit usage of the target user into the target model to obtain the credit limit of the target user, the method further includes: Setting the objective function of the target model based on the predicted credit usage of the target user; Setting a quota constraint condition of the target model based on the predicted quota usage and preset income parameters of the target user; The cost constraint condition of the target model is set based on the preset adjustment number of users.

3. The method according to claim 2, characterized in that The step of inputting the predicted credit usage of the target user into a target model to obtain the credit usage of the target user includes: Inputting the predicted credit limit usage of the target user into the target model to obtain the initial credit limit of the target user; Determining whether the preset income condition is met based on the initial credit limit of the target user; When the judgment result is that the preset income condition is met, the initial credit limit of the target user is determined to be the credit limit of the target user.

4. The method according to claim 3, characterized in that The method further comprises: When the judgment result is that the preset income condition is not met, obtaining updated data of the preset income parameter; Updating the quota constraint condition corresponding to the preset income parameter based on the updated data; The updated initial credit limit of the target user is determined based on the updated credit limit constraint condition, and whether the preset income condition is met is determined again based on the initial credit limit of the target user until the preset income condition is met.

5. The method according to claim 1, characterized in that The acquiring of consumption data of the target user includes: Obtain original consumption information, target user information and industry information of target users; The original consumption information of the target user is supplemented with features based on the information of the target user and the industry information to obtain the consumption data of the target user.

6. The method according to claim 1, characterized in that Before obtaining the consumption data of the target user and presetting the number of adjusted users, the method further includes: Obtain labor cost data and preset cost limits; The preset adjusted number of users is calculated based on the labor cost data and the preset cost limit.

7. The method according to any one of claims 1 to 6, characterized in that Determining a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user, and the preset number of adjustment users includes: Determining a first user among the target users for whom a credit limit adjustment is required based on the predicted credit limit usage of the target users and the credit limit of the target users; According to the size of the first users, users among the first users who meet the preset number of adjusted users are determined as target adjusted users.

8. The method according to claim 1, characterized in that: The method further comprises: Based on the difference between the predicted credit limit usage of the target user and the credit limit of the target user, high-risk users among the target users are determined to monitor the high-risk users.

9. A data distribution device, characterized in that: include: An acquisition module is used to acquire consumption data of target users and preset the number of adjusted users; An input module, used to input the target user's data into a prediction model to obtain an output of the target user's predicted credit usage, wherein the prediction model is trained based on the target user's historical consumption data; Inputting the predicted credit usage of the target user into a target model to obtain the credit usage of the target user output by the target model, wherein the target model is a learning model established based on user consumption information and cost information; A determination module, configured to determine a target adjustment user among the target users based on the predicted quota usage of the target user, the quota of the target user and the preset number of adjustment users; The adjustment module is used to adjust the allocation quota of the target adjustment user based on the quota of the target user.

10. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 8.

11. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images