Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
The terms first, second and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged, as appropriate, such that embodiments of the present application may be implemented in sequences other than those illustrated or described herein, and that the objects identified by "first," "second," etc. are generally of a type, and are not limited to the number of objects, such as the first object may be one or more. Furthermore, in the description and claims, "and/or" means at least one of the connected objects, and the character "/", generally means that the associated object is an "or" relationship.
Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As described in the background, with the development of economy, it is a common collaboration mode that companies and channel merchants cooperate and expand business users. The company can periodically settle commissions to the facilitator based on the number of users expanded by the facilitator, sales amount, etc. How to accurately settle the commission of the user of the channel business is a problem to be solved. In the related art, judgment is usually performed manually according to specific credit, historical sales and cash withdrawal conditions of each channel, and settlement proportion is set for the channel to carry out commission settlement. Thus, there is a problem that the settlement amount accuracy is low. The embodiment of the application provides a settlement method for commissions of users, which can solve the problem of lower settlement amount accuracy caused by manually setting settlement proportion in the related technology.
Specifically, in the user commission settlement method provided by the embodiment of the application, target data of a target user is obtained, wherein the target data comprises portrait information of the target user, or the target data comprises portrait information of the target user and historical settlement information of the target user; inputting the target data into a target machine learning model to obtain a target settlement proportion of the target user; determining a commission settlement amount for the target user based on the target settlement proportion of the target user; the target machine learning model is obtained based on target training data, wherein the target training data comprises real portrait information of a user and real historical presentation information of the user. Therefore, the target machine learning model is trained through the real portrait information of the user and the real historical rendering information of the user, so that the target settlement proportion obtained through the target machine learning model can be higher in accuracy. Further, the determined commission settlement amount has higher accuracy based on the target settlement proportion obtained by the target machine learning model, and the problem of lower settlement amount accuracy caused by manually setting the settlement proportion in the related art is solved to a certain extent.
Because the highest cash-on-hand proportion of each channel quotient is manually researched and judged in the related art, the workload is large, the labor cost is high, and additional risks such as inaccurate judgment, camera bellows transaction and the like exist. The problem of difficult determination and large workload in the related technology can be solved to a certain extent.
In the embodiment of the application, the first settlement proportion can be obtained through the portrait information of the target user, and the target settlement proportion is further obtained according to the first settlement proportion, so that the commission settlement amount of the target user is determined. Therefore, the commission settlement can be carried out according to the self identity information of the target user, and when the target user does not have historical settlement information, the commission settlement can be carried out, so that the commission settlement has more use scenes, and the accuracy of the commission settlement is improved.
In the embodiment of the application, the target settlement proportion of the target user can be obtained based on the portrait information of the target user and the historical settlement information of the target user, that is, the target settlement proportion can comprehensively describe the portrait information of the target user and the historical settlement information of the target user, has higher correlation with the target user, and further ensures that the target commission settlement amount determined based on the target settlement proportion also has higher accuracy.
It should be appreciated that the method for settling the commission of the user provided by the embodiment of the application can be executed by the target device. The target device may be one electronic device, or may be a plurality of electronic devices that are executed in cooperation with each other. The electronic device may be, for example, a server, such as an independent physical server, a server cluster formed by a plurality of servers, and a cloud server capable of performing cloud computing.
The method provided by the embodiment of the application is described in detail through specific embodiments and application scenes thereof with reference to the accompanying drawings.
Fig. 1 is a flowchart of a method for settling a commission of a user according to an embodiment of the present application. As shown in fig. 1, the method for settling a commission of a user provided by the embodiment of the application includes the following steps:
Step 110, obtaining target data of a target user, wherein the target data comprises portrait information of the target user, or the target data comprises portrait information of the target user and historical settlement information of the target user.
In the embodiment of the application, the target user can be any user with commissions to be settled, such as an individual channel user cooperated with a company or an enterprise channel cooperated with the company. The target data includes portrait information of the target user, which may describe intrinsic feature information of the target user. For example, when the target user is a channel user, the portrait information of the target user may include: vendor type (channel/enterprise), channel type (agent channel/distribution channel/allied channel), channel level (primary channel/secondary channel), whether individual (enterprise/individual), service type (development user class/maintenance integration class), and information characteristic inherent to the user of the affiliated channel such as the affiliated local market.
The target data may further include portrait information of the target user and historical settlement information of the target user. The historical settlement information of the target user may be information capable of reflecting the historical settlement behavior of the target user. For example, when the target user is a channel user, the historical settlement information of the target user may include: the historical daily settlement amount, monthly settlement amount, the number of times of presentation, the total amount of presentation, the monthly settlement remaining amount, the rewarding amount, the punishment amount, the settlement proportion, the total contract amount, the tax rate of posting, the number of developing users, the assessment coefficient, whether overpayment, overpayment amount and the like are directly and closely related to the historical monthly agent development condition and performance of the channel agent.
In the embodiment of the application, the portrait information of the target user and/or the historical settlement information of the target user can be subjected to numeric processing, and the portrait information and/or the historical settlement information of the target user are represented in a numeric mode, so that the target data corresponding to the target user is obtained.
In the embodiment of the application, the target device can acquire the target data of the target user from the outside, for example, the target device receives the target data sent by the mobile terminal of the target user. The target device may also obtain target data of the target user from within the target device, e.g., after the target device receives a settlement request of the target user, the target device obtains the target data of the target user from within, so that the target device performs the settlement request of the target user. The embodiment of the present application is not particularly limited thereto. After the target device obtains the target data of the target user, steps 120 to 130 may be executed to settle the commission corresponding to the target user according to the target data of the target user.
Step 120, inputting the target data into a target machine learning model to obtain a target settlement proportion of the target user; the target machine learning model is obtained based on target training data, wherein the target training data comprises real portrait information of a user and real historical presentation information of the user.
In the embodiment of the application, the target settlement proportion is used for settling the commission of the target user, and can represent the proportion that the target user can settle the commission. For example, when the target user is a primary channel, the first settlement ratio may be 0.01. When the target user is a primary channel provider, but overpayment appears in the historical settlement information, the first settlement ratio may be 0.005.
In the embodiment of the application, the target machine learning model can be any machine learning model capable of obtaining a target settlement ratio according to target data. The target machine learning model can be obtained by data fitting of target training data and settlement proportion corresponding to the target training data. The target training data may be the same as target data of the target user.
In the embodiment of the application, the target settlement proportion can be obtained through the target machine learning model at regular intervals, and the target settlement proportion can also be obtained through the target machine learning model at any time. The embodiment of the present application is not particularly limited thereto. After the target settlement proportion is obtained through the target machine learning model, the commission settlement amount of the target user can be further obtained.
And 130, determining the commission settlement amount of the target user based on the target settlement proportion of the target user.
In the embodiment of the present application, in the process of determining the commission settlement amount of the target user based on the target settlement proportion of the target user, the amount to be settled of the target user may be obtained, for example, when the target user is a channel user, payment Fei Mingxi bill data of a downlink user (a user who can be understood to consume through a channel of the target user) corresponding to the target user may be obtained, and the amount to be settled of the target user may be obtained from the downlink user payment detail bill data according to the commission settlement rule. And determining the commission settlement amount of the target user based on the target settlement proportion of the target user and the amount to be settled of the target user.
In the embodiment of the application, the target settlement proportion and/or the amount to be settled of the target user can be periodically acquired by the target equipment, and the commission settlement amount of the target user can be periodically determined. For example, the target device may obtain the target settlement proportion of the target user at the beginning of the month according to the target data of the target user, obtain the amount to be settled of the target user every day, and determine the commission settlement amount of the target user at the end of the month.
In the embodiment of the application, the target settlement proportion and/or the amount to be settled of the target user can be obtained by the target equipment periodically, and the commission settlement amount of the target user is not determined periodically. For example, the target device may obtain the target settlement proportion of the target user at the beginning of the month according to the target data of the target user, and obtain the amount to be settled of the target user every day, and determine the commission settlement amount of the target user at any time period in the month. The embodiment of the present application is not particularly limited thereto.
In the embodiment of the application, the target data of the target user is acquired, wherein the target data comprises portrait information of the target user, or the target data comprises portrait information of the target user and historical settlement information of the target user; inputting the target data into a target machine learning model to obtain a target settlement proportion of the target user; determining a commission settlement amount for the target user based on the target settlement proportion of the target user; the target machine learning model is obtained based on target training data, wherein the target training data comprises real portrait information of a user and real historical presentation information of the user. Therefore, the target machine learning model is trained through the real portrait information of the user and the real historical rendering information of the user, so that the target settlement proportion obtained through the target machine learning model can be higher in accuracy. Further, the determined commission settlement amount has higher accuracy based on the target settlement proportion obtained by the target machine learning model, and the problem of lower settlement amount accuracy caused by manually setting the settlement proportion in the related art is solved to a certain extent.
Fig. 2 is a flowchart of a method for settling a commission of a user according to an embodiment of the present application. As shown in fig. 2, the method for settling a commission of a user provided by the embodiment of the application includes the following steps:
Step 210, obtaining target data of a target user, wherein the target data comprises portrait information of the target user.
In the embodiment of the application, after the target data of the target user is obtained, the target data can be input into a target machine learning model to obtain the target settlement ratio. The target machine learning model includes a first settlement model. Accordingly, after obtaining the target data of the target user, steps 220 to 230 may be performed to obtain the target settlement ratio.
And 220, inputting the portrait information of the target user into the first settlement model to obtain a first settlement proportion.
In the embodiment of the present application, the first settlement ratio may be a settlement ratio confirmed according to the portrait information of the target user. For example, when the target user is a primary channel, the first settlement ratio may be 0.01. When the target user is a secondary channel, the first settlement ratio may be 0.005.
In the embodiment of the present application, the first settlement model may be any machine learning model capable of obtaining a first settlement ratio according to image information of the target user. The first settlement model can be obtained by data fitting of the real portrait information of the user and the settlement proportion corresponding to the real portrait information of the user.
After the first settlement ratio is obtained, step 230 may be performed to obtain a target settlement ratio for the target user based on the first settlement ratio.
And step 230, obtaining the target settlement proportion of the target user based on the first settlement proportion.
In the embodiment of the present application, the target settlement ratio may be the same as the first settlement ratio, or may be different from the first settlement ratio, which is not particularly limited in the embodiment of the present application. If the target settlement ratio is different from the first settlement ratio, other factors, such as information of the cooperation time between the target user and the company, can be considered in the process of calculating the target settlement ratio. For example, when the target user is a primary channel provider for a year of collaboration, the first settlement ratio may be 0.01. When the target user is a primary channel provider for two years of collaboration, the first settlement ratio may be 0.015.
Step 240, determining the commission settlement amount of the target user based on the target settlement proportion of the target user.
In the embodiment of the application, the first settlement proportion can be obtained through the portrait information of the target user, and the target settlement proportion is further obtained according to the first settlement proportion, so that the commission settlement amount of the target user is determined. Therefore, the commission settlement can be carried out according to the self identity information of the target user, and when the target user does not have historical settlement information, the commission settlement can be carried out, so that the commission settlement has more use scenes, and the accuracy of the commission settlement is improved.
Fig. 3 is a flowchart of a method for settling a commission of a user according to an embodiment of the present application. As shown in fig. 3, the method for settling the commission of the user provided by the embodiment of the application comprises the following steps:
Step 310, obtaining target data of a target user, wherein the target data comprises portrait information of the target user and historical settlement information of the target user.
In the embodiment of the application, after the target data of the target user is obtained, the target data can be input into a target machine learning model to obtain the target settlement ratio. The target machine learning model includes a first settlement model and a second settlement model. Accordingly, after obtaining the target data of the target user, steps 320 to 340 may be performed to obtain the target settlement ratio.
And 320, inputting the portrait information of the target user into the first settlement model to obtain a first settlement proportion.
And 330, inputting the historical settlement information of the target user into the second settlement model to obtain a second settlement proportion.
In the embodiment of the present application, the second settlement ratio may be a settlement ratio confirmed according to historical settlement information of the target user. For example, when the monthly settlement amount of the target user is 1000, the second settlement ratio may be 0.01. The first settlement ratio may be 0.005 when the monthly settlement amount of the target user is 500.
In the embodiment of the present application, the second settlement model may be any machine learning model capable of obtaining the second settlement ratio according to the historical settlement information of the target user. The second settlement model can be obtained by data fitting of the real historical rendering information of the user and the settlement proportion corresponding to the real historical rendering information of the user.
After obtaining the second settlement ratio, step 340 may be performed to obtain a target settlement ratio for the target user based on the first settlement ratio and the second settlement ratio.
And step 340, obtaining a target settlement proportion of the target user based on the first settlement proportion and the second settlement proportion.
In the embodiment of the present application, the target settlement ratio may be the same as the first settlement ratio or the second settlement ratio, or may be different from both the first settlement ratio and the second settlement ratio, which is not particularly limited in the embodiment of the present application. The target settlement ratio may be a settlement ratio obtained by integrating the portrait information of the target user and the historical settlement information of the target user.
In the embodiment of the application, the target settlement proportion of the target user is obtained based on the first settlement proportion and the second settlement proportion. The target settlement ratio is obtained, for example, by means of weighted summation, and can also be obtained by means of a machine learning model.
Step 350, determining the commission settlement amount of the target user based on the target settlement proportion of the target user.
In the embodiment of the application, the target settlement proportion of the target user can be obtained based on the first settlement proportion and the second settlement proportion, that is, the target settlement proportion can comprehensively describe the portrait information of the target user and the historical settlement information of the target user, has higher correlation with the target user, and further ensures that the determined target commission settlement amount based on the target settlement proportion also has higher accuracy.
In step 340 provided in the embodiment of the present application, in the process of obtaining the target settlement ratio of the target user based on the first settlement ratio and the second settlement ratio, a weighted summation manner may be used. One specific implementation may be:
And acquiring a first weight corresponding to the first settlement proportion and a second weight corresponding to the second settlement proportion. The first weight may represent an importance of the first settlement ratio, which corresponds to an importance of the portrait information of the target user in obtaining the target settlement ratio. The second weight may represent an importance of the second settlement ratio, which is equivalent to an importance of the historical settlement information of the target user in a process of obtaining the target settlement ratio. The first weight and the second weight may be the same or different.
Then the formula is passed: a=b×b1+c×c1, calculating a target value; wherein a is a target value, B is the first settlement ratio, B1 is the first weight, C is the second settlement ratio, and C1 is the second weight.
And acquiring a target settlement proportion of the target user based on the target value. The target settlement ratio may be the same as the target value or different from the target value, and the embodiment of the present application is not particularly limited. In the embodiment of the application, the target settlement proportion is obtained based on the first settlement proportion, the first weight corresponding to the first settlement proportion, the second settlement proportion and the second weight corresponding to the second settlement proportion, and the more accurate target settlement proportion can be obtained according to the importance of the first settlement proportion and the second settlement proportion.
In the embodiment of the application, in the process of obtaining the target settlement ratio according to the first settlement ratio, the first weight corresponding to the first settlement ratio, the second settlement ratio and the second weight corresponding to the second settlement ratio, the first weight and the second weight are adjustable weights; the first weight gradually decreases over time and the second weight gradually increases over time. When the historical settlement information data of the target user is small in initial settlement, prediction by the first settlement model is accurate based on the portrait information of the target user, the weight corresponding to the first settlement model (i.e., the first weight) can be increased, and the weight corresponding to the second settlement model (i.e., the second weight) can be set to be small. Accordingly, when the historical settlement information data of the target user is more, the prediction based on the second settlement model is more accurate, the corresponding weight (i.e., the second weight) can be increased, and the weight corresponding to the first settlement model can be set to be smaller.
In the embodiment of the application, the number threshold of the historical settlement information data of the target user can be set, and the number of the historical settlement information data of the target user is compared with the target settlement proportion corresponding to different machine learning models and weights, so that the weight of each machine learning model is determined. For example, when the historical settlement data of the target user is 500 pieces, the first weight may be set to 0.7 and the second weight may be set to 0.3, and then the verification is performed through the target settlement ratio. If the verification is passed, 500 pieces of historical settlement data of the target user can be set as a first threshold of the number of historical settlement information data of the target user, the corresponding first weight is set to 0.7, and the second weight is set to 0.3. If the verification is not passed, the method can be appropriately adjusted and re-verified until the verification is passed, 500 pieces of historical settlement data of the target user can be set as a first threshold of the number of historical settlement information data of the target user, the corresponding first weight can be set to 0.8, and the second weight can be set to 0.2. Accordingly, 1000 pieces of the historical settlement data of the target user may also be set as the second threshold value of the number of the historical settlement information data of the target user, the corresponding first weight may be set to 0.65, and the second weight may be set to 0.35. The embodiment of the application does not limit the quantity threshold value of the historical settlement information data of the target user and the second weight corresponding to each threshold value.
In step 340 provided in the embodiment of the present application, in the process of obtaining the target settlement ratio of the target user based on the first settlement ratio and the second settlement ratio, a machine learning model may also be used. One specific implementation may be: the target machine learning model further comprises a third settlement model; and inputting the first settlement proportion and the second settlement proportion into the third settlement model to obtain the target settlement proportion of the target user.
In the embodiment of the present application, the third settlement model may be any machine learning model capable of obtaining the target settlement ratio according to the first settlement ratio and the second settlement ratio. The first settlement model includes a recurrent neural network model, the second settlement model includes a long-term memory network model for processing timing characteristics, and the third settlement model includes a decision tree model. In the process of obtaining the target settlement proportion by the third settlement model, the third settlement model can be obtained by fitting the first real settlement proportion and the second real settlement proportion in the training data with the target settlement proportion in the weighted summation mode. In the process of obtaining the third settlement model by fitting according to the first real settlement proportion and the second real settlement proportion in the training data and the target settlement proportion, other factors can be considered to obtain the third settlement model together, so that the third settlement model can be judged more comprehensively. The target settlement proportion is obtained through the first settlement model, the second settlement model and the third settlement model, so that the target settlement proportion has higher accuracy, and the commission settlement amount obtained based on the target settlement proportion also has higher accuracy.
Fig. 4 is a flowchart of a method for settling a commission of a user according to an embodiment of the present application. As shown in fig. 4, the method for settling the commission of the user provided by the embodiment of the application includes the following steps:
Step 410, obtaining target data of a target user, wherein the target data comprises portrait information of the target user, or the target data comprises portrait information of the target user and historical settlement information of the target user.
If the target data includes portrait information for the target user, steps 420, 430, and 460 may be performed; if the target data includes portrait information of the target user and historical settlement information of the target user, steps 420, 440, 450, and 460 may be performed.
And step 420, inputting the portrait information of the target user into the first settlement model to obtain a first settlement ratio.
And step 430, obtaining a target settlement proportion of the target user based on the first settlement proportion.
Step 440, inputting the historical settlement information of the target user into the second settlement model to obtain a second settlement ratio.
And step 450, obtaining the target settlement proportion of the target user based on the first settlement proportion and the second settlement proportion.
In the 450 provided by the embodiment of the present application, a specific implementation manner of obtaining the target settlement ratio of the target user based on the first settlement ratio and the second settlement ratio may be:
Acquiring a first weight corresponding to the first settlement proportion and a second weight corresponding to the second settlement proportion;
By the formula: a=b×b1+c×c1, calculating a target value; wherein a is a target value, B is the first settlement ratio, B1 is the first weight, C is the second settlement ratio, and C1 is the second weight; the first weight and the second weight are adjustable weights; the first weight gradually decreases with the lapse of time, and the second weight gradually increases with the lapse of time;
and acquiring a target settlement proportion of the target user based on the target value.
In the 450 provided by the embodiment of the present application, a specific implementation manner of obtaining the target settlement ratio of the target user based on the first settlement ratio and the second settlement ratio may be:
the target machine learning model further comprises a third settlement model; and inputting the first settlement proportion and the second settlement proportion into the third settlement model to obtain the target settlement proportion of the target user. The first settlement model includes a recurrent neural network model, the second settlement model includes a long-term memory network model for processing timing characteristics, and the third settlement model includes a decision tree model.
Step 460, determining the commission settlement amount of the target user based on the target settlement proportion of the target user.
The target machine learning model is trained through the real portrait information of the user and the real historical rendering information of the user, so that the target settlement proportion obtained through the target machine learning model has higher accuracy. Further, the determined commission settlement amount has higher accuracy based on the target settlement proportion obtained by the target machine learning model, and the problem of lower settlement amount accuracy caused by manually setting the settlement proportion in the related art is solved to a certain extent.
For better understanding of the user commission settlement method provided in the embodiment of the present application, it should be understood that the present application is not limited thereto.
Fig. 5 is a conceptual diagram of a user commission settlement method according to an embodiment of the application, as shown in fig. 5: in the user commission settlement method provided by the embodiment of the application, the target user can be a channel merchant. User payment statement data for the upstream server is obtained daily through the distributed file system (i.e., the target device), and the commission fee for the facilitator is calculated according to the settlement rules for a preset trigger period (e.g., monthly). The highest safe rendering rate (i.e., target settlement rate) can be obtained by a pre-trained target machine learning model at the beginning of each month based on the portrayal information of the channel (i.e., portrayal information of the target user) and the historical rendering information (i.e., historical settlement information of the target user). The user of the channel (during the month) may initiate a cash withdrawal for the accumulated commission amount up to date, the cash withdrawal ratio not being higher than the highest cash withdrawal ratio (i.e., the target settlement ratio). The settlement method for the user commission provided by the embodiment of the application can combine the portrait information and the historical settlement information of the channel quotient, utilize the offline training of a target machine learning model to predict the highest safe settlement proportion, and solve the settlement excess risk on the premise of shortening the commission settlement period.
In the process of obtaining the safe highest presentation ratio (i.e., target settlement ratio) based on portrait information of a target user and historical settlement information of the target user through a pre-trained target machine learning model, a specific process may be as shown in fig. 6: the target machine learning model may include a first machine learning model (i.e., a first settlement model), a second machine learning model (i.e., a second settlement model), and a third machine learning model (i.e., a third settlement model).
Inputting the portrayal information of the channel merchant (i.e., the portrayal information of the target user) into a first machine learning model trained in advance to obtain a first output value (i.e., a first settlement ratio); inputting the history summary as related information (i.e., history settlement information of the target user) into a pre-trained second machine learning model to obtain a second output value (i.e., a second settlement ratio); the first output value and the second output value are input into a pre-trained third machine learning model. In the third machine learning model, the target settlement ratio can be obtained by a weighted manner. Because the portrait information of the user belongs to non-time sequence characteristics, the portrait information can be processed by applying a common neural network model. For example, the first machine model may use a recurrent neural network model (Recurrent Neural Network, RNN for short), or the like. Accordingly, the information related to the rendering behavior of each period of the history (i.e., the history settlement information of the target user) is a time sequence feature, a neural network model capable of processing the time sequence feature may be applied, a long-short-period memory network model (Long Short Time Memory, abbreviated as LSTM) for processing the time sequence feature may be used for the second machine model, and a decision tree model such as a gradient-lifting decision tree (Gradient Boosting Decision Tree, abbreviated as GBDT) may be used for the third machine model.
In the process of obtaining the target settlement proportion, weights corresponding to the first machine learning model and the second machine model can be dynamically adjusted according to the condition of training data. When the historical presentation data of the initial channel is used is smaller, the prediction based on the first machine learning model is more accurate, the corresponding weight can be increased, and the weight corresponding to the second machine learning model can be set smaller. Conversely, when the historical rendering data of the channel is greater, the prediction based on the second machine learning model may be more accurate, the corresponding weight (i.e., the second weight) may be increased, and the weight (i.e., the first weight) corresponding to the first machine learning model may be decreased. The number threshold of the historical rendering data of the channel can be set, and the number of the historical rendering data of the channel is compared with the threshold of the rendering data corresponding to different machine learning models and weights, so that the weight of each machine learning model is determined.
The target machine learning model may be trained prior to utilizing the pre-trained target machine learning model. And training the machine learning model by using the portrait information of the training channel and the historical current service information of the channel to be predicted to obtain the pre-trained machine learning model. The machine learning model can be trained offline, and then the channel related information can be directly input when the machine learning model is applied, so that the result can be determined.
The settlement method for the user commission provided by the embodiment of the application can be used for predicting the highest safe commission proportion by combining the portrait information and the historical commission behavior of the channel and utilizing a machine learning model, and solving the commission excess risk on the premise of shortening the commission charge period.
Fig. 7 is a block diagram of a settlement apparatus for commissions of users according to an embodiment of the present application. As shown in fig. 7, a settlement apparatus 700 for user commissions provided in an embodiment of the present application includes:
an obtaining module 710, configured to obtain target data of a target user, where the target data includes portrait information of the target user, or the target data includes portrait information of the target user and historical settlement information of the target user;
the processing module 720 is configured to input the target data into a target machine learning model to obtain a target settlement ratio of the target user;
A determining module 730, configured to determine a commission settlement amount of the target user based on a target settlement proportion of the target user;
The target machine learning model is obtained based on target training data, wherein the target training data comprises real portrait information of a user and real historical presentation information of the user.
Optionally, in one embodiment of the present application, the target data includes portrait information of the target user, and the target machine learning model includes a first settlement model; in the process of inputting the target data into a target machine learning model to obtain a target settlement ratio of the target user, the processing module 720 is configured to: inputting the portrait information of the target user into the first settlement model to obtain a first settlement proportion; and obtaining the target settlement proportion of the target user based on the first settlement proportion.
Optionally, in one embodiment of the present application, the target data includes portrait information of the target user and historical settlement information of the target user, and the target machine learning model includes a first settlement model and a second settlement model; in the process of inputting the target data into a target machine learning model to obtain a target settlement ratio of the target user, the processing module 720 is configured to: inputting the portrait information of the target user into the first settlement model to obtain a first settlement proportion; inputting the historical settlement information of the target user into the second settlement model to obtain a second settlement proportion; and obtaining the target settlement proportion of the target user based on the first settlement proportion and the second settlement proportion.
Optionally, in one embodiment of the present application, in the process of obtaining the target settlement ratio of the target user based on the first settlement ratio and the second settlement ratio, the processing module 720 is configured to: acquiring a first weight corresponding to the first settlement proportion and a second weight corresponding to the second settlement proportion; by the formula: a=b×b1+c×c1, calculating a target value; wherein a is a target value, B is the first settlement ratio, B1 is the first weight, C is the second settlement ratio, and C1 is the second weight; and acquiring a target settlement proportion of the target user based on the target value.
Optionally, in one embodiment of the present application, the target machine learning model further includes a third settlement model; in the process of obtaining the target settlement ratio of the target user based on the first settlement ratio and the second settlement ratio, the processing module 720 is configured to: and inputting the first settlement proportion and the second settlement proportion into the third settlement model to obtain the target settlement proportion of the target user.
Optionally, in one embodiment of the present application, the first settlement model includes a recurrent neural network model, the second settlement model includes a long-term memory network model for processing the time series features, and the third settlement model includes a decision tree model.
Optionally, in an embodiment of the present application, the first weight and the second weight are both adjustable weights; the first weight gradually decreases over time and the second weight gradually increases over time.
Fig. 8 is a schematic diagram of an electronic device according to an embodiment of the present application. As shown in fig. 8, an electronic device 800 provided by an embodiment of the application may include a processor 810 and a memory 820. The memory stores a computer program which, when executed, implements steps in any one of the user commission settlement methods provided by the embodiments of the present application (e.g., the user commission settlement method shown in any one of fig. 1 to 4).
The Memory is used for storing programs or data, and may be, but is not limited to, random access Memory (Random Access Memory, RAM), read Only Memory (ROM), programmable Read Only Memory (Programmable Read-Only Memory, PROM), erasable Read Only Memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable Read Only Memory (Electric Erasable Programmable Read-Only Memory, EEPROM), or the like.
The embodiment of the application also provides a computer readable storage medium, on which a program or an instruction is stored, which when executed by a processor, implements each process of the above method embodiment, and can achieve the same technical effects, and in order to avoid repetition, a detailed description is omitted here.
Wherein the processor is a processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium such as a computer Read Only Memory (ROM), random Access Memory (RAM), magnetic or optical disk, etc.
The embodiment of the application further provides a chip, which comprises a processor and a communication interface, wherein the communication interface is coupled with the processor, and the processor is used for running programs or instructions to realize the processes of the embodiment of the method, and can achieve the same technical effects, so that repetition is avoided, and the description is omitted here.
Embodiments of the present application provide a computer program product stored in a storage medium, where the program product is executed by at least one processor to implement the respective processes of the above method embodiments, and achieve the same technical effects, and for avoiding repetition, a detailed description is omitted herein.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. Furthermore, it should be noted that the scope of the methods and apparatus in the embodiments of the present application is not limited to performing the functions in the order shown or discussed, but may also include performing the functions in a substantially simultaneous manner or in an opposite order depending on the functions involved, e.g., the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a computer software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present application.
The embodiments of the present application have been described above with reference to the accompanying drawings, but the present application is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present application and the scope of the claims, which are to be protected by the present application.